WO2023151989A1 - Incorporating conditions into data-collection & ai/ml operations - Google Patents

Incorporating conditions into data-collection & ai/ml operations Download PDF

Info

Publication number
WO2023151989A1
WO2023151989A1 PCT/EP2023/052345 EP2023052345W WO2023151989A1 WO 2023151989 A1 WO2023151989 A1 WO 2023151989A1 EP 2023052345 W EP2023052345 W EP 2023052345W WO 2023151989 A1 WO2023151989 A1 WO 2023151989A1
Authority
WO
WIPO (PCT)
Prior art keywords
metadata
data
model
network
dataset
Prior art date
Application number
PCT/EP2023/052345
Other languages
French (fr)
Inventor
Abdulrahman ALABBASI
Angelo Centonza
Michael Meyer
Alexandros PALAIOS
Philipp GEUER
Original Assignee
Telefonaktiebolaget Lm Ericsson (Publ)
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Telefonaktiebolaget Lm Ericsson (Publ) filed Critical Telefonaktiebolaget Lm Ericsson (Publ)
Publication of WO2023151989A1 publication Critical patent/WO2023151989A1/en

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/147—Network analysis or design for predicting network behaviour
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/16—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using machine learning or artificial intelligence

Definitions

  • Embodiments of the present disclosure relate to methods, controller modules, proxy modules, network nodes and user equipments (UEs), and particularly methods, controller modules, proxy modules, network nodes and UEs for facilitating Artificial Intelligence (AI)ZMachine Learning (ML) operations through the production and/or use of metadata.
  • AI Artificial Intelligence
  • ML Machine Learning
  • the NG-Radio Access Network comprises a set of New Radio base stations (gNBs) connected to the 5GC through the NG interface.
  • gNBs New Radio base stations
  • the NG-RAN Overall Architecture is shown in Figure 1.
  • NG-RAN could also comprise a set of ng- eNBs, an ng-eNB may comprise an ng-eNB-CU and one or more ng-eNB-DU(s). An ng-eNB-CU and an ng-eNB- DU is connected via W1 interface.
  • the general principle described in this section also applies to ng-eNB and W1 interface, if not explicitly specified otherwise.
  • a gNB can support FDD mode, TDD mode or dual mode operation.
  • gNBs can be interconnected through the Xn interface.
  • a gNB may comprise a gNB-CU and one or more gNB-DU(s).
  • a gNB-CU and a gNB-DU is connected via F1 interface.
  • One gNB-DU may be connected to only one gNB-CU.
  • each Cell Identity associated with a subset of PLMNs may correspond to a gNB-DU and the gNB-CU it is connected to, i.e. the corresponding gNB-DUs share the same physical layer cell resources.
  • a gNB-DU may be connected to multiple gNB-CUs by appropriate implementation.
  • NG, Xn and F1 are logical interfaces.
  • the NG and Xn-C interfaces for a gNB comprising a gNB-CU and gNB-DUs, terminate in the gNB-CU.
  • EN-DC the S1-U and X2-C interfaces for a gNB comprising a gNB-CU and gNB-DUs, terminate in the gNB-CU.
  • the gNB-CU and connected gNB-DUs are only visible to other gNBs and the 5GC as a gNB.
  • a possible deployment scenario is described in 3GPP TS 38.300 version 16.2.0 (available at the above address as of 18 January 2023).
  • the node hosting user plane part of NR PDCP may perform user inactivity monitoring and further inform its inactivity or (re)activation to the node having C-plane connection towards the core network (e.g. over E1, X2).
  • the node hosting NR RLC e.g. gNB-DU
  • the node hosting NR RLC may perform user inactivity monitoring and further inform its inactivity or (re)activation to the node hosting control plane, e.g. gNB-CU or gNB-CU-CP.
  • UL PDCP Uplink Packet Data Convergence Protocol
  • Radio Link Outage/Resume for DL and/or UL is indicated via X2-U (for EN-DC), Xn-U (for NG-RAN) and F1-U.
  • the NG-RAN is layered into a Radio Network Layer (RNL) and a Transport Network Layer (TNL).
  • RNL Radio Network Layer
  • TNL Transport Network Layer
  • the NG-RAN architecture i.e. the NG-RAN logical nodes and interfaces between them, is defined as part of the RNL.
  • NG-RAN interface For each NG-RAN interface (NG, Xn, F1) the related TNL protocol and the functionality are specified.
  • the TNL provides services for user plane transport, signalling transport.
  • each NG-RAN node is connected to all AMFs of AMF Sets within an AMF Region supporting at least one slice also supported by the NG-RAN node.
  • a gNB may comprise a gNB-CU-CP, multiple gNB-CU-UPs and multiple gNB-DUs;
  • the gNB-CU-CP is connected to the gNB-DU through the F1-C interface;
  • the gNB-CU-UP is connected to the gNB-DU through the F1-U interface;
  • the gNB-CU-UP is connected to the gNB-CU-CP through the E1 interface;
  • One gNB-DU is connected to only one gNB-CU-CP;
  • One gNB-CU-UP is connected to only one gNB-CU-CP;
  • a gNB-DU and/or a gNB-CU-UP may be connected to multiple gNB-CU-CPs by appropriate implementation.
  • One gNB-DU can be connected to multiple gNB-CU-UPs under the control of the same gNB-CU-CP.
  • One gNB-CU-UP can be connected to multiple DUs under the control of the same gNB-CU- CP.
  • the gNB-CU-CP selects the appropriate gNB-CU-UP(s) for the requested services for the UE. In case of multiple CU-UPs they belong to same security domain as defined in TS 33.210.
  • Data Collection is a function that provides input data to Model training and Model inference functions.
  • AI/ML algorithm specific data preparation e.g., data pre-processing and cleaning, formatting, and transformation
  • Examples of input data may include measurements from UEs or different network entities, feedback from Actor, output from an AI/ML model.
  • Training Data Data needed as input for the AI/ML Model Training function.
  • Inference Data Data needed as input for the AI/ML Model Inference function.
  • Model Training is a function that performs the ML model training, validation, and testing which may generate model performance metrics as part of the model testing procedure.
  • the Model Training function is also responsible for data preparation (e.g. data pre-processing and cleaning, formatting, and transformation) based on Training Data delivered by a Data Collection function, if required.
  • Model Deployment/Update Used to initially deploy a trained, validated, and tested AI/ML model to the Model Inference function or to deliver an updated model to the Model Inference function.
  • Model Inference is a function that provides AI/ML model inference output (e.g. predictions or decisions). It is FFS (for future study) whether it provides model performance feedback to Model Training function.
  • the Model inference function is also responsible for data preparation (e.g. data pre-processing and cleaning, formatting, and transformation) based on Inference Data delivered by a Data Collection function, if required.
  • Output The inference output of the AI/ML model produced by a Model Inference function.
  • Model Performance Feedback Applied if certain information derived from Model Inference function is suitable for improvement of the AI/ML model trained in Model Training function. Feedback from Actor or other network entities (via Data Collection function) may be needed at Model Inference function to create Model Performance Feedback.
  • Actor is a function that receives the output from the Model inference function and triggers or performs corresponding actions.
  • the Actor may trigger actions directed to other entities or to itself.
  • Feedback Information that may be needed to derive training or inference data or performance feedback.
  • Model Training and Model Inference functions may be able to request, if needed, specific information to be used to train or execute the AI/ML algorithm and to avoid reception of unnecessary information. The nature of such information depends on the use case and on the AI/ML algorithm.
  • the Model Inference function may signal the outputs of the model only to nodes that have explicitly requested them (e.g. via subscription), or nodes that are subject to actions based on the output from Model Inference.
  • One of the main principles of the currently discussed Functional Framework for RAN Intelligence is the use of a subscription paradigm, so that information is transferred from a second functional block (e.g. Model Training or Model Inference) to a first functional block once a subscription request from the first functional block has been accepted by the second functional block.
  • a second functional block e.g. Model Training or Model Inference
  • GAN KL Divergence; Average Log-likelihood; Coverage Metric; Inception Score (IS); Modified Inception Score (m-IS); Mode Score; AM Score; Frechet Inception Distance (FID); Maximum Mean Discrepancy (MMD); The Wasserstein Critic; Birthday Paradox Test; Classifier Two-sample Tests (C2ST); Classification Performance; Boundary Distortion; Number of Statistically-Different Bins (NDB); Image Retrieval Performance; Tournament Win Rate and Skill Rating; Normalized Relative Discriminative Score (NRDS); Adversarial Accuracy and Adversarial Divergence; Geometry Score; Reconstruction Error.
  • Some such examples are: Sample bias; Dual Drift; Feature (covariate) drift; Label drift; Concept drift.
  • Problem-1 is how to enable the product unit to test new technologies or schedulers, as ML tech/scheduler needs to be safe before developing a new product. This extending test requires data, interactivity, and training and such processes might be impaired by the lack of opportune datasets.
  • descriptive meta data can be environment characteristic parameters, e.g, speed, associated g N B, requested traffic, number of UEs associated with BSs, etc).
  • Problem-4 The standardized data analytic and Al related protocols (e.g., NWDAF, etc) may not enable an efficient solution to the above problems, as they are more focused towards data and analytics (prediction) extraction. Specifically:
  • NWDAF and the Data Collection functions are currently only able to perform analytics or predictions of data, which is collected from other NFs/OAM: o Some data might be missing (type of data, only available what is specified and from RAN only CAM input possible) o Collected data might only represent specific scenarios as these are restricted from the deployment (i.e. location, numbers of users). Data from many more scenarios which would be interesting might very rarely be collected (too little to train) o NWDAF (Release 17) cannot perform any action itself, only give predictions and provide recommendations.
  • An embodiment of the disclosure provides a method performed by a controller module for processing data.
  • the method comprises sending a request for metadata relating to a network to a proxy module, and responsive to receiving from the proxy module metadata generated by a generative model, using the received metadata as input to a trained model.
  • the method further comprises determining an action to take based on the output of the trained model.
  • a further embodiment of the disclosure provides a method performed by a proxy module for generating data.
  • the method comprises receiving, from a controller module, a request for metadata relating to a network.
  • the method further comprises generating metadata using a generative model, and sending the generated metadata to the controller module.
  • a further embodiment of the disclosure provides a method performed by a controller module for training a model.
  • the method comprises sending a request for metadata relating to a network to a proxy module.
  • the method further comprises, responsive to receiving metadata from the proxy module generated by a generative model, using the metadata as training data to train a model.
  • a further embodiment of the disclosure provides a method performed by a proxy module for training a generative model.
  • the method comprises receiving a dataset and receiving metadata.
  • the method further comprises using the dataset and metadata to train a generative model to generate metadata based on an input dataset.
  • controller modules configured to execute some or all of the methods disclosed herein.
  • This disclosure focusses on AI/ML operations facilitated by the production of metadata, namely data that can be used as inputs to a Model Inference Functions or as training data to a Model Training function, which are not derived from real events and scenarios, but that are deduced by dataset generation processes.
  • such dataset generation may be carried out by a proxy module.
  • proxy module may be a separate function or it may be part of the existing functions identified in the functional framework that RAN3 has defined.
  • This disclosure focusses on two parts: the standardization impact of realizing conditional data generator, forecaster, controller, in term of the factors triggering generation of such metadata, and the methods to exchange the metadata produced by such functions and in terms of utilization of such data by the functions receiving it.
  • Identification of the events that trigger the generation of metadata may be, for example, the identification of a missing set of training data range from the data used to train a model. The latter may cause that the finally trained model would not be able to correctly infer events triggered by inputs in the region of missing training data. For this reason it would be beneficial to produce a set of metadata filling the gap in the training data used to train the model.
  • the node or function receiving such metadata may use it in a way similar to e.g. data derived from field operations for the purpose of e.g. training. However, the node receiving such metadata may also use them differently, for example by giving to such data a lower weight when compared to e.g. data derived from field operations.
  • a Model Training function may use metadata to train a model over a wider range of training data.
  • Signaling of data When a node or function needs data such as training data, inference data or feedback data (feedback to the model performance or to the system performance, for example), the node or function will request functions such as the Data Collection function or the Actor for such data.
  • the functions providing such data may signal, together with the data, an indication that the data signaled are metadata generated by a dataset generator, in order to differentiate such data from, e.g. data derived from field operations.
  • Such signaling can be enriched with details such as the methods used to derive such metadata, the accuracy of such metadata and more.
  • - Conditional-based AI/ML operation is an AI/ML operation that uses specific radio or UE-capability, or environment, or contextual scenarios during its training, inference or data generation/measurement operation.
  • AI/ML Information could be input data to AI/ML model, the AI/ML model itself, training data, Model Performance feedback data, information on System performance impacted by the Al operations.
  • Metadata a set of data generated by a dataset generator and needed to optimize different aspects of an Al process, such as training. These data may be distinguished from other AI/ML information derived from e.g. field measurements.
  • the newly added functionality embedded in existing functions or provided as an independent function, enables the 3GPP framework to request, send, and/or receive different Al -I nformation (e.g., data and models) with distinct meta(conditional)-data operation.
  • the new framework generates data with specific characteristics as being requested from another entity (e.g., Data Collection function, Model Training Function, Model Inference function, controller entity, or forecasting entity, etc) during the learning/training or inference procedure.
  • a mechanism is also proposed to validate the conditionally generated data for the requester entity.
  • Those conditionally generated data e.g., environment states, rewards, or observations
  • Attaching 'proactive' prefix to "action 1 ' means that the need to take an action is predicted ahead of time.
  • Certain embodiments may provide one or more of the following technical advantage(s).
  • the proposed framework may have the advantage of enabling the exchange of conditionally assisted training and data generation to enable a more efficient control operation (or Reinforcement Learning).
  • the proposed framework may enable the following: o Proactive studies and evaluations. For example, what would be the UE throughput if the gNB proactively executed a handover. o Safe-training mechanism that may improve the changes of having a less-risky controller in the deployment phase. o Enriching the datasets enabling more studies of what the initial dataset could potentially support. o Enhancing simulation capabilities. The end-user is empowered by introducing characteristics to the dataset. o Future states of observation x or y can be easily predicted.
  • Figure 1 is a block diagram illustrating NG-RAN Overall Architecture
  • Figure 2 illustrates the overall architecture for separation of gNB-CU-CP and gNB-CU-UP
  • Figure 3 illustrates the Functional Framework for RAN Intelligence
  • Figure 4 illustrates a method in a controller module according to an example
  • Figure 5 illustrates a method in a proxy module according to an example
  • Figure 6 illustrates a method in a controller module according to an example
  • Figure 7 illustrates a method in a proxy module according to an example
  • Figure 8 illustrates AI/ML connected framework/blocks in 3GPP TR 37.817;
  • Figure 9 illustrates a general proposed framework according to an example
  • Figure 10 illustrates an example of methods applied to the 3GPP RAN3 AI/ML framework, case of training metadata
  • Figure 11 illustrates an example of methods applied to the 3GPP RAN3 AI/ML framework, case of inference input metadata
  • Figure 12 illustrates an example of metadata collection in the AI/ML energy saving use case
  • Figure 13 illustrates a flow diagram for a part of the proposed system
  • Figure 14 illustrates an adaptation of Functional Framework for RAN intelligence in 3GPP TR 37.817;
  • Figure 15 illustrates Updated Potential Architecture of the proposed Framework
  • Figure 17 illustrates an example of a communication system
  • Figure 18 shows an example of a UE
  • Figure 19 shows an example of a network node
  • Figure 20 is a block diagram of an example of a host
  • Figure 21 is a block diagram illustrating a virtualization environment
  • Figure 22 is a communication diagram of a host.
  • FIG. 4 depicts a method in accordance with particular embodiments.
  • the method W1 may be performed by a controller module.
  • the controller module may be configured to process data.
  • the method begins at step 402 with sending a request for metadata relating to a network to a proxy module, at step 402, responsive to receiving from the proxy module metadata generated by a generative model, using the received metadata as input to a trained model (to generate an output), and at step 403, determining an action to take based on the output of the trained model.
  • Step 403 may be optional, where for example the controller module may instead send the output of the trained model (e.g. to a further module, node etc).
  • the output of the trained model may comprise at least one of: an environmental state; a reward; an observation.
  • the method may further comprise sending the determined action to at least one of a UE, a network node.
  • the method may further comprise receiving a dataset comprising at least one of measured and simulated data, and wherein the dataset may be additionally used as input to the trained model.
  • the metadata may be given a different weight to a weight given to the dataset.
  • the method may further comprise sending at least one of: information indicating the type of metadata to be generated; information indicating the range within which the metadata is to be generated; at least one condition to be used for the generation of the metadata; information indicating at least one generative model to be used to generate the metadata; information indicating how the generative model is to be trained prior to generating the metadata; a forecasting time interval for generation of the metadata.
  • the method may further comprise receiving at least one of: an indication of the type of data received; for each type of data received, receiving an indication of whether the data comprises generated metadata; an indication of which part of the data comprises generated metadata.
  • the method may further comprise receiving at least one of: information indicating that the metadata is generated data; information indicating the type of metadata generated; information indicating the model used to generate the metadata; information indicating the conditions used to generate the metadata; information indicating how conditions were used to generate the metadata; information on the generative model; metadata values; information indicating the accuracy of the metadata.
  • the method may further comprise using the received information as input to the trained model.
  • the method may be performed by a model inference function.
  • FIG. 5 depicts a method in accordance with particular embodiments.
  • the method 5 may be performed by a proxy module.
  • the proxy module may be configured to generate data.
  • the method may begin at step 502 with receiving, from a controller module, a request for metadata relating to a network, at step 504, generating metadata using a generative model, at step 506, sending the generated metadata to the controller module.
  • the receiving of a request may be optional.
  • the proxy module may instead periodically generate metadata.
  • At least one of a condition and a dataset may be input to the generative model to generate metadata.
  • the dataset may comprise at least one of measured and simulated data.
  • a condition may comprise information relating to a contextual scenario.
  • the method may further comprise sending at least one of: information indicating that the metadata is conditionally generated; information indicating the type of metadata generated; information indicating the model used to generate the metadata; information indicating the conditions used to generate the metadata; information indicating how conditions used to generate the metadata were used; information on the generative model; metadata values; information indicating the accuracy of the metadata.
  • the method may further comprise receiving at least one of: information indicating the type of metadata to be generated; information indicating the range within which the metadata is to be generated; at least one condition to be used for the generation of the metadata; information indicating at least one generative model to be used to generate the metadata; information indicating how the generative model is to be trained prior to generating the metadata, and wherein the method further comprises using the received information as input to the generative model.
  • the method may further comprise receiving feedback on the result of an action instructed by the controller module to at least one of a UE, a network node.
  • the generated metadata may comprise at least one of: information relating to data missing from a dataset; forecast metadata.
  • the method may be performed by a model inference function.
  • the method may further comprise receiving at the proxy module a dataset.
  • the method may further comprise updating the dataset based on the metadata output from the generative model.
  • the system may further comprise a validation module, and the method further comprises processing the metadata by the validation module before updating the dataset.
  • the validation module may update the dataset if the metadata fulfils an event criteria.
  • An event criteria may comprise at least one of: a rare event criteria; a quality wise criteria; extra desired data; metadata is derived using a new experiment.
  • An experiment may comprise at least one of: altering network configuration; adding or removing base stations; increasing or reducing coverage; altering radio environment; introducing coverage holes; changing propagation characteristics; adding or removing frequency layers; new channel ranks; user dynamics; more throttling at the cells; dynamic/constant type of traffic; measurement campaign that contains new margins of KPIs; shorter or longer margin of latency; shorter or longer margin of throughput; shorter or longer margin of reliability; shorter or longer margin of SINR; shorter or longer margin of bandwidth; shorter or longer margin of center frequency.
  • Figure 6 depicts a method in accordance with particular embodiments.
  • the method 6 may be performed by a controller module.
  • the controller module may be configured to train a model.
  • the method may begin at step 602 with sending a request for metadata relating to a network to a proxy module, then at step 604, responsive to receiving metadata from the proxy module generated by a generative model, using the metadata as training data to train a model.
  • the model may be trained to output information usable to determine an action to take based on input data.
  • the training of the model may be further based on data comprising at least one of measured and simulated data.
  • the steps of the method may be repeated to retrain the model in response to receiving new metadata.
  • the trained model may be used in the method according to the embodiments described above, and wherein feedback on the performance of the model is used to retrain the model.
  • the model may be initially trained using a measured or simulated dataset.
  • the method may further comprise sending at least one of: a request for a type of training data; an indication of the range of values within which each training data type is required; an indication that the training data can further comprise generated metadata; information indicating the conditions for metadata generation; information related to the generative model to be used to generate the metadata.
  • the method may further comprise receiving at least one of: an indication of the type of data received; for each type of data received, receiving an indication of whether the data comprises generated metadata; an indication of which part of the data comprises generated metadata; an indication of the accuracy of the metadata; information on the generative model used to generate the metadata; and wherein the method further comprises using the received information to train the model.
  • the method may further comprise sending information on the trained model.
  • the method may be performed by a model training function.
  • the controller module may be comprised in an Operation, Administration, and Management entity, OAM.
  • Figure 7 depicts a method in accordance with particular embodiments.
  • the method 7 may be performed by a proxy module.
  • the proxy module may be configured to train a generative model.
  • the method may begin at step 702 with receiving a dataset, at step 704, receiving metadata, at step 706, using the dataset and metadata to train a generative model to generate metadata based on an input dataset.
  • the received metadata may be at least one of: received explicitly; received implicitly; architectural based metadata.
  • the training may be further performed based on a condition comprising information relating to a contextual scenario.
  • the method may further comprise receiving characteristics of an experiment and using the characteristics to train the generative model.
  • the method may further comprise receiving an action and using the action to train the generative model.
  • the steps of the method may be repeated to retrain the generative model in response to receiving at least one of new metadata, a new dataset.
  • the metadata may comprise at least one of: an observation; a state; a reward.
  • the generative model may comprise at least one of: a general adversarial network, GAN; a generative minimization network; a variational auto-encoder; a conditional GAN; an InfoGAN; a Wasserstein GAN; a neural network; a deep neural network.
  • the method may be implemented in a RAN framework.
  • the methods may relate to any of: handover; energy consumption; traffic characteristics measurement; core network measurements; network load measurements; network performance measurements; slice related measurements; UE related analytics; UE congestions related measurements; QoS sustainability measurements.
  • a network node comprising at least one of a controller module and a proxy module, and wherein the network node is configured to perform any of the methods described herein.
  • the methods 4-7 may be performed by a network node (e.g. the network node 1710 or network node 1900 as described later with reference to Figures 17 and 19 respectively).
  • a UE comprising at least one of a controller module and a proxy module, and wherein the UE is configured to perform any of the methods described herein.
  • the method 4-7 may be performed by a UE or wireless device (e.g. the UE 1712 or UE 1800 as described later with reference to Figures 17 and 18 respectively).
  • Optimal handover (HO) decisions use-case is considered to elaborate the efficiency of the use-case described herein. This example is included for reasons of clarity. The person skilled in the art can understand that the methods are applicable to any other AI/ML supported use case.
  • proxy module is used to identify the functionality that generates metadata.
  • functionalities may be part of one or more of the existing functions in the AI/ML framework or it could be a separate function interacting with existing functionalities.
  • the proxy module may generate conditional data (related to HO decisions). Upon request from the Model Inference function, the proxy module may signal back such conditional data (or metadata) to the Model Inference function.
  • the Model Inference function also identified herein as "controller module”, may then use such conditionally generated input, to make proactive decision on when the handover shall take place, so the UE has a smaller drop in the overall received QoS.
  • This block represents the data collection module (as also called in 3GPP TR 37.817), however in this invention, many entities that could exist in the dataset source block are considered, e.g., such entities could be categorized into:
  • Data generating entity that can be probed to send or receive measurement data
  • This entity is what is referred to herein as a “Proxy Module”, which is also explained below.
  • Realistic datasets entity responsible for collecting the measurement obtained via requests from a radio node or a conducting measurement campaign or some data collection procedure. Once the framework executes, the dataset may keep updating and being enriched.
  • This proposed entity may be placed as part of the data collection block or at the model training or inferences host as in Error! Reference source not found.8.
  • the proxy module comprises a generative AI/ML model that receives information from different entities across the AI/ML framework and send back Al-I nformation to different entities.
  • the purpose of the generative model is to produce additional (virtual) data based on given conditions and descriptive metadata.
  • the generated (virtual) data is input to the controller module (model training or inference host).
  • model training or inference host A possible implementation of such generative model is training a generative adversial network, although this disclosure does not exclude all other generative methods and models.
  • the following information could be send-out of the PM module: o
  • the requested forecasted measurement namely the metadata generated by the PM, constituting data that are missing at the Controller Module, and that are useful to optimize the AI/ML process o
  • Information describing that the data signaled are metadata, namely they are conditionally generated data and not, e.g. data derived from field tests. Such information may also include the type of metadata signaled, e.g. type of UE measurement, type of RAN measurement, type of system performance measurement.
  • Al-information describing the models which used to generate such meta-data or conditions.
  • Assistant information represented as the conditions which helped the PM to generate the forecasted measurement or requested Al-information
  • such information could be represented as: o Contextual information considered to generate the metadata provided:
  • ⁇ UE radio capability related context information o Details on how each of those contextual/conditional information was used in training the generative model. o Given a transmitted Al-I nformation (model or Al-Data as output), PM may send the difference between this version of Al -I nformation and previous one that was sent without the operation/involvement of PM.
  • the following information could be received by the PM module: o Request from inference or training host (or real data-source) to generate specific (virtual) measurements/metadata o
  • the request may include the type of metadata required, e.g. cell resource utilization, UE RSRP cell measurements, etc. o
  • the request may include the range within which the requested metadata shall be generated, e.g. for UE RSRP measurements, the range could be -110dB and -120dB o
  • Specific conditions to be used for the generation of metadata e.g. specific radio conditions, e.g. assumed pathloss, or specific load conditions, e.g. PRB utilization.
  • Specific generative models to be used namely specifying the model to use to generate the metadata o
  • Corresponding training for the model to be used to generate metadata namely details on how the model has been/could be trained before generating metadata.
  • Controller module [0152]
  • controller module The purpose of the controller module is to take the data and conditions produced from the proxy module and pass/suggest actions and observe states of the environment. For example, such an action can be the decision to make a handover to a new cell.
  • a General Proposed Framework is shown in Figure 9.
  • the Model Training Function requests training data from the Data Collection function.
  • data may comprise one or more of the following: o
  • An indication for training data types e.g. RSRP measurements, cell Load measurements etc.
  • An indication of the range of values within which each training data type is required o
  • An indication that the training data needed may also comprise metadata o
  • the conditions for such metadata generation may be indicated, e.g. radio conditions, load conditions, transport network conditions o
  • the Model Training function may also indicate details concerning the generative model that the Data Collection function could adopt. Such details may point at a specific generative model or they may point at one or more model characteristics
  • the Data Collection function processes the request from the Model Training Function and derives the data required.
  • the Data Collection function signals to the Model Training function the training data required.
  • Such signalling may include one or more of the following: o
  • the type of data provided to the Model Training function e.g. RSRP measurements, cell load measurements, etc.
  • an indication of whether the data include conditionally generated metadata or not o
  • For each type of data a list of data values that has not been generated in a conditional way, as well as a list of conditionally generated metadata values o
  • an optional accuracy or uncertainty score indicating with what level of error such data may compare with real non-conditionally generated data o
  • the Model Training function is able to further train an AI/ML model and to improve its predictions accuracy also due to the conditionally generated data, which may fill the gaps of non- conditionally generated training data ranges available.
  • Model Inference Function requests model inference input data from the Data Collection function.
  • Such data may comprise one or more of the following: o An indication for input data types, e.g. RSRP measurements, cell Load measurements etc. o An indication of the range of values within which each input data type is required o An indication that the training data needed may also comprise metadata o
  • the conditions for such metadata generation may be indicated, e.g. radio conditions, load conditions, transport network conditions o
  • the Model Inference function may also indicate details concerning the generative model that the Data Collection function could adopt. Such details may point at a specific generative model or they may point at one or more model characteristics [0162] In 2, the Data Collection function processes the request form the Model Inference Function and derives the data required.
  • the Data Collection function signals to the Model Inference function the input data required.
  • Such signalling may include one or more of the following: o
  • the type of data provided to the Model Inference function e.g. RSRP measurements, cell load measurements, etc.
  • o For each type of data, an indication of whether the data include conditionally generated metadata or not o
  • a list of data values that has not been generated in a conditional way, as well as a list of conditionally generated metadata values o
  • an optional accuracy or uncertainty score indicating with what level of error such data may compare with real non-conditionally generated data o
  • details about the generative model used to generate such metadata For each type of metadata signalled, details about the generative model used to generate such metadata
  • the Model Inference function is able to run inference that would not otherwise be possible due to the rare availability of the requested metadata inputs.
  • Such inference process may derive model outputs that may reveal system behaviours and performance in rarely occurring use cases and events. This can be used to optimise the system, for
  • Figure 12 shows an example of the methods described on metadata generation and usage.
  • the 0AM hosting the Model Training function trains a model in step 5 thanks to training data previously received from the RAN.
  • the 0AM system may request to the RAN further conditionally generated training metadata, as per the description in section 6.1.4 [in TS 38.300],
  • the RAN provides such conditionally generated training metadata in step 6b, as described in section 6.1.4 [in TS 38.300], With such metadata the 0AM is able to retrain the Al /ML Model and deploy an update of such model to the RAN (step 6c and 6d, respectively).
  • step 7a and 8a the RAN is able to request to other RAN nodes and to the UE for metadata inputs. Such request follows the descriptions in section 6.1.4. in steps 7b and 8b, the RAN receives the requested metadata as described in section 6.1.4. the RAN is therefore able to sum metadata input based predictions in step 9 and eventually to use such predictions to optimize the system performance.
  • the RAN (or in general the node hosting the model inference function) may include the result of the metadata generated predictions and signal it to the model training function and/or to the 0AM.
  • the model training function and/or the 0AM may optimize the system on the basis of such generated predictions.
  • Such scenarios can be table based, where in the entries of specific characteristics (i.e. number of connected users, utilization of Base Stations, area of deployment, burstiness, volume, RTT) have labels. Based on those labels the NWDAF can find out which are the network entities and for which parts of the network the data collection shall take place.
  • PM send observations to Control module based on the current requirements. For example, when a timer for data collection has expired or the quality metric was reached.
  • CM can request more data or extend the time window to allow more measurements in case the collected data set did not meet specific criteria.
  • CM send proactive action towards the PM (and the%) that can be used to enrich the dataset.
  • the filter/validation module receive updating sample... some interactions between the following entities occur:
  • NWPNFConsumer sends a request of NWDAF related measurement from Translatelnterface.
  • the Translatelnterface send a confirmation of registration, and inform whether to register at NWDAF or RANDAF.
  • NWPNFConsumer requests NWDAFRelatedMeasurement (or one of its alternatives as described above).
  • the NWDAF may need for each NF targeted instance to subscribe to CAM services to retrieve the target NF resource usage and NF resources configuration following steps captured in clause 6.2.3.2 (in TS-23.288) for data collection from CAM. Steps 2-5 may be skipped when e.g. the NWDAF already has the requested analytics.
  • NWPNF network proxy NF
  • the NWDAF subscribes to changes on the load and status of NF instances registered in NRF and identified by their NF id from NRF using Nnrf_NFManagement_NFStatusSubscribe service operation for each NF instance.
  • NRF notifies NWDAF of changes on the load and status of the requested NWPNFConsumer instances by using Nnrf_NFManagement_NFStatusNotify service operation.
  • the NWDAF derives requested analytics.
  • the NWDAF provide requested NWPNFConsumer load analytics to the NWPNFConsumer along with the corresponding Validity Period, using either the Nnwdaf_Analyticslnfo_Request response or Nnwdaf_AnalyticsSubscription_Subscribe response, depending on the service used in step 1. (17-20). If at step 1 the NF has subscribed to receive continuous reporting of NF load analytics, the NWDAF may generate new analytics and, when relevant according to the Analytics target period and Reporting Threshold, provide them along with the corresponding Validity Period to the NF upon reception of notification of new NF load information from 0AM or NRF.
  • Dataset original and updated.
  • the dataset module is the collected measurement obtained via a measurement campaign or over some simulation.
  • Proxy Module comprises a generative model.
  • a generative Al model is trained using the existing dataset with conditioning and meta-data learning possibility.
  • the utilized Al generative model receives input from controller module to steer the latent space of generative distribution and dynamically generates different datasets outputs.
  • Proxy module generation process The proxy module focuses on generating
  • the proxy module could contain a time series generating functionality block. Such block will further consider o Aperiodic forecasted (in time or place) generation of samples, such that based on a request from (e.g.,) Controller module it generates a forecasted sample [t,t+T] for event based scenario. o Tweaking the periodicity of generating the samples and send it to the interaction module.
  • Proxy module [0186] Many generative models could be used in the proxy modules, for instance:
  • Generative Adversarial Network which utilizes the concept of adversary between the generator and discriminator. Where the generator uses gradient descent steps toward minimizing discriminator loss, while the discriminator performs a gradient ascent step toward maximizing discriminator loss. Hence, there is no guarantee of convergence.
  • Generative Minimization network has both generator and discriminator to minimize a loss function rather than working in opposite direction in a competitive fashion, which enable better guarantee of convergence.
  • Variational auto-encoder is another technique that can be used to generate, and enrich the dataset based on the measurement campaign.
  • Any appropriate generative model may be used, for example, conditional GANs, InfoGANs, Wasserstein GANs, and any other type of methods that falls under the same class of machine learning algorithms, including deep neural network approaches.
  • the controller module is connected to the Proxy module to: a) Receive the environment observations generated by the Proxy module. b) Send 1) proactive actions and 2) metadata to the Proxy module. These can be conditional arguments towards to proxy module and do not need to be generated at the same time. c) Send requests on forecasting time intervals to the proxy module. d) Send requests for generation of specific mix of scenarios that help the reinforcement learning in the controller module to gain more experience.
  • controller module is similar to the proxy module and same methods can be applied here as were described above ("Implementation of Proxy module”).
  • the proxy module is connected to: o Controller Module, o PM receives from CM: the CM's actions and other processed meta-data. o PM transmits to CM: the CM's observations, states, rewards. o Updated Dataset o PM receives from updated dataset, its training input to the discriminator o PM sends to updated dataset, its generated output, via validation module o Validation Module o PM sends to updated dataset, its generated output, via validation module
  • the input of the training is: o updated dataset (or the dataset at its first phase), o meta-data (that refer to conditions and also describe dataset characteristics), section 6.1.6.9 illustrates more on the meta-data. o Characteristics of new experiment if required. Section 6.1.6.6 elaborates more on such characteristics.
  • the output comprises: o All Observations that will be used in the CM training, those observationswill depend on the addressed use-case o Potential reward that will be part of input towards the CM training. More specifically at the initialization phase the reward is one of the actions of the actual dataset and in the inference phase is the actions of the training phase of the CM. o In initialization phase, the reward will be the one on the actions of the actual dataset o In inference phase, the reward will be the one on the actions of the training phase of the CM.
  • the training of controller module is briefly explained here.
  • the training starts once the proxy module has reached some specific criterion.
  • the input is the observations as these were generated from the proxy module.
  • these can be CSI, RSRP, GPS measurements.
  • the reward can be the target that is being optimized for the specific use-case.
  • Such target can be for example the maximization of the throughput.
  • the output of the controller module are the actions that it takes. In one example that is changing the association of the UE with another gNB.
  • Reference source not found.9 illustrates a basic architecture where the main three components (dataset, proxy module, and controller module) are connected to one another as discussed in a previous section, to train of the proxy and controller module, to provide virtual observation from proxy, and provide proactive action from the controller module.
  • Such basic framework also considers the possibility of updating the dataset via feedback loop from the proxy module to the newly block "Updated Real-Dataset”. This is beneficial since the controller might request from proxy some new mix of dataset scenarios to enrich its RL agent learning and experience.
  • Second architecture is an extension of the first one, and it described in Error! Reference source not found.15.
  • the first extension of the basic framework is the possibility of adding a control and validation mechanism to control how the generated samples from the proxy module can be used to enrich the updated Real- Dataset. Such a control mechanism is described further in section 6.1.6.4.
  • the second extension of the framework is to have the possibility of using a new dataset (e.g., from real measurement campaign) to enrich the experience of the proxy module.
  • the third extension is the possibility/functionality to request a new measurement campaign with specific scenarios (details are in section 6.1.6.5) to enrich the experience of proxy and controller modules.
  • Section [0212] further describes the characteristics of new experiments.
  • Figure 16 shows Elaborative features of the mobility experiment, in particular relations between different features. Proxy module could learn interdependencies and generate new samples.
  • Example from real-experiment related to Proxy module activity o Nb UE: Experiment shows KPI metrics (e.g., 'TB Size'/ 'serving_cell_snr_x_mean_maj') for specific number of UE performing DL data transmission, with discrete values (like: 10, 20).
  • the proposed Proxy module may generate similar KPI metrics (e.g., 'TB Size'/ 'serving_cell_snr_x_mean_maj') for different number of UEs in the DL after learning the dependencies among them.
  • o MCS Experiment shows KPI metrics ('MCS7 'serving_cell_snr_x_mean_maj').
  • the proposed Proxy module may learn the relationship between both KPI metrics and also given other metadata.
  • the proposed Proxy module will be able to generate new 'serving_cell_snr_x_mean_maj' for in the data nonexistent 'MCS' values and vice versa.
  • Example from real experiment related to Controller module activity o MCS: Controller Module decides to take action in form of increasing 'MCS'. Since the resulting data is not included in the original dataset, the Proxy Module generates the new "sample” and provides the virtual observation, including states (snr, 'TB Size', 'Num RBs'.%) and rewards ('throughput_DL') of the previous controller action, to the Controller Module, which can then decide on the new action (another 'MCS'index).
  • PM proxy module
  • Such latent variable could be found via fixing specific KPI (TP, Latency, Reliability, Coverage, etc) for training/updating the latent variable encoder.
  • TP Latency, Reliability, Coverage, etc
  • o Architectural based metadata which could be found via considering the following architecture technology: o D-MIMO + Dual Connectivity + mmWave + SISO + Side link
  • Figure 17 shows an example of a communication system 1700 in accordance with some embodiments.
  • the communication system 1700 includes a telecommunication network 1702 that includes an access network 1704, such as a radio access network (RAN), and a core network 1706, which includes one or more core network nodes 1708.
  • the access network 1704 includes one or more access network nodes, such as network nodes 1710a and 1710b (one or more of which may be generally referred to as network nodes 1710), or any other similar 3 rd Generation Partnership Project (3GPP) access node or non-3GPP access point.
  • 3GPP 3 rd Generation Partnership Project
  • the network nodes 1710 facilitate direct or indirect connection of user equipment (UE), such as by connecting UEs 1712a, 1712b, 1712c, and 1712d (one or more of which may be generally referred to as UEs 1712) to the core network 1706 over one or more wireless connections.
  • UE user equipment
  • Example wireless communications over a wireless connection include transmitting and/or receiving wireless signals using electromagnetic waves, radio waves, infrared waves, and/or other types of signals suitable for conveying information without the use of wires, cables, or other material conductors.
  • the communication system 1700 may include any number of wired or wireless networks, network nodes, UEs, and/or any other components or systems that may facilitate or participate in the communication of data and/or signals whether via wired or wireless connections.
  • the communication system 1700 may include and/or interface with any type of communication, telecommunication, data, cellular, radio network, and/or other similar type of system.
  • the UEs 1712 may be any of a wide variety of communication devices, including wireless devices arranged, configured, and/or operable to communicate wirelessly with the network nodes 1710 and other communication devices.
  • the network nodes 1710 are arranged, capable, configured, and/or operable to communicate directly or indirectly with the UEs 1712 and/or with other network nodes or equipment in the telecommunication network 1702 to enable and/or provide network access, such as wireless network access, and/or to perform other functions, such as administration in the telecommunication network 1702.
  • the core network 1706 connects the network nodes 1710 to one or more hosts, such as host 1716. These connections may be direct or indirect via one or more intermediary networks or devices. In other examples, network nodes may be directly coupled to hosts.
  • the core network 1706 includes one more core network nodes (e.g., core network node 1708) that are structured with hardware and software components. Features of these components may be substantially similar to those described with respect to the UEs, network nodes, and/or hosts, such that the descriptions thereof are generally applicable to the corresponding components of the core network node 1708.
  • Example core network nodes include functions of one or more of a Mobile Switching Center (MSC), Mobility Management Entity (MME), Home Subscriber Server (HSS), Access and Mobility Management Function (AMF), Session Management Function (SMF), Authentication Server Function (AUSF), Subscription Identifier De-concealing function (SIDF), Unified Data Management (UDM), Security Edge Protection Proxy (SEPP), Network Exposure Function (NEF), and/or a User Plane Function (UPF).
  • MSC Mobile Switching Center
  • MME Mobility Management Entity
  • HSS Home Subscriber Server
  • AMF Access and Mobility Management Function
  • SMF Session Management Function
  • AUSF Authentication Server Function
  • SIDF Subscription Identifier De-concealing function
  • UDM Unified Data Management
  • SEPP Security Edge Protection Proxy
  • NEF Network Exposure Function
  • UPF User Plane Function
  • the host 1716 may be under the ownership or control of a service provider other than an operator or provider of the access network 1704 and/or the telecommunication network 1702, and may be operated by the service provider or on behalf of the service provider.
  • the host 1716 may host a variety of applications to provide one or more services. Examples of such applications include the provision of live and/or pre-recorded audio/video content, data collection services, for example, retrieving and compiling data on various ambient conditions detected by a plurality of UEs, analytics functionality, social media, functions for controlling or otherwise interacting with remote devices, functions for an alarm and surveillance center, or any other such function performed by a server.
  • the communication system 1700 of Figure 17 enables connectivity between the UEs, network nodes, and hosts.
  • the communication system may be configured to operate according to predefined rules or procedures, such as specific standards that include, but are not limited to: Global System for Mobile Communications (GSM); Universal Mobile Telecommunications System (UMTS); Long Term Evolution (LTE), and/or other suitable 2G, 3G, 4G, 5G standards, or any applicable future generation standard (e.g., 6G); wireless local area network (WLAN) standards, such as the Institute of Electrical and Electronics Engineers (IEEE) 802.11 standards (WiFi); and/or any other appropriate wireless communication standard, such as the Worldwide Interoperability for Microwave Access (WiMax), Bluetooth, Z-Wave, Near Field Communication (NFC) ZigBee, LiFi, and/or any low-power wide-area network (LPWAN) standards such as LoRa and Sigfox.
  • GSM Global System for Mobile Communications
  • UMTS Universal Mobile Telecommunications System
  • LTE Long Term Evolution
  • the telecommunication network 1702 is a cellular network that implements 3GPP standardized features. Accordingly, the telecommunications network 1702 may support network slicing to provide different logical networks to different devices that are connected to the telecommunication network 1702. For example, the telecommunications network 1702 may provide Ultra Reliable Low Latency Communication (URLLC) services to some UEs, while providing Enhanced Mobile Broadband (eMBB) services to other UEs, and/or Massive Machine Type Communication (mMTC)ZMassive loT services to yet further UEs.
  • URLLC Ultra Reliable Low Latency Communication
  • eMBB Enhanced Mobile Broadband
  • mMTC Massive Machine Type Communication
  • the UEs 1712 are configured to transmit and/or receive information without direct human interaction.
  • a UE may be designed to transmit information to the access network 1704 on a predetermined schedule, when triggered by an internal or external event, or in response to requests from the access network 1704.
  • a UE may be configured for operating in single- or multi-RAT or multistandard mode.
  • a UE may operate with any one or combination of Wi-Fi, NR (New Radio) and LTE, i.e. being configured for multi-radio dual connectivity (MR-DC), such as E-UTRAN (Evolved-UMTS Terrestrial Radio Access Network) New Radio - Dual Connectivity (EN-DC).
  • MR-DC multi-radio dual connectivity
  • the hub 1714 communicates with the access network 1704 to facilitate indirect communication between one or more UEs (e.g., UE 1712c and/or 1712d) and network nodes (e.g., network node 1710b).
  • the hub 1714 may be a controller, router, a content source and analytics node, or any of the other communication devices described herein regarding UEs.
  • the hub 1714 may be a broadband router enabling access to the core network 1706 for the UEs.
  • the hub 1714 may be a controller that sends commands or instructions to one or more actuators in the UEs.
  • the hub 1714 may be a data collector that acts as temporary storage for UE data and, in some embodiments, may perform analysis or other processing of the data.
  • the hub 1714 may be a content source. For example, for a UE that is a VR headset, display, loudspeaker or other media delivery device, the hub 1714 may retrieve VR assets, video, audio, or other media or data related to sensory information via a network node, which the hub 1714 then provides to the UE either directly, after performing local processing, and/or after adding additional local content.
  • the hub 1714 acts as a proxy server or orchestrator for the UEs, in particular in if one or more of the UEs are low energy loT devices.
  • the hub 1714 may have a constant/persistent or intermittent connection to the network node 1710b.
  • the hub 1714 may also allow for a different communication scheme and/or schedule between the hub 1714 and UEs (e.g., UE 1712c and/or 1712d), and between the hub 1714 and the core network 1706.
  • the hub 1714 is connected to the core network 1706 and/or one or more UEs via a wired connection.
  • the hub 1714 may be configured to connect to an M2M service provider over the access network 1704 and/or to another UE over a direct connection.
  • UEs may establish a wireless connection with the network nodes 1710 while still connected via the hub 1714 via a wired or wireless connection.
  • the hub 1714 may be a dedicated hub - that is, a hub whose primary function is to route communications to/from the UEs from/to the network node 1710b.
  • the hub 1714 may be a non-dedicated hub - that is, a device which is capable of operating to route communications between the UEs and network node 1710b, but which is additionally capable of operating as a communication start and/or end point for certain data channels.
  • FIG. 18 shows a UE 1800 in accordance with some embodiments.
  • a UE refers to a device capable, configured, arranged and/or operable to communicate wirelessly with network nodes and/or other UEs.
  • Examples of a UE include, but are not limited to, a smart phone, mobile phone, cell phone, voice over IP (VoIP) phone, wireless local loop phone, desktop computer, personal digital assistant (PDA), wireless camera, gaming console or device, music storage device, playback appliance, wearable terminal device, wireless endpoint, mobile station, tablet, laptop, laptop-embedded equipment (LEE), laptop-mounted equipment (LME), smart device, wireless customer-premise equipment (CPE), vehicle-mounted or vehicle embedded/integrated wireless device, etc.
  • VoIP voice over IP
  • PDA personal digital assistant
  • LME laptop-embedded equipment
  • LME laptop-mounted equipment
  • CPE wireless customer-premise equipment
  • UEs identified by the 3rd Generation Partnership Project (3GPP), including a narrow band internet of things (NB-loT) UE, a machine type communication (MTC) UE, and/or an enhanced MTC (eMTC) UE.
  • 3GPP 3rd Generation Partnership Project
  • NB-loT narrow band internet of things
  • MTC machine type communication
  • eMTC enhanced MTC
  • a UE may support device-to-device (D2D) communication, for example by implementing a 3GPP standard for sidelink communication, Dedicated Short-Range Communication (DSRC), vehicle-to-vehicle (V2V), vehicle-to-infrastructure (V2I), or vehicle-to-everything (V2X).
  • D2D device-to-device
  • DSRC Dedicated Short-Range Communication
  • V2V vehicle-to-vehicle
  • V2I vehicle-to-infrastructure
  • V2X vehicle-to-everything
  • a UE may not necessarily have a user in the sense of a human user who owns and/or operates the relevant device.
  • a UE may represent a device that is intended for sale to, or operation by, a human user but which may not, or which may not initially, be associated with a specific human user (e.g., a smart sprinkler controller).
  • a UE may represent a device that is not intended for sale
  • the UE 1800 includes processing circuitry 1802 that is operatively coupled via a bus 1804 to an input/output interface 1806, a power source 1808, a memory 1810, a communication interface 1812, and/or any other component, or any combination thereof.
  • Certain UEs may utilize all or a subset of the components shown in Figure 18. The level of integration between the components may vary from one UE to another UE. Further, certain UEs may contain multiple instances of a component, such as multiple processors, memories, transceivers, transmitters, receivers, etc.
  • the processing circuitry 1802 is configured to process instructions and data and may be configured to implement any sequential state machine operative to execute instructions stored as machine-readable computer programs in the memory 1810.
  • the processing circuitry 1802 may be implemented as one or more hardware-implemented state machines (e.g., in discrete logic, field-programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), etc.); programmable logic together with appropriate firmware; one or more stored computer programs, general-purpose processors, such as a microprocessor or digital signal processor (DSP), together with appropriate software; or any combination of the above.
  • the processing circuitry 1802 may include multiple central processing units (CPUs).
  • the processing circuitry 1802 may be operable to provide, either alone or in conjunction with other UE 1800 components, such as the memory 1810, UE 1800 functionality.
  • the processing circuitry 1802 may be configured to cause the UE 1802 to perform the methods as described with reference to Figure 4 - 7.
  • the input/output interface 1806 may be configured to provide an interface or interfaces to an input device, output device, or one or more input and/or output devices.
  • Examples of an output device include a speaker, a sound card, a video card, a display, a monitor, a printer, an actuator, an emitter, a smartcard, another output device, or any combination thereof.
  • An input device may allow a user to capture information into the UE 1800.
  • Examples of an input device include a touch-sensitive or presence-sensitive display, a camera (e.g., a digital camera, a digital video camera, a web camera, etc.), a microphone, a sensor, a mouse, a trackball, a directional pad, a trackpad, a scroll wheel, a smartcard, and the like.
  • the presencesensitive display may include a capacitive or resistive touch sensor to sense input from a user.
  • a sensor may be, for instance, an accelerometer, a gyroscope, a tilt sensor, a force sensor, a magnetometer, an optical sensor, a proximity sensor, a biometric sensor, etc., or any combination thereof.
  • An output device may use the same type of interface port as an input device. For example, a Universal Serial Bus (USB) port may be used to provide an input device and an output device.
  • USB Universal Serial Bus
  • the power source 1808 is structured as a battery or battery pack. Other types of power sources, such as an external power source (e.g., an electricity outlet), photovoltaic device, or power cell, may be used.
  • the power source 1808 may further include power circuitry for delivering power from the power source 1808 itself, and/or an external power source, to the various parts of the UE 1800 via input circuitry or an interface such as an electrical power cable. Delivering power may be, for example, for charging of the power source 1808.
  • Power circuitry may perform any formatting, converting, or other modification to the power from the power source 1808 to make the power suitable for the respective components of the UE 1800 to which power is supplied.
  • the memory 1810 may be or be configured to include memory such as random access memory (RAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), magnetic disks, optical disks, hard disks, removable cartridges, flash drives, and so forth.
  • the memory 1810 includes one or more application programs 1814, such as an operating system, web browser application, a widget, gadget engine, or other application, and corresponding data 1816.
  • the memory 1810 may store, for use by the UE 1800, any of a variety of various operating systems or combinations of operating systems.
  • the memory 1810 may be configured to include a number of physical drive units, such as redundant array of independent disks (RAID), flash memory, USB flash drive, external hard disk drive, thumb drive, pen drive, key drive, high-density digital versatile disc (HD-DVD) optical disc drive, internal hard disk drive, Blu-Ray optical disc drive, holographic digital data storage (HDDS) optical disc drive, external mini-dual in-line memory module (DIMM), synchronous dynamic random access memory (SDRAM), external micro-DIMM SDRAM, smartcard memory such as tamper resistant module in the form of a universal integrated circuit card (UICC) including one or more subscriber identity modules (SIMs), such as a USIM and/or ISIM, other memory, or any combination thereof.
  • RAID redundant array of independent disks
  • HD-DVD high-density digital versatile disc
  • HDDS holographic digital data storage
  • DIMM external mini-dual in-line memory module
  • SDRAM synchronous dynamic random access memory
  • SDRAM synchronous dynamic random access memory
  • the UICC may for example be an embedded UICC (eUlCC), integrated UICC (iUICC) or a removable UICC commonly known as ‘SIM card.
  • the memory 1810 may allow the UE 1800 to access instructions, application programs and the like, stored on transitory or non-transitory memory media, to offload data, or to upload data.
  • An article of manufacture, such as one utilizing a communication system may be tangibly embodied as or in the memory 1810, which may be or comprise a device-readable storage medium.
  • the processing circuitry 1802 may be configured to communicate with an access network or other network using the communication interface 1812.
  • the communication interface 1812 may comprise one or more communication subsystems and may include or be communicatively coupled to an antenna 1822.
  • the communication interface 1812 may include one or more transceivers used to communicate, such as by communicating with one or more remote transceivers of another device capable of wireless communication (e.g., another UE or a network node in an access network).
  • Each transceiver may include a transmitter 1818 and/or a receiver 1820 appropriate to provide network communications (e.g., optical, electrical, frequency allocations, and so forth).
  • the transmitter 1818 and receiver 1820 may be coupled to one or more antennas (e.g., antenna 1822) and may share circuit components, software or firmware, or alternatively be implemented separately.
  • communication functions of the communication interface 1812 may include cellular communication, Wi-Fi communication, LPWAN communication, data communication, voice communication, multimedia communication, short-range communications such as Bluetooth, near-field communication, location-based communication such as the use of the global positioning system (GPS) to determine a location, another like communication function, or any combination thereof.
  • GPS global positioning system
  • Communications may be implemented in according to one or more communication protocols and/or standards, such as IEEE 802.11 , Code Division Multiplexing Access (CDMA), Wideband Code Division Multiple Access (WCDMA), GSM, LTE, New Radio (NR), UMTS, WiMax, Ethernet, transmission control protocol/internet protocol (TCP/IP), synchronous optical networking (SONET), Asynchronous Transfer Mode (ATM), QUIC, Hypertext Transfer Protocol (HTTP), and so forth.
  • CDMA Code Division Multiplexing Access
  • WCDMA Wideband Code Division Multiple Access
  • GSM Global System for Mobile communications
  • LTE Long Term Evolution
  • NR New Radio
  • UMTS Worldwide Interoperability for Microwave Access
  • WiMax Ethernet
  • TCP/IP transmission control protocol/internet protocol
  • SONET synchronous optical networking
  • ATM Asynchronous Transfer Mode
  • QUIC Hypertext Transfer Protocol
  • HTTP Hypertext Transfer Protocol
  • a UE may provide an output of data captured by its sensors, through its communication interface 1812, via a wireless connection to a network node.
  • Data captured by sensors of a UE can be communicated through a wireless connection to a network node via another UE.
  • the output may be periodic (e.g., once every 15 minutes if it reports the sensed temperature), random (e.g., to even out the load from reporting from several sensors), in response to a triggering event (e.g., when moisture is detected an alert is sent), in response to a request (e.g., a user initiated request), or a continuous stream (e.g., a live video feed of a patient).
  • a UE comprises an actuator, a motor, or a switch, related to a communication interface configured to receive wireless input from a network node via a wireless connection.
  • the states of the actuator, the motor, or the switch may change.
  • the UE may comprise a motor that adjusts the control surfaces or rotors of a drone in flight according to the received input or controls a robotic arm performing a medical procedure according to the received input.
  • a UE when in the form of an Internet of Things (loT) device, may be a device for use in one or more application domains, these domains comprising, but not limited to, city wearable technology, extended industrial application and healthcare.
  • loT device are devices which are or which are embedded in: a connected refrigerator or freezer, a TV, a connected lighting device, an electricity meter, a robot vacuum cleaner, a voice controlled smart speaker, a home security camera, a motion detector, a thermostat, a smoke detector, a door/window sensor, a flood/moisture sensor, an electrical door lock, a connected doorbell, an air conditioning system like a heat pump, an autonomous vehicle, a surveillance system, a weather monitoring device, a vehicle parking monitoring device, an electric vehicle charging station, a smart watch, a fitness tracker, a head-mounted display for Augmented Reality (AR) or Virtual Reality (VR), a wearable for tactile augmentation or sensory enhancement, a water sprinkler, an animal- or item
  • AR Augmented Reality
  • VR
  • a UE may represent a machine or other device that performs monitoring and/or measurements, and transmits the results of such monitoring and/or measurements to another UE and/or a network node.
  • the UE may in this case be an M2M device, which may in a 3GPP context be referred to as an MTC device.
  • the UE may implement the 3GPP NB- loT standard.
  • a UE may represent a vehicle, such as a car, a bus, a truck, a ship and an airplane, or other equipment that is capable of monitoring and/or reporting on its operational status or other functions associated with its operation.
  • a first UE might be or be integrated in a drone and provide the drone's speed information (obtained through a speed sensor) to a second UE that is a remote controller operating the drone.
  • the first UE may adjust the throttle on the drone (e.g. by controlling an actuator) to increase or decrease the drone's speed.
  • the first and/or the second UE can also include more than one of the functionalities described above.
  • a UE might comprise the sensor and the actuator, and handle communication of data for both the speed sensor and the actuators.
  • FIG 19 shows a network node 1900 in accordance with some embodiments.
  • network node refers to equipment capable, configured, arranged and/or operable to communicate directly or indirectly with a UE and/or with other network nodes or equipment, in a telecommunication network.
  • network nodes include, but are not limited to, access points (APs) (e.g., radio access points), base stations (BSs) (e.g., radio base stations, Node Bs, evolved Node Bs (eNBs) and NR NodeBs (gNBs)).
  • APs access points
  • BSs base stations
  • Node Bs Node Bs
  • eNBs evolved Node Bs
  • gNBs NR NodeBs
  • Base stations may be categorized based on the amount of coverage they provide (or, stated differently, their transmit power level) and so, depending on the provided amount of coverage, may be referred to as femto base stations, pico base stations, micro base stations, or macro base stations.
  • a base station may be a relay node or a relay donor node controlling a relay.
  • a network node may also include one or more (or all) parts of a distributed radio base station such as centralized digital units and/or remote radio units (RRUs), sometimes referred to as Remote Radio Heads (RRHs). Such remote radio units may or may not be integrated with an antenna as an antenna integrated radio.
  • RRUs remote radio units
  • RRHs Remote Radio Heads
  • Such remote radio units may or may not be integrated with an antenna as an antenna integrated radio.
  • Parts of a distributed radio base station may also be referred to as nodes in a distributed antenna system (DAS).
  • DAS distributed antenna system
  • network nodes include multiple transmission point (multi-TRP) 5G access nodes, multistandard radio (MSR) equipment such as MSR BSs, network controllers such as radio network controllers (RNCs) or base station controllers (BSCs), base transceiver stations (BTSs), transmission points, transmission nodes, multi-cell/multicast coordination entities (MCEs), Operation and Maintenance (O&M) nodes, Operations Support System (OSS) nodes, Self-Organizing Network (SON) nodes, positioning nodes (e.g., Evolved Serving Mobile Location Centers (E-SMLCs)), and/or Minimization of Drive Tests (MDTs).
  • MSR multistandard radio
  • RNCs radio network controllers
  • BSCs base station controllers
  • BTSs base transceiver stations
  • OFDM Operation and Maintenance
  • OSS Operations Support System
  • SON Self-Organizing Network
  • positioning nodes e.g., Evolved Serving Mobile Location Centers (E-SMLCs)
  • the network node 1900 includes processing circuitry 1902, a memory 1904, a communication interface 1906, and a power source 1908, and/or any other component, or any combination thereof.
  • the network node 1900 may be composed of multiple physically separate components (e.g., a NodeB component and a RNC component, or a BTS component and a BSC component, etc.), which may each have their own respective components.
  • the network node 1900 comprises multiple separate components (e.g., BTS and BSC components)
  • one or more of the separate components may be shared among several network nodes. For example, a single RNC may control multiple NodeBs.
  • each unique NodeB and RNC pair may in some instances be considered a single separate network node.
  • the network node 1900 may be configured to support multiple radio access technologies (RATs).
  • RATs radio access technologies
  • some components may be duplicated (e.g., separate memory 1904 for different RATs) and some components may be reused (e.g., a same antenna 1910 may be shared by different RATs).
  • the network node 1900 may also include multiple sets of the various illustrated components for different wireless technologies integrated into network node 1900, for example GSM, WCDMA, LTE, NR, WiFi, Zigbee, Z-wave, LoRaWAN, Radio Frequency Identification (RFID) or Bluetooth wireless technologies. These wireless technologies may be integrated into the same or different chip or set of chips and other components within network node 1900.
  • RFID Radio Frequency Identification
  • the processing circuitry 1902 may comprise a combination of one or more of a microprocessor, controller, microcontroller, central processing unit, digital signal processor, application-specific integrated circuit, field programmable gate array, or any other suitable computing device, resource, or combination of hardware, software and/or encoded logic operable to provide, either alone or in conjunction with other network node 1900 components, such as the memory 1904, network node 1900 functionality.
  • the processing circuitry 1902 may be configured to cause the network node to perform the methods as described with reference to Figure 4-7.
  • the processing circuitry 1902 includes a system on a chip (SOC). In some embodiments, the processing circuitry 1902 includes one or more of radio frequency (RF) transceiver circuitry 1912 and baseband processing circuitry 1914. In some embodiments, the radio frequency (RF) transceiver circuitry 1912 and the baseband processing circuitry 1914 may be on separate chips (or sets of chips), boards, or units, such as radio units and digital units. In alternative embodiments, part or all of RF transceiver circuitry 1912 and baseband processing circuitry 1914 may be on the same chip or set of chips, boards, or units.
  • SOC system on a chip
  • the processing circuitry 1902 includes one or more of radio frequency (RF) transceiver circuitry 1912 and baseband processing circuitry 1914.
  • the radio frequency (RF) transceiver circuitry 1912 and the baseband processing circuitry 1914 may be on separate chips (or sets of chips), boards, or units, such as radio units and digital units. In alternative embodiments, part or all of RF transceiver
  • the memory 1904 may comprise any form of volatile or non-volatile computer-readable memory including, without limitation, persistent storage, solid-state memory, remotely mounted memory, magnetic media, optical media, random access memory (RAM), read-only memory (ROM), mass storage media (for example, a hard disk), removable storage media (for example, a flash drive, a Compact Disk (CD) or a Digital Video Disk (DVD)), and/or any other volatile or non-volatile, non-transitory device-readable and/or computer-executable memory devices that store information, data, and/or instructions that may be used by the processing circuitry 1902.
  • volatile or non-volatile computer-readable memory including, without limitation, persistent storage, solid-state memory, remotely mounted memory, magnetic media, optical media, random access memory (RAM), read-only memory (ROM), mass storage media (for example, a hard disk), removable storage media (for example, a flash drive, a Compact Disk (CD) or a Digital Video Disk (DVD)), and/or any other volatile or non-
  • the memory 1904 may store any suitable instructions, data, or information, including a computer program, software, an application including one or more of logic, rules, code, tables, and/or other instructions capable of being executed by the processing circuitry 1902 and utilized by the network node 1900.
  • the memory 1904 may be used to store any calculations made by the processing circuitry 1902 and/or any data received via the communication interface 1906.
  • the processing circuitry 1902 and memory 1904 is integrated.
  • the communication interface 1906 is used in wired or wireless communication of signaling and/or data between a network node, access network, and/or UE. As illustrated, the communication interface 1906 comprises port(s)/terminal(s) 1916 to send and receive data, for example to and from a network over a wired connection.
  • the communication interface 1906 also includes radio front-end circuitry 1918 that may be coupled to, or in certain embodiments a part of, the antenna 1910. Radio front-end circuitry 1918 comprises filters 1920 and amplifiers 1922. The radio front-end circuitry 1918 may be connected to an antenna 1910 and processing circuitry 1902. The radio front-end circuitry may be configured to condition signals communicated between antenna 1910 and processing circuitry 1902.
  • the radio front-end circuitry 1918 may receive digital data that is to be sent out to other network nodes or UEs via a wireless connection.
  • the radio front-end circuitry 1918 may convert the digital data into a radio signal having the appropriate channel and bandwidth parameters using a combination of filters 1920 and/or amplifiers 1922.
  • the radio signal may then be transmitted via the antenna 1910.
  • the antenna 1910 may collect radio signals which are then converted into digital data by the radio front-end circuitry 1918.
  • the digital data may be passed to the processing circuitry 1902.
  • the communication interface may comprise different components and/or different combinations of components.
  • the network node 1900 does not include separate radio front-end circuitry 1918, instead, the processing circuitry 1902 includes radio front-end circuitry and is connected to the antenna 1910.
  • the processing circuitry 1902 includes radio front-end circuitry and is connected to the antenna 1910.
  • all or some of the RF transceiver circuitry 1912 is part of the communication interface 1906.
  • the communication interface 1906 includes one or more ports or terminals 1916, the radio front-end circuitry 1918, and the RF transceiver circuitry 1912, as part of a radio unit (not shown), and the communication interface 1906 communicates with the baseband processing circuitry 1914, which is part of a digital unit (not shown).
  • the antenna 1910 may include one or more antennas, or antenna arrays, configured to send and/or receive wireless signals.
  • the antenna 1910 may be coupled to the radio front-end circuitry 1918 and may be any type of antenna capable of transmitting and receiving data and/or signals wirelessly.
  • the antenna 1910 is separate from the network node 1900 and connectable to the network node 1900 through an interface or port.
  • the antenna 1910, communication interface 1906, and/or the processing circuitry 1902 may be configured to perform any receiving operations and/or certain obtaining operations described herein as being performed by the network node. Any information, data and/or signals may be received from a UE, another network node and/or any other network equipment. Similarly, the antenna 1910, the communication interface 1906, and/or the processing circuitry 1902 may be configured to perform any transmitting operations described herein as being performed by the network node. Any information, data and/or signals may be transmitted to a UE, another network node and/or any other network equipment.
  • the power source 1908 provides power to the various components of network node 1900 in a form suitable for the respective components (e.g., at a voltage and current level needed for each respective component).
  • the power source 1908 may further comprise, or be coupled to, power management circuitry to supply the components of the network node 1900 with power for performing the functionality described herein.
  • the network node 1900 may be connectable to an external power source (e.g., the power grid, an electricity outlet) via an input circuitry or interface such as an electrical cable, whereby the external power source supplies power to power circuitry of the power source 1908.
  • the power source 1908 may comprise a source of power in the form of a battery or battery pack which is connected to, or integrated in, power circuitry. The battery may provide backup power could the external power source fail.
  • Embodiments of the network node 1900 may include additional components beyond those shown in Figure 19 for providing certain aspects of the network node's functionality, including any of the functionality described herein and/or any functionality necessary to support the subject matter described herein.
  • the network node 1900 may include user interface equipment to allow input of information into the network node 1900 and to allow output of information from the network node 1900. This may allow a user to perform diagnostic, maintenance, repair, and other administrative functions for the network node 1900.
  • FIG 20 is a block diagram of a host 2000, which may be an embodiment of the host 1716 of Figure 17, in accordance with various aspects described herein.
  • the host 2000 may be or comprise various combinations hardware and/or software, including a standalone server, a blade server, a cloud- implemented server, a distributed server, a virtual machine, container, or processing resources in a server farm.
  • the host 2000 may provide one or more services to one or more UEs.
  • the host 2000 includes processing circuitry 2002 that is operatively coupled via a bus 2004 to an input/output interface 2006, a network interface 2008, a power source 2010, and a memory 2012.
  • Other components may be included in other embodiments. Features of these components may be substantially similar to those described with respect to the devices of previous figures, such as Figures 18 and 19, such that the descriptions thereof are generally applicable to the corresponding components of host 2000.
  • the memory 2012 may include one or more computer programs including one or more host application programs 2014 and data 2016, which may include user data, e.g., data generated by a UE for the host 2000 or data generated by the host 2000 for a UE.
  • Embodiments of the host 2000 may utilize only a subset or all of the components shown.
  • the host application programs 2014 may be implemented in a container-based architecture and may provide support for video codecs (e.g., Versatile Video Coding (VVC), High Efficiency Video Coding (HEVC), Advanced Video Coding (AVC), MPEG, VP9) and audio codecs (e.g., FLAG, Advanced Audio Coding (AAC), MPEG, G.711), including transcoding for multiple different classes, types, or implementations of UEs (e.g., handsets, desktop computers, wearable display systems, heads-up display systems).
  • the host application programs 2014 may also provide for user authentication and licensing checks and may periodically report health, routes, and content availability to a central node, such as a device in or on the edge of a core network.
  • the host 2000 may select and/or indicate a different host for over- the-top services for a UE.
  • the host application programs 2014 may support various protocols, such as the HTTP Live Streaming (HLS) protocol, Real-Time Messaging Protocol (RTMP), Real-Time Streaming Protocol (RTSP), Dynamic Adaptive Streaming over HTTP (MPEG-DASH), etc.
  • HTTP Live Streaming HLS
  • RTMP Real-Time Messaging Protocol
  • RTSP Real-Time Streaming Protocol
  • MPEG-DASH Dynamic Adaptive Streaming over HTTP
  • FIG. 21 is a block diagram illustrating a virtualization environment 2100 in which functions implemented by some embodiments may be virtualized.
  • virtualizing means creating virtual versions of apparatuses or devices which may include virtualizing hardware platforms, storage devices and networking resources.
  • virtualization can be applied to any device described herein, or components thereof, and relates to an implementation in which at least a portion of the functionality is implemented as one or more virtual components.
  • Some or all of the functions described herein may be implemented as virtual components executed by one or more virtual machines (VMs) implemented in one or more virtual environments 2100 hosted by one or more of hardware nodes, such as a hardware computing device that operates as a network node, UE, core network node, or host.
  • VMs virtual machines
  • the virtual node does not require radio connectivity (e.g., a core network node or host)
  • the node may be entirely virtualized.
  • Hardware 2104 includes processing circuitry, memory that stores software and/or instructions executable by hardware processing circuitry, and/or other hardware devices as described herein, such as a network interface, input/output interface, and so forth.
  • Software may be executed by the processing circuitry to instantiate one or more virtualization layers 2106 (also referred to as hypervisors or virtual machine monitors (VMMs)), provide VMs 2108a and 2108b (one or more of which may be generally referred to as VMs 2108), and/or perform any of the functions, features and/or benefits described in relation with some embodiments described herein.
  • the virtualization layer 2106 may present a virtual operating platform that appears like networking hardware to the VMs 2108.
  • the VMs 2108 comprise virtual processing, virtual memory, virtual networking or interface and virtual storage, and may be run by a corresponding virtualization layer 2106.
  • a virtualization layer 2106 Different embodiments of the instance of a virtual appliance 2102 may be implemented on one or more of VMs 2108, and the implementations may be made in different ways.
  • Virtualization of the hardware is in some contexts referred to as network function virtualization (NFV).
  • NFV network function virtualization
  • NFV may be used to consolidate many network equipment types onto industry standard high volume server hardware, physical switches, and physical storage, which can be located in data centers, and customer premise equipment.
  • a VM 2108 may be a software implementation of a physical machine that runs programs as if they were executing on a physical, non-virtualized machine.
  • Each of the VMs 2108, and that part of hardware 2104 that executes that VM be it hardware dedicated to that VM and/or hardware shared by that VM with others of the VMs, forms separate virtual network elements.
  • a virtual network function is responsible for handling specific network functions that run in one or more VMs 2108 on top of the hardware 2104 and corresponds to the application 2102.
  • Hardware 2104 may be implemented in a standalone network node with generic or specific components. Hardware 2104 may implement some functions via virtualization. Alternatively, hardware 2104 may be part of a larger cluster of hardware (e.g. such as in a data center or CPE) where many hardware nodes work together and are managed via management and orchestration 2110, which, among others, oversees lifecycle management of applications 2102.
  • hardware 2104 is coupled to one or more radio units that each include one or more transmitters and one or more receivers that may be coupled to one or more antennas. Radio units may communicate directly with other hardware nodes via one or more appropriate network interfaces and may be used in combination with the virtual components to provide a virtual node with radio capabilities, such as a radio access node or a base station.
  • some signaling can be provided with the use of a control system 2112 which may alternatively be used for communication between hardware nodes and radio units.
  • Figure 22 shows a communication diagram of a host 2202 communicating via a network node 2204 with a UE 2206 over a partially wireless connection in accordance with some embodiments.
  • host 2202 Like host 2000, embodiments of host 2202 include hardware, such as a communication interface, processing circuitry, and memory.
  • the host 2202 also includes software, which is stored in or accessible by the host 2202 and executable by the processing circuitry.
  • the software includes a host application that may be operable to provide a service to a remote user, such as the UE 2206 connecting via an over-the-top (OTT) connection 2250 extending between the UE 2206 and host 2202.
  • OTT over-the-top
  • a host application may provide user data which is transmitted using the OTT connection 2250.
  • the network node 2204 includes hardware enabling it to communicate with the host 2202 and UE 2206.
  • the connection 2260 may be direct or pass through a core network (like core network 1706 of Figure 17) and/or one or more other intermediate networks, such as one or more public, private, or hosted networks.
  • a core network like core network 1706 of Figure 17
  • one or more other intermediate networks such as one or more public, private, or hosted networks.
  • an intermediate network may be a backbone network or the Internet.
  • the UE 2206 includes hardware and software, which is stored in or accessible by UE 2206 and executable by the UE's processing circuitry.
  • the software includes a client application, such as a web browser or operatorspecific "app” that may be operable to provide a service to a human or non-human user via UE 2206 with the support of the host 2202.
  • a client application such as a web browser or operatorspecific "app” that may be operable to provide a service to a human or non-human user via UE 2206 with the support of the host 2202.
  • an executing host application may communicate with the executing client application via the OTT connection 2250 terminating at the UE 2206 and host 2202.
  • the UE's client application may receive request data from the host's host application and provide user data in response to the request data.
  • the OTT connection 2250 may transfer both the request data and the user data.
  • the UE's client application may interact with the user to generate the user data that it provides to the host application through the OTT connection 2
  • the OTT connection 2250 may extend via a connection 2260 between the host 2202 and the network node 2204 and via a wireless connection 2270 between the network node 2204 and the UE 2206 to provide the connection between the host 2202 and the UE 2206.
  • the connection 2260 and wireless connection 2270, over which the OTT connection 2250 may be provided, have been drawn abstractly to illustrate the communication between the host 2202 and the UE 2206 via the network node 2204, without explicit reference to any intermediary devices and the precise routing of messages via these devices.
  • the host 2202 provides user data, which may be performed by executing a host application.
  • the user data is associated with a particular human user interacting with the UE 2206.
  • the user data is associated with a UE 2206 that shares data with the host 2202 without explicit human interaction.
  • the host 2202 initiates a transmission carrying the user data towards the UE 2206.
  • the host 2202 may initiate the transmission responsive to a request transmitted by the UE 2206.
  • the request may be caused by human interaction with the UE 2206 or by operation of the client application executing on the UE 2206.
  • the transmission may pass via the network node 2204, in accordance with the teachings of the embodiments described throughout this disclosure. Accordingly, in step 2212, the network node 2204 transmits to the UE 2206 the user data that was carried in the transmission that the host 2202 initiated, in accordance with the teachings of the embodiments described throughout this disclosure. In step 2214, the UE 2206 receives the user data carried in the transmission, which may be performed by a client application executed on the UE 2206 associated with the host application executed by the host 2202.
  • the UE 2206 executes a client application which provides user data to the host 2202.
  • the user data may be provided in reaction or response to the data received from the host 2202.
  • the UE 2206 may provide user data, which may be performed by executing the client application.
  • the client application may further consider user input received from the user via an input/output interface of the UE 2206. Regardless of the specific manner in which the user data was provided, the UE 2206 initiates, in step 2218, transmission of the user data towards the host 2202 via the network node 2204.
  • the network node 2204 receives user data from the UE 2206 and initiates transmission of the received user data towards the host 2202.
  • the host 2202 receives the user data carried in the transmission initiated by the UE 2206.
  • One or more of the various embodiments improve the performance of OTT services provided to the UE 2206 using the OTT connection 2250, in which the wireless connection 2270 forms the last segment. More precisely, the teachings of these embodiments may improve the amount and type of data available to a network to determine actions to be taken and the consequences thereof, and thereby provide benefits such as a more efficient control operation.
  • factory status information may be collected and analyzed by the host 2202.
  • the host 2202 may process audio and video data which may have been retrieved from a UE for use in creating maps.
  • the host 2202 may collect and analyze real-time data to assist in controlling vehicle congestion (e.g., controlling traffic lights).
  • the host 2202 may store surveillance video uploaded by a UE.
  • the host 2202 may store or control access to media content such as video, audio, VR or AR which it can broadcast, multicast or unicast to UEs.
  • the host 2202 may be used for energy pricing, remote control of non-time critical electrical load to balance power generation needs, location services, presentation services (such as compiling diagrams etc. from data collected from remote devices), or any other function of collecting, retrieving, storing, analyzing and/or transmitting data.
  • a measurement procedure may be provided for the purpose of monitoring data rate, latency and other factors on which the one or more embodiments improve.
  • the measurement procedure and/or the network functionality for reconfiguring the OTT connection may be implemented in software and hardware of the host 2202 and/or UE 2206.
  • sensors (not shown) may be deployed in or in association with other devices through which the OTT connection 2250 passes; the sensors may participate in the measurement procedure by supplying values of the monitored quantities exemplified above, or supplying values of other physical quantities from which software may compute or estimate the monitored quantities.
  • the reconfiguring of the OTT connection 2250 may include message format, retransmission settings, preferred routing etc.; the reconfiguring need not directly alter the operation of the network node 2204. Such procedures and functionalities may be known and practiced in the art.
  • measurements may involve proprietary UE signaling that facilitates measurements of throughput, propagation times, latency and the like, by the host 2202.
  • the measurements may be implemented in that software causes messages to be transmitted, in particular empty or 'dummy' messages, using the OTT connection 2250 while monitoring propagation times, errors, etc.
  • computing devices described herein may include the illustrated combination of hardware components, other embodiments may comprise computing devices with different combinations of components. It is to be understood that these computing devices may comprise any suitable combination of hardware and/or software needed to perform the tasks, features, functions and methods disclosed herein. Determining, calculating, obtaining or similar operations described herein may be performed by processing circuitry, which may process information by, for example, converting the obtained information into other information, comparing the obtained information or converted information to information stored in the network node, and/or performing one or more operations based on the obtained information or converted information, and as a result of said processing making a determination.
  • processing circuitry may process information by, for example, converting the obtained information into other information, comparing the obtained information or converted information to information stored in the network node, and/or performing one or more operations based on the obtained information or converted information, and as a result of said processing making a determination.
  • computing devices may comprise multiple different physical components that make up a single illustrated component, and functionality may be partitioned between separate components.
  • a communication interface may be configured to include any of the components described herein, and/or the functionality of the components may be partitioned between the processing circuitry and the communication interface.
  • non-computationally intensive functions of any of such components may be implemented in software or firmware and computationally intensive functions may be implemented in hardware.
  • processing circuitry executing instructions stored on in memory, which in certain embodiments may be a computer program product in the form of a non-transitory computer-readable storage medium.
  • some or all of the functionality may be provided by the processing circuitry without executing instructions stored on a separate or discrete device-readable storage medium, such as in a hard-wired manner.
  • the processing circuitry can be configured to perform the described functionality.
  • the benefits provided by such functionality are not limited to the processing circuitry alone or to other components of the computing device, but are enjoyed by the computing device as a whole, and/or by end users and a wireless network generally. The following numbered embodiments provide additional information on the disclosure.
  • a method performed by a controller module for processing data comprising: sending a request for metadata relating to a network to a proxy module; responsive to receiving from the proxy module metadata generated by a generative model, using the received metadata as input to a trained model; and determining an action to take based on the output of the trained model.
  • the output of the trained model comprises at least one of: an environmental state; a reward; an observation.
  • a method performed by a proxy module for generating data comprising: receiving, from a controller module, a request for metadata relating to a network; generating metadata using a generative model; and sending the generated metadata to the controller module.
  • a condition comprises information relating to a contextual scenario.
  • the method further comprising: sending at least one of: information indicating that the metadata is conditionally generated; information indicating the type of metadata generated; information indicating the model used to generate the metadata; information indicating the conditions used to generate the metadata; information indicating how conditions used to generate the metadata were used; information on the generative model; metadata values; information indicating the accuracy of the metadata.
  • the method further comprising: receiving at least one of: information indicating the type of metadata to be generated; information indicating the range within which the metadata is to be generated; at least one condition to be used for the generation of the metadata; information indicating at least one generative model to be used to generate the metadata; information indicating how the generative model is to be trained prior to generating the metadata, and wherein the method further comprises using the received information as input to the generative model.
  • the generated metadata comprises at least one of: information relating to data missing from a dataset; forecast metadata.
  • a method in a system comprising a controller module configured to perform the method according to any of embodiment 1 to 10 and a proxy module configured to perform the method according to any of embodiment 11 to 19.
  • method further comprises updating the dataset based on the metadata output from the generative model.
  • an event criteria comprises at least one of: a rare event criteria; a quality wise criteria; extra desired data; metadata is derived using a new experiment.
  • an experiment comprises at least one of: altering network configuration; adding or removing base stations; increasing or reducing coverage; altering radio environment; introducing coverage holes; changing propagation characteristics; adding or removing frequency layers; new channel ranks; user dynamics; more throttling at the cells; dynamic/constant type of traffic; measurement campaign that contains new margins of KPIs; shorter or longer margin of latency; shorter or longer margin of throughput; shorter or longer margin of reliability; shorter or longer margin of SI NR; shorter or longer margin of bandwidth; shorter or longer margin of center frequency.
  • a method performed by a controller module for training a model comprising: sending a request for metadata relating to a network to a proxy module; responsive to receiving metadata from the proxy module generated by a generative model, using the metadata as training data to train a model.
  • the method further comprises: receiving at least one of: an indication of the type of data received; for each type of data received, receiving an indication of whether the data comprises generated metadata; an indication of which part of the data comprises generated metadata; an indication of the accuracy of the metadata; information on the generative model used to generate the metadata; and wherein the method further comprises using the received information to train the model.
  • controller module is comprised in an Operation, Administration, and Management entity, OAM.
  • a method performed by a proxy module for training a generative model comprising: receiving a dataset; receiving metadata; using the dataset and metadata to train a generative model to generate metadata based on an input dataset.
  • the metadata comprises at least one of: an observation; a state; a reward
  • the generative model comprises at least one of: a general adversarial network, GAN; a generative minimization network; a variational auto-encoder; a conditional GAN; an InfoGAN; a Wasserstein GAN; a neural network; a deep neural network. 46. The method of any preceding embodiment, wherein the method is implemented in a RAN framework.
  • the method relates to at least one of: handover; energy consumption; traffic characteristics measurement; core network measurements; network load measurements; network performance measurements; slice related measurements; UE related analytics; UE congestions related measurements; QoS sustainability measurements.
  • a method in a system comprising a controller module and a proxy module, the system to perform the method according to any preceding embodiment.
  • a method in a network node comprising at least one of a proxy module and a controller module, the network node to perform any of embodiments 1 to 48.
  • a method in a UE comprising at least one of a proxy module and a controller module, the UE to perform any of embodiments 1 to 48.
  • a controller module for processing data and/or for training a model comprising: processing circuitry configured to cause the controller module to perform any of the steps of any of embodiments 1 to 48; power supply circuitry configured to supply power to the processing circuitry.
  • a proxy module for generating data and/or for training a generative model comprising: processing circuitry configured to cause the proxy module to perform any of the steps of any of embodiments 1 to 48; power supply circuitry configured to supply power to the processing circuitry.
  • a network node comprising: at least one of a controller module and a proxy module; processing circuitry configured to cause the network node to perform any of the steps of any of embodiments 1 to 48; power supply circuitry configured to supply power to the processing circuitry.
  • a user equipment comprising: at least one of a controller module and a proxy module; processing circuitry configured to cause the user equipment to perform any of the steps of any of embodiments 1 to 48; and power supply circuitry configured to supply power to the processing circuitry.
  • a user equipment comprising: an antenna configured to send and receive wireless signals; radio front-end circuitry connected to the antenna and to processing circuitry, and configured to condition signals communicated between the antenna and the processing circuitry; the processing circuitry being configured to perform any of the steps of any of embodiments 1 to 48; an input interface connected to the processing circuitry and configured to allow input of information into the UE to be processed by the processing circuitry; an output interface connected to the processing circuitry and configured to output information from the UE that has been processed by the processing circuitry; and a battery connected to the processing circuitry and configured to supply power to the UE.
  • UE user equipment
  • a host configured to operate in a communication system to provide an over-the-top (OTT) service, the host comprising: processing circuitry configured to provide user data; and a network interface configured to initiate transmission of the user data to a cellular network for transmission to a user equipment (UE), wherein the UE comprises a communication interface and processing circuitry, the communication interface and processing circuitry of the UE being configured to perform any of the steps of any of the embodiments 1 to 48 to receive the user data from the host.
  • OTT over-the-top
  • the cellular network further includes a network node configured to communicate with the UE to transmit the user data to the UE from the host.
  • any of embodiments 56 and 57 wherein: the processing circuitry of the host is configured to execute a host application, thereby providing the user data; and the host application is configured to interact with a client application executing on the UE, the client application being associated with the host application.
  • UE user equipment
  • a host configured to operate in a communication system to provide an over-the-top (OTT) service, the host comprising: processing circuitry configured to provide user data; and a network interface configured to initiate transmission of the user data to a cellular network for transmission to a user equipment (UE), wherein the UE comprises a communication interface and processing circuitry, the communication interface and processing circuitry of the UE being configured to perform any of the steps of any of embodiments 1 to 48 to transmit the user data to the host.
  • OTT over-the-top
  • the host of embodiment 62, wherein the cellular network further includes a network node configured to communicate with the UE to transmit the user data from the UE to the host.
  • the processing circuitry of the host is configured to execute a host application, thereby providing the user data; and the host application is configured to interact with a client application executing on the UE, the client application being associated with the host application.
  • UE user equipment
  • invention 65 further comprising: at the host, executing a host application associated with a client application executing on the UE to receive the user data from the UE.
  • a host configured to operate in a communication system to provide an over-the-top (OTT) service, the host comprising: processing circuitry configured to provide user data; and a network interface configured to initiate transmission of the user data to a network node in a cellular network for transmission to a user equipment (UE), the network node having a communication interface and processing circuitry, the processing circuitry of the network node configured to perform any of the operations of any of embodiments 1 to 48 to transmit the user data from the host to the UE.
  • OTT over-the-top
  • the processing circuitry of the host is configured to execute a host application that provides the user data; and the UE comprises processing circuitry configured to execute a client application associated with the host application to receive the transmission of user data from the host.
  • UE user equipment
  • a communication system comprising any of a proxy module, a controller module, a validation module, the communication system configured to perform any of the steps of embodiments 1 to 48.
  • a communication system configured to provide an over-the-top service, the communication system comprising: a host comprising: processing circuitry configured to provide user data for a user equipment (UE), the user data being associated with the over-the-top service; and a network interface configured to initiate transmission of the user data toward a cellular network node for transmission to the UE, the network node having a communication interface and processing circuitry, the processing circuitry of the network node configured to perform any of the operations of any of embodiments 1 to 48 to transmit the user data from the host to the UE.
  • a host comprising: processing circuitry configured to provide user data for a user equipment (UE), the user data being associated with the over-the-top service; and a network interface configured to initiate transmission of the user data toward a cellular network node for transmission to the UE, the network node having a communication interface and processing circuitry, the processing circuitry of the network node configured to perform any of the operations of any of embodiments 1 to 48 to transmit the user data from the host to the UE.
  • the communication system of embodiment 74 further comprising: the network node; and/or the user equipment.
  • a host configured to operate in a communication system to provide an over-the-top (OTT) service, the host comprising: processing circuitry configured to initiate receipt of user data; and a network interface configured to receive the user data from a network node in a cellular network, the network node having a communication interface and processing circuitry, the processing circuitry of the network node configured to perform any of the operations of any of embodiments 1 to 48 to receive the user data from a user equipment (UE) for the host.
  • OTT over-the-top
  • the processing circuitry of the host is configured to execute a host application, thereby providing the user data; and the host application is configured to interact with a client application executing on the UE, the client application being associated with the host application.
  • UE user equipment

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Databases & Information Systems (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

Methods and apparatuses for facilitating Artificial Intelligence/Machine Learning operations through the production and/or use of metadata. A method performed by a controller module for processing data comprises sending a request for metadata relating to a network to a proxy module. The method further comprises responsive to receiving from the proxy module metadata generated by a generative model, using the received metadata as input to a trained model. The method also comprises determining an action to take based on the output of the trained model.

Description

Incorporating Conditions into Data-Collection & AI/ML Operations
Technical Field
[0001] Embodiments of the present disclosure relate to methods, controller modules, proxy modules, network nodes and user equipments (UEs), and particularly methods, controller modules, proxy modules, network nodes and UEs for facilitating Artificial Intelligence (AI)ZMachine Learning (ML) operations through the production and/or use of metadata.
Background
[0002] NG-RAN Overall Architecture
[0003] The NG-Radio Access Network (RAN) comprises a set of New Radio base stations (gNBs) connected to the 5GC through the NG interface. The NG-RAN Overall Architecture is shown in Figure 1.
[0004] As specified in 3rd Generation Partnership Project (3GPP) Technical Specification (TS) 38.300 V16.8.0 (available at https://portal.3gpp.org/desktopmodules/Specifications
/SpecificationDetails.aspx?specificationld=3191 as of 18 January 2023), NG-RAN could also comprise a set of ng- eNBs, an ng-eNB may comprise an ng-eNB-CU and one or more ng-eNB-DU(s). An ng-eNB-CU and an ng-eNB- DU is connected via W1 interface. The general principle described in this section also applies to ng-eNB and W1 interface, if not explicitly specified otherwise.
[0005] A gNB can support FDD mode, TDD mode or dual mode operation. gNBs can be interconnected through the Xn interface. A gNB may comprise a gNB-CU and one or more gNB-DU(s). A gNB-CU and a gNB-DU is connected via F1 interface. One gNB-DU may be connected to only one gNB-CU.
[0006] In a case of network sharing with multiple cell ID broadcast, each Cell Identity associated with a subset of PLMNs may correspond to a gNB-DU and the gNB-CU it is connected to, i.e. the corresponding gNB-DUs share the same physical layer cell resources.
[0007] For resiliency, a gNB-DU may be connected to multiple gNB-CUs by appropriate implementation.
[0008] NG, Xn and F1 are logical interfaces.
[0009] For NG-RAN, the NG and Xn-C interfaces for a gNB comprising a gNB-CU and gNB-DUs, terminate in the gNB-CU. For EN-DC, the S1-U and X2-C interfaces for a gNB comprising a gNB-CU and gNB-DUs, terminate in the gNB-CU. The gNB-CU and connected gNB-DUs are only visible to other gNBs and the 5GC as a gNB. A possible deployment scenario is described in 3GPP TS 38.300 version 16.2.0 (available at the above address as of 18 January 2023).
[0010] The node hosting user plane part of NR PDCP (e.g. gNB-CU, gNB-CU-UP, and for EN-DC, MeNB or SgNB depending on the bearer split) may perform user inactivity monitoring and further inform its inactivity or (re)activation to the node having C-plane connection towards the core network (e.g. over E1, X2). The node hosting NR RLC (e.g. gNB-DU) may perform user inactivity monitoring and further inform its inactivity or (re)activation to the node hosting control plane, e.g. gNB-CU or gNB-CU-CP. [0011] Uplink Packet Data Convergence Protocol (UL PDCP) configuration (i.e. how the UE uses the UL at the assisting node) is indicated via X2-C (for EN-DC), Xn-C (for NG-RAN) and F1-C. Radio Link Outage/Resume for DL and/or UL is indicated via X2-U (for EN-DC), Xn-U (for NG-RAN) and F1-U.
[0012] The NG-RAN is layered into a Radio Network Layer (RNL) and a Transport Network Layer (TNL).
[0013] The NG-RAN architecture, i.e. the NG-RAN logical nodes and interfaces between them, is defined as part of the RNL.
[0014] For each NG-RAN interface (NG, Xn, F1) the related TNL protocol and the functionality are specified. The TNL provides services for user plane transport, signalling transport.
[0015] In NG-Flex configuration, each NG-RAN node is connected to all AMFs of AMF Sets within an AMF Region supporting at least one slice also supported by the NG-RAN node. The AMF Set and the AMF Region are defined in 3GPP TS 23.501 v.17.3.0, available at https://portal.3gpp.org/desktopmodules/Specifications/SpecificationDetails.aspx? specificationld=3144 as of 30 January 2023.
[0016] If security protection for control plane and user plane data on TNL of NG-RAN interfaces is to be supported, NDS/IP 3GPP TS 33.501 v.17.7.0 may be applied, available at https://portal.3gpp. org/desktopmodules/Specifications/SpecificationDetails.aspx?specificationld=3169 as of 30 January 2023.
[0017] Overall architecture for separation of gNB-CU-CP and gNB-CU-UP
[0018] The overall architecture for separation of gNB-CU-CP and gNB-CU-UP is depicted in Error! Reference source not found..
A gNB may comprise a gNB-CU-CP, multiple gNB-CU-UPs and multiple gNB-DUs;
[0019] The gNB-CU-CP is connected to the gNB-DU through the F1-C interface;
[0020] The gNB-CU-UP is connected to the gNB-DU through the F1-U interface;
[0021] The gNB-CU-UP is connected to the gNB-CU-CP through the E1 interface;
[0022] One gNB-DU is connected to only one gNB-CU-CP;
[0023] One gNB-CU-UP is connected to only one gNB-CU-CP;
[0024] NOTE 1 : For resiliency, a gNB-DU and/or a gNB-CU-UP may be connected to multiple gNB-CU-CPs by appropriate implementation. One gNB-DU can be connected to multiple gNB-CU-UPs under the control of the same gNB-CU-CP. One gNB-CU-UP can be connected to multiple DUs under the control of the same gNB-CU- CP.
[0025] NOTE 2: The connectivity between a gNB-CU-UP and a gNB-DU is established by the gNB-CU-CP using Bearer Context Management functions.
[0026] NOTE 3: The gNB-CU-CP selects the appropriate gNB-CU-UP(s) for the requested services for the UE. In case of multiple CU-UPs they belong to same security domain as defined in TS 33.210.
[0027] NOTE 4: Data forwarding between gNB-CU-UPs during Intra-gNB-CU-CP handover within a gNB may be supported by Xn-U. [0028] Published technology and ongoing 3GPP discussions
[0029] The Functional Framework for RAN Intelligence, as captured in 3GPP document R3-221014 (available at https://www.3gpp.org/ftp/TSG_RAN/WG3_lu/TSGR3_114bis-e/Docs/?sortby=sizerev as of 18 January 2023) is shown in Figure 3. The description of each functional block is reported below.
[0030] Data Collection is a function that provides input data to Model training and Model inference functions. AI/ML algorithm specific data preparation (e.g., data pre-processing and cleaning, formatting, and transformation) is not carried out in the Data Collection function.
Examples of input data may include measurements from UEs or different network entities, feedback from Actor, output from an AI/ML model.
[0031] Training Data: Data needed as input for the AI/ML Model Training function.
[0032] Inference Data: Data needed as input for the AI/ML Model Inference function.
[0033] Model Training is a function that performs the ML model training, validation, and testing which may generate model performance metrics as part of the model testing procedure. The Model Training function is also responsible for data preparation (e.g. data pre-processing and cleaning, formatting, and transformation) based on Training Data delivered by a Data Collection function, if required.
[0034] Model Deployment/Update: Used to initially deploy a trained, validated, and tested AI/ML model to the Model Inference function or to deliver an updated model to the Model Inference function.
[0035] Note: Details of the Model Deployment/Update process as well as the use case specific AI/ML models transferred via this process are out of RAN3 Rel-17 study scope. The feasibility to single-vendor or multi-vendor environment has not been studied in RAN3 Rel-17 study.
[0036] Model Inference is a function that provides AI/ML model inference output (e.g. predictions or decisions). It is FFS (for future study) whether it provides model performance feedback to Model Training function. The Model inference function is also responsible for data preparation (e.g. data pre-processing and cleaning, formatting, and transformation) based on Inference Data delivered by a Data Collection function, if required.
[0037] Output: The inference output of the AI/ML model produced by a Model Inference function.
[0038] Note: Details of inference output are use case specific.
[0039] (FFS) Model Performance Feedback: Applied if certain information derived from Model Inference function is suitable for improvement of the AI/ML model trained in Model Training function. Feedback from Actor or other network entities (via Data Collection function) may be needed at Model Inference function to create Model Performance Feedback.
[0040] Note: Details of the Model Performance Feedback process are out of RAN3 Rel-17 study scope.
[0041] Actor is a function that receives the output from the Model inference function and triggers or performs corresponding actions. The Actor may trigger actions directed to other entities or to itself.
[0042] Feedback: Information that may be needed to derive training or inference data or performance feedback.
[0043] In the same document, the following high-level principles are described: [0044] The Model Training and Model Inference functions may be able to request, if needed, specific information to be used to train or execute the AI/ML algorithm and to avoid reception of unnecessary information. The nature of such information depends on the use case and on the AI/ML algorithm.
[0045] The Model Inference function may signal the outputs of the model only to nodes that have explicitly requested them (e.g. via subscription), or nodes that are subject to actions based on the output from Model Inference.
[0046] One of the main principles of the currently discussed Functional Framework for RAN Intelligence is the use of a subscription paradigm, so that information is transferred from a second functional block (e.g. Model Training or Model Inference) to a first functional block once a subscription request from the first functional block has been accepted by the second functional block.
[0047] Handling of datasets used within the AI/ML framework
[0048] Recent development in AI/ML algorithm shows the benefit of using generative models to learn dataset distribution. Several techniques for learning to generate dataset are considered state-of-the-art in this direction, such as variational auto-encoder and generative adversarial network.
[0049] On the other hand, reinforcement learning has been investigated for a long time. The most important factor for the operation of the RL is to have an interactive environment where the agent places its action and observes a vector of environment states and rewards. It is also possible, to replace the interactive environment with a simulator that receives action and respond with states and reward. However, if there is an existing dataset without a simulator then it becomes challenging to apply RL to such dataset.
[0050] Here components that are related to the later section in this disclosure are presented. Different methods are provided that can be used at different part of the invention.
[0051] Performance evaluation of the Generative adversarial network (GAN).
[0052] The following are techniques to evaluate the GAN: KL Divergence; Average Log-likelihood; Coverage Metric; Inception Score (IS); Modified Inception Score (m-IS); Mode Score; AM Score; Frechet Inception Distance (FID); Maximum Mean Discrepancy (MMD); The Wasserstein Critic; Birthday Paradox Test; Classifier Two-sample Tests (C2ST); Classification Performance; Boundary Distortion; Number of Statistically-Different Bins (NDB); Image Retrieval Performance; Tournament Win Rate and Skill Rating; Normalized Relative Discriminative Score (NRDS); Adversarial Accuracy and Adversarial Divergence; Geometry Score; Reconstruction Error.
[0053] On stability of dataset:
[0054] Several techniques could be considered to check the stability of a dataset, such as usage of classifier/clusters to identify whether the data is imbalanced or not w.r.t specific parameters, as in, Anomaly Detection or Outlier Detection.
[0055] It may be determined whether the updated dataset has one or more of data drift characteristics and apply a necessary countermeasure. Some such examples are: Sample bias; Dual Drift; Feature (covariate) drift; Label drift; Concept drift.
[0056] There currently exist certain challenge(s). [0057] Problem-1 is how to enable the product unit to test new technologies or schedulers, as ML tech/scheduler needs to be safe before developing a new product. This extending test requires data, interactivity, and training and such processes might be impaired by the lack of opportune datasets.
[0058] Problem-2 occurs when having passive dataset, as the product unit may not be able to experimentally test the new technologies on the existing dataset. (As is described in relation to the methods below, enriching an existing dataset with extra (non-existing) scenario, enhances the generality of the product testing, hence make it safer before developing the end product.)
[0059] Problem-3 occurs when utilizing legacy generative models in the context of Problem-1 and Problem-2. In particular, the following limitations appear:
1 . Mode collapse & Fidelity problem;
2. Learn only the distribution (or a statistical "summary”) of the data ignoring any other type of correlations in the data like the spatial & traffic ones;
3. Learn how to segment a continuous imbalanced distribution, in order to learn to generate each segment;
4. Generate not only channel etc, but also other OSI-layer measurements.
5. It is hard to generate descriptive meta-data to enable interaction as needed with the controller. For example, descriptive meta data can be environment characteristic parameters, e.g, speed, associated g N B, requested traffic, number of UEs associated with BSs, etc).
[0060] Problem-4 The standardized data analytic and Al related protocols (e.g., NWDAF, etc) may not enable an efficient solution to the above problems, as they are more focused towards data and analytics (prediction) extraction. Specifically:
• NWDAF and the Data Collection functions are currently only able to perform analytics or predictions of data, which is collected from other NFs/OAM: o Some data might be missing (type of data, only available what is specified and from RAN only CAM input possible) o Collected data might only represent specific scenarios as these are restricted from the deployment (i.e. location, numbers of users). Data from many more scenarios which would be interesting might very rarely be collected (too little to train) o NWDAF (Release 17) cannot perform any action itself, only give predictions and provide recommendations.
• RANDAF currently discussed, also with opportunity to perform actions and access to all RAN data: o Still, data might be missing o For Reinforcement Learning for the training one would need to change e.g. configuration and collect the rewards, which is a risk and challenge for a live network. Hence a problem is how to make data available that would represent a scenario where some parameter has changed, where such scenario may be too risky to run in a live network as it might cause severe disruptions and downgraded performance • It is difficult to extract information from DAFs and data collection functions by using descriptive meta data. [0061] It is an object of the present disclosure to provide methods, controller modules, proxy modules, network nodes and UEs for facilitating Artificial Intelligence (AI)ZMachine Learning (ML) operations though the production and/or use of metadata.
[0062] An embodiment of the disclosure provides a method performed by a controller module for processing data. The method comprises sending a request for metadata relating to a network to a proxy module, and responsive to receiving from the proxy module metadata generated by a generative model, using the received metadata as input to a trained model. The method further comprises determining an action to take based on the output of the trained model.
[0063] A further embodiment of the disclosure provides a method performed by a proxy module for generating data. The method comprises receiving, from a controller module, a request for metadata relating to a network. The method further comprises generating metadata using a generative model, and sending the generated metadata to the controller module.
[0064] A further embodiment of the disclosure provides a method performed by a controller module for training a model. The method comprises sending a request for metadata relating to a network to a proxy module. The method further comprises, responsive to receiving metadata from the proxy module generated by a generative model, using the metadata as training data to train a model.
[0065] A further embodiment of the disclosure provides a method performed by a proxy module for training a generative model. The method comprises receiving a dataset and receiving metadata. The method further comprises using the dataset and metadata to train a generative model to generate metadata based on an input dataset.
[0066] Further embodiments of the disclosure provide controller modules, proxy modules, network nodes and UEs configured to execute some or all of the methods disclosed herein.
[0067] Certain aspects of the disclosure and their embodiments may provide solutions to these or other challenges.
[0068] Herein, two main aspects are discussed relating to 1) exchanging conditional information related to training and generation of datasets, 2) implementation aspects of changing data set from passive to interactive one.
[0069] This disclosure focusses on AI/ML operations facilitated by the production of metadata, namely data that can be used as inputs to a Model Inference Functions or as training data to a Model Training function, which are not derived from real events and scenarios, but that are deduced by dataset generation processes.
[0070] In one method described herein, such dataset generation may be carried out by a proxy module. Such proxy module may be a separate function or it may be part of the existing functions identified in the functional framework that RAN3 has defined.
[0071] This disclosure focusses on two parts: the standardization impact of realizing conditional data generator, forecaster, controller, in term of the factors triggering generation of such metadata, and the methods to exchange the metadata produced by such functions and in terms of utilization of such data by the functions receiving it.
The methods forming this part of the invention can be summarized as:
[0072] Identification of the events that trigger the generation of metadata. Such events may be, for example, the identification of a missing set of training data range from the data used to train a model. The latter may cause that the finally trained model would not be able to correctly infer events triggered by inputs in the region of missing training data. For this reason it would be beneficial to produce a set of metadata filling the gap in the training data used to train the model.
[0073] Usage of data: The node or function receiving such metadata may use it in a way similar to e.g. data derived from field operations for the purpose of e.g. training. However, the node receiving such metadata may also use them differently, for example by giving to such data a lower weight when compared to e.g. data derived from field operations. In the example case of model training, a Model Training function may use metadata to train a model over a wider range of training data.
[0074] Signaling of data: When a node or function needs data such as training data, inference data or feedback data (feedback to the model performance or to the system performance, for example), the node or function will request functions such as the Data Collection function or the Actor for such data. The functions providing such data may signal, together with the data, an indication that the data signaled are metadata generated by a dataset generator, in order to differentiate such data from, e.g. data derived from field operations. Such signaling can be enriched with details such as the methods used to derive such metadata, the accuracy of such metadata and more.
[0075] Implementation aspects of conditional generation and controllers in interactive closed loops.
[0076] Note:
[0077] - Conditional-based AI/ML operation, is an AI/ML operation that uses specific radio or UE-capability, or environment, or contextual scenarios during its training, inference or data generation/measurement operation.
[0078] - AI/ML Information, could be input data to AI/ML model, the AI/ML model itself, training data, Model Performance feedback data, information on System performance impacted by the Al operations.
[0079] Metadata: a set of data generated by a dataset generator and needed to optimize different aspects of an Al process, such as training. These data may be distinguished from other AI/ML information derived from e.g. field measurements.
[0080] In order to realize a conditional-based AI/ML operation, including impacted operations in data collection (realistic or virtual), measurement forecaster, and control blocks, modifications to the AI/ML framework proposed in 3GPP are introduced. These modification impact data collection, model training functions and/or model inference functions. The changes may also imply the addition of an independent function.
[0081] The newly added functionality, embedded in existing functions or provided as an independent function, enables the 3GPP framework to request, send, and/or receive different Al -I nformation (e.g., data and models) with distinct meta(conditional)-data operation.
[0082] This disclosure focusses on two parts: 1. The standardization impact of realizing conditional data generator, forecaster, controller, in term of exchanging: a) Specific parameters of such conditional Al -I nformation . b) Range of such conditional Al-I nformation c) Difference between current conditional Al-lnformation and previous ones.
2. The factors triggering such data generation
3. The behaviour of a node receiving such data
4. Implementation aspects of conditional generation and controllers in interactive closed loops. a. The new framework generates data with specific characteristics as being requested from another entity (e.g., Data Collection function, Model Training Function, Model Inference function, controller entity, or forecasting entity, etc) during the learning/training or inference procedure. A mechanism is also proposed to validate the conditionally generated data for the requester entity. Those conditionally generated data (e.g., environment states, rewards, or observations) enable the controller or Model Training/Model Inference function for efficient learning/training on how to take proactive actions. Attaching 'proactive' prefix to "action1' means that the need to take an action is predicted ahead of time. Such pre-knowledge result in improving network performance, for instance, in UE mobility handover scenario, knowing when the UE should be handed over to a target cell, or knowing its trajectory, enables the network to pre-trigger the HO procedure, hence reducing HO process time to 0 msec. b. A flowchart is also proposed explaining the interaction between the proposed modules and 3GPP specifications, e.g., data analytic functions (DAF).
[0083] Certain embodiments may provide one or more of the following technical advantage(s).
[0084] From 3GPP perspective, the proposed framework may have the advantage of enabling the exchange of conditionally assisted training and data generation to enable a more efficient control operation (or Reinforcement Learning).
[0085] From Deployment perspective, the proposed framework may enable the following: o Proactive studies and evaluations. For example, what would be the UE throughput if the gNB proactively executed a handover. o Safe-training mechanism that may improve the changes of having a less-risky controller in the deployment phase. o Enriching the datasets enabling more studies of what the initial dataset could potentially support. o Enhancing simulation capabilities. The end-user is empowered by introducing characteristics to the dataset. o Future states of observation x or y can be easily predicted.
Brief Description of the Drawings [0086] For a better understanding of the embodiments of the present disclosure, and to show how it may be put into effect, reference will now be made, by way of example only, to the accompanying drawings, in which:
[0087] Figure 1 is a block diagram illustrating NG-RAN Overall Architecture;
[0088] Figure 2 illustrates the overall architecture for separation of gNB-CU-CP and gNB-CU-UP
[0089] Figure 3 illustrates the Functional Framework for RAN Intelligence
[0090] Figure 4 illustrates a method in a controller module according to an example;
[0091] Figure 5 illustrates a method in a proxy module according to an example;
[0092] Figure 6 illustrates a method in a controller module according to an example;
[0093] Figure 7 illustrates a method in a proxy module according to an example;
[0094] Figure 8 illustrates AI/ML connected framework/blocks in 3GPP TR 37.817;
[0095] Figure 9 illustrates a general proposed framework according to an example;
[0096] Figure 10 illustrates an example of methods applied to the 3GPP RAN3 AI/ML framework, case of training metadata;
[0097] Figure 11 illustrates an example of methods applied to the 3GPP RAN3 AI/ML framework, case of inference input metadata;
[0098] Figure 12 illustrates an example of metadata collection in the AI/ML energy saving use case;
[0099] Figure 13 illustrates a flow diagram for a part of the proposed system;
[0100] Figure 14 illustrates an adaptation of Functional Framework for RAN intelligence in 3GPP TR 37.817;
[0101] Figure 15 illustrates Updated Potential Architecture of the proposed Framework;
[0102] Figure 16. illustrates elaborative features of the mobility experiment;
[0103] Figure 17 illustrates an example of a communication system;
[0104] Figure 18 shows an example of a UE;
[0105] Figure 19 shows an example of a network node;
[0106] Figure 20 is a block diagram of an example of a host;
[0107] Figure 21 is a block diagram illustrating a virtualization environment; and
[0108] Figure 22 is a communication diagram of a host.
ADDITIONAL EXPLANATION
[0109] Some of the embodiments contemplated herein will now be described more fully with reference to the accompanying drawings. Embodiments are provided by way of example to convey the scope of the subject matter to those skilled in the art.
[0110] Figure 4 depicts a method in accordance with particular embodiments. The method W1 may be performed by a controller module. The controller module may be configured to process data. The method begins at step 402 with sending a request for metadata relating to a network to a proxy module, at step 402, responsive to receiving from the proxy module metadata generated by a generative model, using the received metadata as input to a trained model (to generate an output), and at step 403, determining an action to take based on the output of the trained model. Step 403 may be optional, where for example the controller module may instead send the output of the trained model (e.g. to a further module, node etc).
[0111] The output of the trained model may comprise at least one of: an environmental state; a reward; an observation. The method may further comprise sending the determined action to at least one of a UE, a network node. The method may further comprise receiving a dataset comprising at least one of measured and simulated data, and wherein the dataset may be additionally used as input to the trained model. The metadata may be given a different weight to a weight given to the dataset.
[0112] The method may further comprise sending at least one of: information indicating the type of metadata to be generated; information indicating the range within which the metadata is to be generated; at least one condition to be used for the generation of the metadata; information indicating at least one generative model to be used to generate the metadata; information indicating how the generative model is to be trained prior to generating the metadata; a forecasting time interval for generation of the metadata.
[0113] The method may further comprise receiving at least one of: an indication of the type of data received; for each type of data received, receiving an indication of whether the data comprises generated metadata; an indication of which part of the data comprises generated metadata.
[0114] The method may further comprise receiving at least one of: information indicating that the metadata is generated data; information indicating the type of metadata generated; information indicating the model used to generate the metadata; information indicating the conditions used to generate the metadata; information indicating how conditions were used to generate the metadata; information on the generative model; metadata values; information indicating the accuracy of the metadata.
[0115] The method may further comprise using the received information as input to the trained model.
[0116] The method may be performed by a model inference function.
[0117] Figure 5 depicts a method in accordance with particular embodiments. The method 5 may be performed by a proxy module. The proxy module may be configured to generate data. The method may begin at step 502 with receiving, from a controller module, a request for metadata relating to a network, at step 504, generating metadata using a generative model, at step 506, sending the generated metadata to the controller module. The receiving of a request may be optional. For example, the proxy module may instead periodically generate metadata.
[0118] At least one of a condition and a dataset may be input to the generative model to generate metadata. The dataset may comprise at least one of measured and simulated data. A condition may comprise information relating to a contextual scenario.
[0119] The method may further comprise sending at least one of: information indicating that the metadata is conditionally generated; information indicating the type of metadata generated; information indicating the model used to generate the metadata; information indicating the conditions used to generate the metadata; information indicating how conditions used to generate the metadata were used; information on the generative model; metadata values; information indicating the accuracy of the metadata.
[0120] The method may further comprise receiving at least one of: information indicating the type of metadata to be generated; information indicating the range within which the metadata is to be generated; at least one condition to be used for the generation of the metadata; information indicating at least one generative model to be used to generate the metadata; information indicating how the generative model is to be trained prior to generating the metadata, and wherein the method further comprises using the received information as input to the generative model.
[0121] The method may further comprise receiving feedback on the result of an action instructed by the controller module to at least one of a UE, a network node. The generated metadata may comprise at least one of: information relating to data missing from a dataset; forecast metadata. The method may be performed by a model inference function.
[0122] There is also provided a method in a system, the system comprising a controller module and a proxy module configured to perform the methods described herein.
[0123] The method may further comprise receiving at the proxy module a dataset. The method may further comprise updating the dataset based on the metadata output from the generative model. The system may further comprise a validation module, and the method further comprises processing the metadata by the validation module before updating the dataset.
[0124] The validation module may update the dataset if the metadata fulfils an event criteria. An event criteria may comprise at least one of: a rare event criteria; a quality wise criteria; extra desired data; metadata is derived using a new experiment. An experiment may comprise at least one of: altering network configuration; adding or removing base stations; increasing or reducing coverage; altering radio environment; introducing coverage holes; changing propagation characteristics; adding or removing frequency layers; new channel ranks; user dynamics; more throttling at the cells; dynamic/constant type of traffic; measurement campaign that contains new margins of KPIs; shorter or longer margin of latency; shorter or longer margin of throughput; shorter or longer margin of reliability; shorter or longer margin of SINR; shorter or longer margin of bandwidth; shorter or longer margin of center frequency.
[0125] Figure 6 depicts a method in accordance with particular embodiments. The method 6 may be performed by a controller module. The controller module may be configured to train a model. The method may begin at step 602 with sending a request for metadata relating to a network to a proxy module, then at step 604, responsive to receiving metadata from the proxy module generated by a generative model, using the metadata as training data to train a model.
[0126] The model may be trained to output information usable to determine an action to take based on input data. The training of the model may be further based on data comprising at least one of measured and simulated data. The steps of the method may be repeated to retrain the model in response to receiving new metadata. The trained model may be used in the method according to the embodiments described above, and wherein feedback on the performance of the model is used to retrain the model. The model may be initially trained using a measured or simulated dataset.
[0127] The method may further comprise sending at least one of: a request for a type of training data; an indication of the range of values within which each training data type is required; an indication that the training data can further comprise generated metadata; information indicating the conditions for metadata generation; information related to the generative model to be used to generate the metadata.
[0128] The method may further comprise receiving at least one of: an indication of the type of data received; for each type of data received, receiving an indication of whether the data comprises generated metadata; an indication of which part of the data comprises generated metadata; an indication of the accuracy of the metadata; information on the generative model used to generate the metadata; and wherein the method further comprises using the received information to train the model. The method may further comprise sending information on the trained model. The method may be performed by a model training function. The controller module may be comprised in an Operation, Administration, and Management entity, OAM.
[0129] Figure 7 depicts a method in accordance with particular embodiments. The method 7 may be performed by a proxy module. The proxy module may be configured to train a generative model. The method may begin at step 702 with receiving a dataset, at step 704, receiving metadata, at step 706, using the dataset and metadata to train a generative model to generate metadata based on an input dataset.
[0130] The received metadata may be at least one of: received explicitly; received implicitly; architectural based metadata. The training may be further performed based on a condition comprising information relating to a contextual scenario. The method may further comprise receiving characteristics of an experiment and using the characteristics to train the generative model. The method may further comprise receiving an action and using the action to train the generative model. The steps of the method may be repeated to retrain the generative model in response to receiving at least one of new metadata, a new dataset. The metadata may comprise at least one of: an observation; a state; a reward. The generative model may comprise at least one of: a general adversarial network, GAN; a generative minimization network; a variational auto-encoder; a conditional GAN; an InfoGAN; a Wasserstein GAN; a neural network; a deep neural network. The method may be implemented in a RAN framework.
[0131] The methods may relate to any of: handover; energy consumption; traffic characteristics measurement; core network measurements; network load measurements; network performance measurements; slice related measurements; UE related analytics; UE congestions related measurements; QoS sustainability measurements.
[0132] There is provided a method in a system, the system comprising a controller module and a proxy module, the system to perform the method according to any of the methods described herein.
[0133] There may be further provided a network node, wherein the network node comprises at least one of a controller module and a proxy module, and wherein the network node is configured to perform any of the methods described herein. The methods 4-7 may be performed by a network node (e.g. the network node 1710 or network node 1900 as described later with reference to Figures 17 and 19 respectively).
[0134] There may be further provided a UE, wherein the UE comprises at least one of a controller module and a proxy module, and wherein the UE is configured to perform any of the methods described herein. The method 4-7 may be performed by a UE or wireless device (e.g. the UE 1712 or UE 1800 as described later with reference to Figures 17 and 18 respectively).
[0135] Considered Technical Use-case for Elaboration purpose
[0136] Optimal handover (HO) decisions use-case is considered to elaborate the efficiency of the use-case described herein. This example is included for reasons of clarity. The person skilled in the art can understand that the methods are applicable to any other AI/ML supported use case.
[0137] In the example below the term "proxy module” is used to identify the functionality that generates metadata. However, such functionalities may be part of one or more of the existing functions in the AI/ML framework or it could be a separate function interacting with existing functionalities.
[0138] The proxy module may generate conditional data (related to HO decisions). Upon request from the Model Inference function, the proxy module may signal back such conditional data (or metadata) to the Model Inference function. The Model Inference function, also identified herein as "controller module”, may then use such conditionally generated input, to make proactive decision on when the handover shall take place, so the UE has a smaller drop in the overall received QoS.
[0139] Other Technical Use-case (not considered in detailed elaboration)
[0140] Assess the possibility of implementing RL and interacting with the proxy. o Real Deployment scenario o HO decision: commanding the proxy to move the UE to another gNB and generate the associated data. o MCS selection: commanding the proxy to reselect a different MCS index for the UE and generate the associated data. o Extreme Scenarios Training o Add/remove BSs o Add/remove UEs o Many potential scenarios for UE mobilities o Performance evaluation - Latency spikes. Commanding the proxy to generate more handover conditions where latency spikes occur.
[0141] Main Embodiment
[0142] Modification to functionalities in 3GPP framework in TR 37.817 (available at https://portal.3gpp.org/desktopmodules/Specifications/SpecificationDetails.aspx?specificationld=3817 as of 18 January 2023) as shown in Error! Reference source not found.8, are proposed to enable conditional generation of virtual measurement or data, or proactive control decision (in short, this is what is called in later sections converting a passive dataset to an inter-active dataset). Two aspects are focussed on: 1) Enablement of the exchange of conditional meta-parameter and its associated Al-I nformation, and 2) the implementation aspect of such interactive framework. The proposed framework composed of three main parts (as illustrated in Error! Reference source not found.9):
[0143] Dataset Sources:
[0144] This block represents the data collection module (as also called in 3GPP TR 37.817), however in this invention, many entities that could exist in the dataset source block are considered, e.g., such entities could be categorized into:
1. Data generating entity (that can be probed to send or receive measurement data), which can be represented as: a) Actual network nodes (legacy system) b) AI/ML entity (located at specific node) that uses transmitted conditions to generate the requested measurement for a specific scenario. This entity is what is referred to herein as a “Proxy Module”, which is also explained below.
2. Realistic datasets entity, responsible for collecting the measurement obtained via requests from a radio node or a conducting measurement campaign or some data collection procedure. Once the framework executes, the dataset may keep updating and being enriched.
[0145] Proxy module (PM)
[0146] This proposed entity may be placed as part of the data collection block or at the model training or inferences host as in Error! Reference source not found.8.
[0147] The proxy module comprises a generative AI/ML model that receives information from different entities across the AI/ML framework and send back Al-I nformation to different entities.
[0148] The purpose of the generative model is to produce additional (virtual) data based on given conditions and descriptive metadata.
[0149] The generated (virtual) data is input to the controller module (model training or inference host). A possible implementation of such generative model is training a generative adversial network, although this disclosure does not exclude all other generative methods and models.
[0150] The following information could be send-out of the PM module: o The requested forecasted measurement, namely the metadata generated by the PM, constituting data that are missing at the Controller Module, and that are useful to optimize the AI/ML process o Information describing that the data signaled are metadata, namely they are conditionally generated data and not, e.g. data derived from field tests. Such information may also include the type of metadata signaled, e.g. type of UE measurement, type of RAN measurement, type of system performance measurement. o Al-information, describing the models which used to generate such meta-data or conditions. o Assistant information (represented as the conditions which helped the PM to generate the forecasted measurement or requested Al-information), such information could be represented as: o Contextual information considered to generate the metadata provided:
■ Radio environment context information
■ UE radio capability related context information o Details on how each of those contextual/conditional information was used in training the generative model. o Given a transmitted Al-I nformation (model or Al-Data as output), PM may send the difference between this version of Al -I nformation and previous one that was sent without the operation/involvement of PM.
[0151] The following information could be received by the PM module: o Request from inference or training host (or real data-source) to generate specific (virtual) measurements/metadata o The request may include the type of metadata required, e.g. cell resource utilization, UE RSRP cell measurements, etc. o The request may include the range within which the requested metadata shall be generated, e.g. for UE RSRP measurements, the range could be -110dB and -120dB o Specific conditions to be used for the generation of metadata, e.g. specific radio conditions, e.g. assumed pathloss, or specific load conditions, e.g. PRB utilization. o Specific generative models to be used, namely specifying the model to use to generate the metadata o Corresponding training for the model to be used to generate metadata, namely details on how the model has been/could be trained before generating metadata.
[0152] Controller module:
[0153] The purpose of the controller module is to take the data and conditions produced from the proxy module and pass/suggest actions and observe states of the environment. For example, such an action can be the decision to make a handover to a new cell. A General Proposed Framework is shown in Figure 9.
[0154] Possible embodiment related to the 3GPP RAN AI/ML framework
[0155] In Figure 10 an example of how the methods herein described can be mapped to the functional framework defined by 3GPP RAN3 is shown.
[0156] In 1, the Model Training Function requests training data from the Data Collection function. Such data may comprise one or more of the following: o An indication for training data types, e.g. RSRP measurements, cell Load measurements etc. o An indication of the range of values within which each training data type is required o An indication that the training data needed may also comprise metadata o In case metadata are accepted by the Model Training Functions, the conditions for such metadata generation may be indicated, e.g. radio conditions, load conditions, transport network conditions o In case metadata are accepted by the Model Training Functions, the Model Training function may also indicate details concerning the generative model that the Data Collection function could adopt. Such details may point at a specific generative model or they may point at one or more model characteristics [0157] In 2, the Data Collection function processes the request from the Model Training Function and derives the data required.
[0158] In 3, the Data Collection function signals to the Model Training function the training data required. Such signalling may include one or more of the following: o The type of data provided to the Model Training function (e.g. RSRP measurements, cell load measurements, etc.) o For each type of data, an indication of whether the data include conditionally generated metadata or not o For each type of data, a list of data values that has not been generated in a conditional way, as well as a list of conditionally generated metadata values o For each type of metadata signalled, an optional accuracy or uncertainty score indicating with what level of error such data may compare with real non-conditionally generated data o For each type of metadata signalled, details about the generative model used to generate such metadata [0159] With the training data received, the Model Training function is able to further train an AI/ML model and to improve its predictions accuracy also due to the conditionally generated data, which may fill the gaps of non- conditionally generated training data ranges available.
[0160] In Figure 11 an example of how the methods herein described can be mapped to the functional framework defined by 3GPP RAN3 is shown.
[0161] In 1, the Model Inference Function requests model inference input data from the Data Collection function.
Such data may comprise one or more of the following: o An indication for input data types, e.g. RSRP measurements, cell Load measurements etc. o An indication of the range of values within which each input data type is required o An indication that the training data needed may also comprise metadata o In case metadata are accepted by the Model Inference Functions, the conditions for such metadata generation may be indicated, e.g. radio conditions, load conditions, transport network conditions o In case metadata are accepted by the Model Inference Functions, the Model Inference function may also indicate details concerning the generative model that the Data Collection function could adopt. Such details may point at a specific generative model or they may point at one or more model characteristics [0162] In 2, the Data Collection function processes the request form the Model Inference Function and derives the data required.
[0163] In 3, the Data Collection function signals to the Model Inference function the input data required. Such signalling may include one or more of the following: o The type of data provided to the Model Inference function (e.g. RSRP measurements, cell load measurements, etc.) o For each type of data, an indication of whether the data include conditionally generated metadata or not o For each type of data, a list of data values that has not been generated in a conditional way, as well as a list of conditionally generated metadata values o For each type of metadata signalled, an optional accuracy or uncertainty score indicating with what level of error such data may compare with real non-conditionally generated data o For each type of metadata signalled, details about the generative model used to generate such metadata [0164] With the input data received, the Model Inference function is able to run inference that would not otherwise be possible due to the rare availability of the requested metadata inputs. Such inference process may derive model outputs that may reveal system behaviours and performance in rarely occurring use cases and events. This can be used to optimise the system, for example to ensure that certain harmful events are prevented as much as possible.
[0165] Possible embodiment related to the 3GPP RAN AI/ML energy saving use case
[0166] Figure 12 shows an example of the methods described on metadata generation and usage.
[0167] In the figure above, The 0AM, hosting the Model Training function trains a model in step 5 thanks to training data previously received from the RAN.
[0168] In step 6a, the 0AM system may request to the RAN further conditionally generated training metadata, as per the description in section 6.1.4 [in TS 38.300], The RAN provides such conditionally generated training metadata in step 6b, as described in section 6.1.4 [in TS 38.300], With such metadata the 0AM is able to retrain the Al /ML Model and deploy an update of such model to the RAN (step 6c and 6d, respectively).
[0169] In step 7a and 8a the RAN is able to request to other RAN nodes and to the UE for metadata inputs. Such request follows the descriptions in section 6.1.4. in steps 7b and 8b, the RAN receives the requested metadata as described in section 6.1.4. the RAN is therefore able to sum metadata input based predictions in step 9 and eventually to use such predictions to optimize the system performance.
[0170] In step 10, the RAN (or in general the node hosting the model inference function) may include the result of the metadata generated predictions and signal it to the model training function and/or to the 0AM. The model training function and/or the 0AM may optimize the system on the basis of such generated predictions.
[0171] Illustrative Implementation embodiments
[0172] Impact of the proposed framework on the relevant 3GPP standardized entities (e.g., OAM/DAF)
[0173] In this section it is described how the framework can be supported from a 3GPP network and what type of signaling messages can enable the interaction between the three main parts of this proposed framework.
[0174] Impact of information exchange between controller and proxy module on DAF/OAM entities: 1 . CM Request virtual (or real) observation of specific characteristic (time and period , request of mixed scenarios for further learning, others) o Mixed scenarios need to be translated into criteria that NWDAF is capable of retrieving. Such scenarios can be table based, where in the entries of specific characteristics (i.e. number of connected users, utilization of Base Stations, area of deployment, burstiness, volume, RTT) have labels. Based on those labels the NWDAF can find out which are the network entities and for which parts of the network the data collection shall take place.
2. PM send observations to Control module based on the current requirements. For example, when a timer for data collection has expired or the quality metric was reached.
3. CM can request more data or extend the time window to allow more measurements in case the collected data set did not meet specific criteria.
4. CM send proactive action towards the PM (and the...) that can be used to enrich the dataset.
[0175] Impact of information exchange between proxy module and updated module (or filtering module) on DAF/OAM entities:
5. When PM decide to update the dataset, the filter/validation module receive updating sample... some interactions between the following entities occur:
1 . Proxy module
2. Filter module
3. Updated Dataset module
[0176] Error! Reference source not found.13Error! Reference source not found, illustrated a flow diagram of part of the proposed invention and how its mapped to 3GPP (especially TS-23.288), with new potential messages. As illustrated above, the impact of the proposed framework on 3GPP messaging exchange is large, here the focus is limited to a limited use-case. To be precise, the described part focuses on enabling the proxy module (called ‘NWPNFConsunT Error! Reference source not found., or simply another network function "NF”) to request data from NWDF (via Tranlatelnterface which map to with network or RAN proxy module). Error! Reference source not found.13 elaborates on the case where:
1 . Translate Interface entity map the consumer to the NWDAF (not RANDAF).
2. RL observation report or input request are not considered (however RL could be consider yet as another NF).
3. How the proxy module saves analytics is not considered.
[0177] Other alternative for the requested analytics could be (as shown in TS-23.288):
1. Traffic characteristics measurement
2. Core network measurements
3. Network load measurements
4. Network performance measurements
5. Slice related measurements
6. UE related analytics 7. UE congestions related measurements
8. QoS sustainability measurements
[0178] The steps in Error! Reference source not found.13Error! Reference source not found, are described as:
[0179] 1. NWPNFConsumer sends a request of NWDAF related measurement from Translatelnterface.
2. The Translatelnterface, send a confirmation of registration, and inform whether to register at NWDAF or RANDAF.
(3-4). NWPNFConsumer registers at NWDAF,
5. NWPNFConsumer requests NWDAFRelatedMeasurement (or one of its alternatives as described above).
If the requested analytics are available in CAM:
(6-9). If the request is authorized, and in order to provide the requested analytics, the NWDAF may need for each NF targeted instance to subscribe to CAM services to retrieve the target NF resource usage and NF resources configuration following steps captured in clause 6.2.3.2 (in TS-23.288) for data collection from CAM. Steps 2-5 may be skipped when e.g. the NWDAF already has the requested analytics.
Else:
(10-12). Requested Info is obtained from network proxy NF (NWPNF).
13. The NWDAF subscribes to changes on the load and status of NF instances registered in NRF and identified by their NF id from NRF using Nnrf_NFManagement_NFStatusSubscribe service operation for each NF instance.
14. NRF notifies NWDAF of changes on the load and status of the requested NWPNFConsumer instances by using Nnrf_NFManagement_NFStatusNotify service operation.
15. The NWDAF derives requested analytics.
16. The NWDAF provide requested NWPNFConsumer load analytics to the NWPNFConsumer along with the corresponding Validity Period, using either the Nnwdaf_Analyticslnfo_Request response or Nnwdaf_AnalyticsSubscription_Subscribe response, depending on the service used in step 1. (17-20). If at step 1 the NF has subscribed to receive continuous reporting of NF load analytics, the NWDAF may generate new analytics and, when relevant according to the Analytics target period and Reporting Threshold, provide them along with the corresponding Validity Period to the NF upon reception of notification of new NF load information from 0AM or NRF.
[0180] Error! Reference source not found.14 describes how the functional framework of RAN intelligence in 3GPP TR 37.817 is adapted to the proposed framework. o Both the proposed updated dataset and Proxy module blocks can be mapped to the Data-source block of the functional framework of RAN intelligence in TR 37.817. o The Controller module block can be mapped to both model training host and model inference host blocks of the functional framework of RAN intelligence in TR 37.817. o The Actor block of the functional framework of RAN intelligence in TR 37.817, may simply be the gNB or UE.
[0181] Detailed description of the proposed framework:
[0182] Dataset (original and updated). The dataset module is the collected measurement obtained via a measurement campaign or over some simulation.
[0183] Proxy Module. The proxy module comprises a generative model. The following characteristics describe the proxy module: o A generative Al model is trained using the existing dataset with conditioning and meta-data learning possibility. o The utilized Al generative model receives input from controller module to steer the latent space of generative distribution and dynamically generates different datasets outputs.
[0184] Proxy module generation process: The proxy module focuses on generating
1 . Current samples of the targeted data.
2. Forecasted samples (in time, location or any other appropriate dimension) of the targeted data. In this context, the proxy module could contain a time series generating functionality block. Such block will further consider o Aperiodic forecasted (in time or place) generation of samples, such that based on a request from (e.g.,) Controller module it generates a forecasted sample [t,t+T] for event based scenario. o Tweaking the periodicity of generating the samples and send it to the interaction module.
[0185] Implementation of Proxy module [0186] Many generative models could be used in the proxy modules, for instance:
1 . Generative Adversarial Network, which utilizes the concept of adversary between the generator and discriminator. Where the generator uses gradient descent steps toward minimizing discriminator loss, while the discriminator performs a gradient ascent step toward maximizing discriminator loss. Hence, there is no guarantee of convergence.
2. Generative Minimization network has both generator and discriminator to minimize a loss function rather than working in opposite direction in a competitive fashion, which enable better guarantee of convergence.
3. Variational auto-encoder is another technique that can be used to generate, and enrich the dataset based on the measurement campaign.
[0187] Any appropriate generative model may be used, for example, conditional GANs, InfoGANs, Wasserstein GANs, and any other type of methods that falls under the same class of machine learning algorithms, including deep neural network approaches.
[0188] Controller module
[0189] The controller module is connected to the Proxy module to: a) Receive the environment observations generated by the Proxy module. b) Send 1) proactive actions and 2) metadata to the Proxy module. These can be conditional arguments towards to proxy module and do not need to be generated at the same time. c) Send requests on forecasting time intervals to the proxy module. d) Send requests for generation of specific mix of scenarios that help the reinforcement learning in the controller module to gain more experience.
[0190] Implementation of Controller module
[0191] The implementation of the controller module is similar to the proxy module and same methods can be applied here as were described above ("Implementation of Proxy module”).
[0192] Training the proxy and the controller module
[0193] In this section training of the proxy and the controller module is described. At the beginning the training of the proxy module starts until it reaches some defined quality threshold. After that point the training continues with the interaction between the proxy and the controller module.
[0194] Training of Proxy:
[0195] The proxy module is connected to: o Controller Module, o PM receives from CM: the CM's actions and other processed meta-data. o PM transmits to CM: the CM's observations, states, rewards. o Updated Dataset o PM receives from updated dataset, its training input to the discriminator o PM sends to updated dataset, its generated output, via validation module o Validation Module o PM sends to updated dataset, its generated output, via validation module
[0196] In case of training of the proxy module, the input, output of PM is discussed below. The input of the training is: o updated dataset (or the dataset at its first phase), o meta-data (that refer to conditions and also describe dataset characteristics), section 6.1.6.9 illustrates more on the meta-data. o Characteristics of new experiment if required. Section 6.1.6.6 elaborates more on such characteristics.
The output comprises: o All Observations that will be used in the CM training, those observationswill depend on the addressed use-case o Potential reward that will be part of input towards the CM training. More specifically at the initialization phase the reward is one of the actions of the actual dataset and in the inference phase is the actions of the training phase of the CM. o In initialization phase, the reward will be the one on the actions of the actual dataset o In inference phase, the reward will be the one on the actions of the training phase of the CM.
[0197] Training of Controller:
[0198] The training of controller module is briefly explained here. The training starts once the proxy module has reached some specific criterion. The input is the observations as these were generated from the proxy module. For example, these can be CSI, RSRP, GPS measurements.
[0199] Since one possible implementation are Reinforcement learning type of techniques the reward can be the target that is being optimized for the specific use-case. Such target can be for example the maximization of the throughput.
[0200] The output of the controller module are the actions that it takes. In one example that is changing the association of the UE with another gNB.
[0201] Performance evaluation of the Proxy Module.
[0202] In this section, one-or-many techniques (as described in above) may be used to evaluate the proposed GAN.
[0203] Potential architectures:
(First baseline architecture is described in Error! Reference source not found.Q)
[0204] Error! Reference source not found.9 illustrates a basic architecture where the main three components (dataset, proxy module, and controller module) are connected to one another as discussed in a previous section, to train of the proxy and controller module, to provide virtual observation from proxy, and provide proactive action from the controller module. Such basic framework also considers the possibility of updating the dataset via feedback loop from the proxy module to the newly block "Updated Real-Dataset”. This is beneficial since the controller might request from proxy some new mix of dataset scenarios to enrich its RL agent learning and experience.
[0205] Second architecture is an extension of the first one, and it described in Error! Reference source not found.15.
[0206] In Error! Reference source not found.15, potential extensions to the initial framework are presented that offer more control of the different processes. The three extensions are shown with a,b, and c. a) The first extension of the basic framework is the possibility of adding a control and validation mechanism to control how the generated samples from the proxy module can be used to enrich the updated Real- Dataset. Such a control mechanism is described further in section 6.1.6.4. b) The second extension of the framework is to have the possibility of using a new dataset (e.g., from real measurement campaign) to enrich the experience of the proxy module. c) The third extension is the possibility/functionality to request a new measurement campaign with specific scenarios (details are in section 6.1.6.5) to enrich the experience of proxy and controller modules.
[0207] Control/Validation mechanism block
[0208] In this section, specific mechanisms are proposed to enhance the existing/updated dataset with new data which contain a mix of scenarios using the proxy module. Such mechanisms can be summarized as follows: o Rare event criteria to enrich existing dataset. If the samples of proxy module pass the rare event criterion, it can be added to the updated dataset. A rare event can be defined, for example, as: o Below a certain frequency of occurrence. o Which the associated results in extreme targeted KPI (latency, reliability, huge number of nodes, etc). o Quality wise criteria to enrich the existing dataset. This is summarized as: o reconstruction error of an initial distribution, meeting specific dataset characteristics (i.e. the dataset is not imbalanced) o The quality of the samples can be tested via communication theory baseline, e.g.,
■ If the sample SINR maps to the known SINR formula with a x% margin of error.
■ If the sample throughput maps to the known throughput formula with a x% margin of error.
■ If the sample link budge maps to the known link budget formula with a x% margin of error. o Extra desired descriptive data (not existing) which is constructed via combining/mixing of meta-data. For instance: o Experiment-1 uses 2 cars and 3 interferers (exist in original measurement campaign) o Experiment-1 uses 6 cars and 8 interferers (exist in original measurement campaign) o Experiment-1 uses 4 cars and 4 interferes (Virtual mixing, does not exist in the original measurement campaign) o Inclusion of other experiment to enrich the update o Missing meta-data can decide what are the experiments needed
■ Factory, City, Indoor o This is needed when both the original and updated dataset are not enough for generality of training RL.
[0209] Section [0212] further describes the characteristics of new experiments.
[0210] On stability of updated dataset
[0211] In order to check the stability of the updated dataset, one or more of the techniques described above may be utilized, such as in section 2.1.3.3.
[0212] Potential characteristics and examples of new experiments.
[0213] In this section, characteristics of the new experiment which may be used to enrich the existing/updated dataset are identified. A summary example of such characteristics could be: o Network configuration o Add or remove of BSs o Increasing or reducing coverage o Radio environment o Introducing of coverage holes o Change of propagation characteristics o Adding or removing frequency layers. o New channel ranks o User's dynamics o More throttling at the cells o Dynamic/constant type of traffic o Measurements campaign that contains new margins of KPIs o Shorter & longer margin of latency, throughput, reliability, SINR, bandwidth, center frequency
[0214] Examples of the updated experiment
[0215] In this section, an example is provided based on real collected dataset from a large test network. Column names are provided as the input features, as these were collected from different network nodes. This section illustrates an exemplary use-case for a mobility /handover scenario.
[0216] Current columns in the dataframe (1sec granularity): o UE/Device specific measurements/readings
■ 'devicejd', 'measjd', 'serving_cell_id', 'speed_kmh' o Radio Specific measurement ■ 'serving_cell_rsrp_x_mean_maj', 'serving_cell_rsrq_x_mean_maj',
'serving_cell_rssi_x_mean_maj', 'serving_cell_snr_x_mean_maj',
'serving_cell_rsrp_1_med_maj'
■ 'rankjndex', 'cqi_O', 'cqi_1', o Environment
■ 'apparentjemperature', 'area', 'cloud_cover', 'cog', 'dew_point', 'humidity', 'lat', 'Ion', 'postal_code', 'precipjntensity', 'precip_probability', 'pressure', , 'street_name', 'temperature',
'traffic_distance', 'trafficjamjactor', 'uvjndex', 'visibility', 'wind_speed', o Serving and Neighboring cells characteristics
■ 'lat_serving_ceH', 'lon_serving_cell','azimuth_serving_cell', 'distance_serving_cell', 'maj_cell_frac','serving_cell_rsrp_1_med_maj', 'serving_cell_rsrp_1_std_maj', 'serving_cell_rsrp_2_med_maj',
'serving_cell_rsrp_2_std_maj',
'serving_cell_rsrq_1_med_maj', 'serving_cell_rsrq_1_std_maj',
'serving_cell_rsrq_2_med_maj', 'serving_cell_rsrq_2_std_maj',
'serving_cell_rssi_1_med_maj', 'serving_cell_rssi_1_std_maj',
'serving_cell_rssi_2_med_maj', 'serving_cell_rssi_2_std_maj',
'serving_cell_snr_1_med_maj', 'serving_cell_snr_1_std_maj',
'serving_cell_snr_2_med_maj', 'serving_cell_snr_2_std_maj',
'serving_cell_id_unique', 'cell_load_UL',
'cell Joad_DL', 'UE_UL', 'UE_DL', 'neighbor_cell_id_0',
'neighbor_celi_frac_0', 'neighbor_celi_rsrp_mean_0',
'neighbor_celi_rsrp_med_0', 'neighbor_celi_rsrp_std_0',
'neighbor_celi_rsrq_mean_0', 'neighbor_celi_rsrq_med_0',
'neig h bor_cel l_rsrq_std_O', ' neig h bor_cel l_i d_1 ',
'neigh bor_cell_frac_1 ', 'neigh bor_cell_rsrp_mean_1 ',
'neigh bor_cell_rsrp_med_1 ', 'neigh bor_cell_rsrp_std_1 ',
'neigh bor_cell_rsrq_mean_1 ', 'neigh bor_cell_rsrq_med_1 ',
'neigh bor_cell_rsrq_std_1 ', 'neigh bor_cel l_id_2',
'neighbor_cell_frac_2', 'neighbor_cell_rsrp_mean_2',
'neighbor_cell_rsrp_med_2', 'neighbor_cell_rsrp_std_2',
'neighbor_cell_rsrq_mean_2', 'neighbor_cell_rsrq_med_2',
'neighbor_cell_rsrq_std_2', 'neighbor_cell_id_3',
'neighbor_cell_frac_3', 'neighbor_cell_rsrp_mean_3',
'neighbor_cell_rsrp_med_3', 'neighbor_cell_rsrp_std_3',
'neighbor_cell_rsrq_mean_3', 'neighbor_cell_rsrq_med_3',
'neighbor_cell_rsrq_std_3', 'neighbor_cell_id_4',
'neighbor_cell_frac_4', 'neighbor_cell_rsrp_mean_4',
'neighbor_cell_rsrp_med_4', 'neighbor_cell_rsrp_std_4',
'neighbor_cell_rsrq_mean_4', 'neighbor_cell_rsrq_med_4',
'neighbor_cell_rsrq_std_4', 'neigh_0_cell_load_UL',
'neigh_0_cell_load_DL', 'neigh_O_UE_UL', 'neigh_O_UE_DL',
'neigh_1_cell_load_UL', 'neigh_1_cell_load_DL', 'neigh_1_UE_UL',
'neigh_1_UE_DL', 'neigh_2_cell_load_UL', 'neigh_2_cell_load_DL',
'neigh_2_UE_UL', 'neigh_2_UE_DL', 'neigh_3_cell_load_UL',
'neigh_3_cell_load_DL', 'neigh_3_UE_UL', 'neigh_3_UE_DL',
'neigh_4_cell_load_UL', 'neigh_4_cell_load_DL', 'neigh_4_UE_UL',
'neigh_4_UE_DL', o Traffic
'throughput_DL', 'delay_mean_DL', 'delay_std_DL',
'packetloss_DL', 'throughput_UL', 'delay_mean_UL', 'delay_std_UL',
'packetloss_UL', 'env', 'target_dl', 'target_ul' o PDSCH further columns: ['ACK/NACK Decision', 'CRC Result', 'Did Recombining',
'Discarded reTx Present', 'HARQ ID', 'MCS', 'Modulation Type',
'NDI', 'Num RBs', 'RNTI Type', 'RV', 'TB Index', 'TB Size']
[0217] Figure 16 shows Elaborative features of the mobility experiment, in particular relations between different features. Proxy module could learn interdependencies and generate new samples.
[0218] Example from real-experiment related to Proxy module activity: o Nb UE: Experiment shows KPI metrics (e.g., 'TB Size'/ 'serving_cell_snr_x_mean_maj') for specific number of UE performing DL data transmission, with discrete values (like: 10, 20...). The proposed Proxy module may generate similar KPI metrics (e.g., 'TB Size'/ 'serving_cell_snr_x_mean_maj') for different number of UEs in the DL after learning the dependencies among them. o MCS: Experiment shows KPI metrics ('MCS7 'serving_cell_snr_x_mean_maj'). The proposed Proxy module may learn the relationship between both KPI metrics and also given other metadata. The proposed Proxy module will be able to generate new 'serving_cell_snr_x_mean_maj' for in the data nonexistent 'MCS' values and vice versa.
[0219] Example from real experiment related to Controller module activity: o MCS: Controller Module decides to take action in form of increasing 'MCS'. Since the resulting data is not included in the original dataset, the Proxy Module generates the new "sample” and provides the virtual observation, including states (snr, 'TB Size', 'Num RBs'....) and rewards ('throughput_DL') of the previous controller action, to the Controller Module, which can then decide on the new action (another 'MCS'index).
[0220] On Granularity of Proxy Model (PM) Output to Controller module
[0221] In this section, several options of the proxy module (PM) to generate data to be used by the controller are proposed. o PM could send data to Controller model based on: a. Controller preference, might require i. Prediction of temporal/spatial/event/etc b. Distribution of output of PM to CM
I. Uniform
II. biased o Temporal granularity (user dynamics?) a. Start/end of the requested samples b. Msec/sec/hours/days c. Frequency of samples o Spatial granularity (Propagation dimension) a. Focus of PM could be on generating data from
I. different geographical, others? ii. Different propagation environment, others? If not specified in meta data (indoor, urban, sub-urban, factory, highway) o Event granularity a. More or less handover, b. UE connection establishment events c. Retransmission events
[0222] Example meta-data/Conditions for proxy module
[0223] In this section, methods of finding meta-data that is necessary for training PM and enriching data from PM to controller and dataset are outlined. Several methods are proposed for obtaining metadata: o Explicitly meta-data, depending on the deployment and application, for instance: o Contextual information + Rank of the channel + radio distance (range, not exact) + frequencies -t-neighbor network load, interference + capacity + Aggregated traffic scenarios +Speed + New margins of KPIs + Shorter & longer margin of latency, throughput, reliability, SI NR, bandwidth, center frequency o Implicitly — >e.g., latent variable -^MAE (or AE). Such latent variable could be found via fixing specific KPI (TP, Latency, Reliability, Coverage, etc) for training/updating the latent variable encoder. o Architectural based metadata, which could be found via considering the following architecture technology: o D-MIMO + Dual Connectivity + mmWave + SISO + Side link
[0224] Figure 17 shows an example of a communication system 1700 in accordance with some embodiments.
[0225] In the example, the communication system 1700 includes a telecommunication network 1702 that includes an access network 1704, such as a radio access network (RAN), and a core network 1706, which includes one or more core network nodes 1708. The access network 1704 includes one or more access network nodes, such as network nodes 1710a and 1710b (one or more of which may be generally referred to as network nodes 1710), or any other similar 3rd Generation Partnership Project (3GPP) access node or non-3GPP access point. The network nodes 1710 facilitate direct or indirect connection of user equipment (UE), such as by connecting UEs 1712a, 1712b, 1712c, and 1712d (one or more of which may be generally referred to as UEs 1712) to the core network 1706 over one or more wireless connections.
[0226] Example wireless communications over a wireless connection include transmitting and/or receiving wireless signals using electromagnetic waves, radio waves, infrared waves, and/or other types of signals suitable for conveying information without the use of wires, cables, or other material conductors. Moreover, in different embodiments, the communication system 1700 may include any number of wired or wireless networks, network nodes, UEs, and/or any other components or systems that may facilitate or participate in the communication of data and/or signals whether via wired or wireless connections. The communication system 1700 may include and/or interface with any type of communication, telecommunication, data, cellular, radio network, and/or other similar type of system. [0227] The UEs 1712 may be any of a wide variety of communication devices, including wireless devices arranged, configured, and/or operable to communicate wirelessly with the network nodes 1710 and other communication devices. Similarly, the network nodes 1710 are arranged, capable, configured, and/or operable to communicate directly or indirectly with the UEs 1712 and/or with other network nodes or equipment in the telecommunication network 1702 to enable and/or provide network access, such as wireless network access, and/or to perform other functions, such as administration in the telecommunication network 1702.
[0228] In the depicted example, the core network 1706 connects the network nodes 1710 to one or more hosts, such as host 1716. These connections may be direct or indirect via one or more intermediary networks or devices. In other examples, network nodes may be directly coupled to hosts. The core network 1706 includes one more core network nodes (e.g., core network node 1708) that are structured with hardware and software components. Features of these components may be substantially similar to those described with respect to the UEs, network nodes, and/or hosts, such that the descriptions thereof are generally applicable to the corresponding components of the core network node 1708. Example core network nodes include functions of one or more of a Mobile Switching Center (MSC), Mobility Management Entity (MME), Home Subscriber Server (HSS), Access and Mobility Management Function (AMF), Session Management Function (SMF), Authentication Server Function (AUSF), Subscription Identifier De-concealing function (SIDF), Unified Data Management (UDM), Security Edge Protection Proxy (SEPP), Network Exposure Function (NEF), and/or a User Plane Function (UPF).
[0229] The host 1716 may be under the ownership or control of a service provider other than an operator or provider of the access network 1704 and/or the telecommunication network 1702, and may be operated by the service provider or on behalf of the service provider. The host 1716 may host a variety of applications to provide one or more services. Examples of such applications include the provision of live and/or pre-recorded audio/video content, data collection services, for example, retrieving and compiling data on various ambient conditions detected by a plurality of UEs, analytics functionality, social media, functions for controlling or otherwise interacting with remote devices, functions for an alarm and surveillance center, or any other such function performed by a server.
[0230] As a whole, the communication system 1700 of Figure 17 enables connectivity between the UEs, network nodes, and hosts. In that sense, the communication system may be configured to operate according to predefined rules or procedures, such as specific standards that include, but are not limited to: Global System for Mobile Communications (GSM); Universal Mobile Telecommunications System (UMTS); Long Term Evolution (LTE), and/or other suitable 2G, 3G, 4G, 5G standards, or any applicable future generation standard (e.g., 6G); wireless local area network (WLAN) standards, such as the Institute of Electrical and Electronics Engineers (IEEE) 802.11 standards (WiFi); and/or any other appropriate wireless communication standard, such as the Worldwide Interoperability for Microwave Access (WiMax), Bluetooth, Z-Wave, Near Field Communication (NFC) ZigBee, LiFi, and/or any low-power wide-area network (LPWAN) standards such as LoRa and Sigfox. [0231] In some examples, the telecommunication network 1702 is a cellular network that implements 3GPP standardized features. Accordingly, the telecommunications network 1702 may support network slicing to provide different logical networks to different devices that are connected to the telecommunication network 1702. For example, the telecommunications network 1702 may provide Ultra Reliable Low Latency Communication (URLLC) services to some UEs, while providing Enhanced Mobile Broadband (eMBB) services to other UEs, and/or Massive Machine Type Communication (mMTC)ZMassive loT services to yet further UEs.
[0232] In some examples, the UEs 1712 are configured to transmit and/or receive information without direct human interaction. For instance, a UE may be designed to transmit information to the access network 1704 on a predetermined schedule, when triggered by an internal or external event, or in response to requests from the access network 1704. Additionally, a UE may be configured for operating in single- or multi-RAT or multistandard mode. For example, a UE may operate with any one or combination of Wi-Fi, NR (New Radio) and LTE, i.e. being configured for multi-radio dual connectivity (MR-DC), such as E-UTRAN (Evolved-UMTS Terrestrial Radio Access Network) New Radio - Dual Connectivity (EN-DC).
[0233] In the example illustrated in Figure 17, the hub 1714 communicates with the access network 1704 to facilitate indirect communication between one or more UEs (e.g., UE 1712c and/or 1712d) and network nodes (e.g., network node 1710b). In some examples, the hub 1714 may be a controller, router, a content source and analytics node, or any of the other communication devices described herein regarding UEs. For example, the hub 1714 may be a broadband router enabling access to the core network 1706 for the UEs. As another example, the hub 1714 may be a controller that sends commands or instructions to one or more actuators in the UEs. Commands or instructions may be received from the UEs, network nodes 1710, or by executable code, script, process, or other instructions in the hub 1714. As another example, the hub 1714 may be a data collector that acts as temporary storage for UE data and, in some embodiments, may perform analysis or other processing of the data. As another example, the hub 1714 may be a content source. For example, for a UE that is a VR headset, display, loudspeaker or other media delivery device, the hub 1714 may retrieve VR assets, video, audio, or other media or data related to sensory information via a network node, which the hub 1714 then provides to the UE either directly, after performing local processing, and/or after adding additional local content. In still another example, the hub 1714 acts as a proxy server or orchestrator for the UEs, in particular in if one or more of the UEs are low energy loT devices.
[0234] The hub 1714 may have a constant/persistent or intermittent connection to the network node 1710b. The hub 1714 may also allow for a different communication scheme and/or schedule between the hub 1714 and UEs (e.g., UE 1712c and/or 1712d), and between the hub 1714 and the core network 1706. In other examples, the hub 1714 is connected to the core network 1706 and/or one or more UEs via a wired connection. Moreover, the hub 1714 may be configured to connect to an M2M service provider over the access network 1704 and/or to another UE over a direct connection. In some scenarios, UEs may establish a wireless connection with the network nodes 1710 while still connected via the hub 1714 via a wired or wireless connection. In some embodiments, the hub 1714 may be a dedicated hub - that is, a hub whose primary function is to route communications to/from the UEs from/to the network node 1710b. In other embodiments, the hub 1714 may be a non-dedicated hub - that is, a device which is capable of operating to route communications between the UEs and network node 1710b, but which is additionally capable of operating as a communication start and/or end point for certain data channels.
[0235] Figure 18 shows a UE 1800 in accordance with some embodiments. As used herein, a UE refers to a device capable, configured, arranged and/or operable to communicate wirelessly with network nodes and/or other UEs. Examples of a UE include, but are not limited to, a smart phone, mobile phone, cell phone, voice over IP (VoIP) phone, wireless local loop phone, desktop computer, personal digital assistant (PDA), wireless camera, gaming console or device, music storage device, playback appliance, wearable terminal device, wireless endpoint, mobile station, tablet, laptop, laptop-embedded equipment (LEE), laptop-mounted equipment (LME), smart device, wireless customer-premise equipment (CPE), vehicle-mounted or vehicle embedded/integrated wireless device, etc. Other examples include any UE identified by the 3rd Generation Partnership Project (3GPP), including a narrow band internet of things (NB-loT) UE, a machine type communication (MTC) UE, and/or an enhanced MTC (eMTC) UE.
[0236] A UE may support device-to-device (D2D) communication, for example by implementing a 3GPP standard for sidelink communication, Dedicated Short-Range Communication (DSRC), vehicle-to-vehicle (V2V), vehicle-to-infrastructure (V2I), or vehicle-to-everything (V2X). In other examples, a UE may not necessarily have a user in the sense of a human user who owns and/or operates the relevant device. Instead, a UE may represent a device that is intended for sale to, or operation by, a human user but which may not, or which may not initially, be associated with a specific human user (e.g., a smart sprinkler controller). Alternatively, a UE may represent a device that is not intended for sale to, or operation by, an end user but which may be associated with or operated for the benefit of a user (e.g., a smart power meter).
[0237] The UE 1800 includes processing circuitry 1802 that is operatively coupled via a bus 1804 to an input/output interface 1806, a power source 1808, a memory 1810, a communication interface 1812, and/or any other component, or any combination thereof. Certain UEs may utilize all or a subset of the components shown in Figure 18. The level of integration between the components may vary from one UE to another UE. Further, certain UEs may contain multiple instances of a component, such as multiple processors, memories, transceivers, transmitters, receivers, etc.
[0238] The processing circuitry 1802 is configured to process instructions and data and may be configured to implement any sequential state machine operative to execute instructions stored as machine-readable computer programs in the memory 1810. The processing circuitry 1802 may be implemented as one or more hardware-implemented state machines (e.g., in discrete logic, field-programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), etc.); programmable logic together with appropriate firmware; one or more stored computer programs, general-purpose processors, such as a microprocessor or digital signal processor (DSP), together with appropriate software; or any combination of the above. For example, the processing circuitry 1802 may include multiple central processing units (CPUs). The processing circuitry 1802 may be operable to provide, either alone or in conjunction with other UE 1800 components, such as the memory 1810, UE 1800 functionality. For example, the processing circuitry 1802 may be configured to cause the UE 1802 to perform the methods as described with reference to Figure 4 - 7.
[0239] In the example, the input/output interface 1806 may be configured to provide an interface or interfaces to an input device, output device, or one or more input and/or output devices. Examples of an output device include a speaker, a sound card, a video card, a display, a monitor, a printer, an actuator, an emitter, a smartcard, another output device, or any combination thereof. An input device may allow a user to capture information into the UE 1800. Examples of an input device include a touch-sensitive or presence-sensitive display, a camera (e.g., a digital camera, a digital video camera, a web camera, etc.), a microphone, a sensor, a mouse, a trackball, a directional pad, a trackpad, a scroll wheel, a smartcard, and the like. The presencesensitive display may include a capacitive or resistive touch sensor to sense input from a user. A sensor may be, for instance, an accelerometer, a gyroscope, a tilt sensor, a force sensor, a magnetometer, an optical sensor, a proximity sensor, a biometric sensor, etc., or any combination thereof. An output device may use the same type of interface port as an input device. For example, a Universal Serial Bus (USB) port may be used to provide an input device and an output device.
[0240] In some embodiments, the power source 1808 is structured as a battery or battery pack. Other types of power sources, such as an external power source (e.g., an electricity outlet), photovoltaic device, or power cell, may be used. The power source 1808 may further include power circuitry for delivering power from the power source 1808 itself, and/or an external power source, to the various parts of the UE 1800 via input circuitry or an interface such as an electrical power cable. Delivering power may be, for example, for charging of the power source 1808. Power circuitry may perform any formatting, converting, or other modification to the power from the power source 1808 to make the power suitable for the respective components of the UE 1800 to which power is supplied.
[0241] The memory 1810 may be or be configured to include memory such as random access memory (RAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), magnetic disks, optical disks, hard disks, removable cartridges, flash drives, and so forth. In one example, the memory 1810 includes one or more application programs 1814, such as an operating system, web browser application, a widget, gadget engine, or other application, and corresponding data 1816. The memory 1810 may store, for use by the UE 1800, any of a variety of various operating systems or combinations of operating systems.
[0242] The memory 1810 may be configured to include a number of physical drive units, such as redundant array of independent disks (RAID), flash memory, USB flash drive, external hard disk drive, thumb drive, pen drive, key drive, high-density digital versatile disc (HD-DVD) optical disc drive, internal hard disk drive, Blu-Ray optical disc drive, holographic digital data storage (HDDS) optical disc drive, external mini-dual in-line memory module (DIMM), synchronous dynamic random access memory (SDRAM), external micro-DIMM SDRAM, smartcard memory such as tamper resistant module in the form of a universal integrated circuit card (UICC) including one or more subscriber identity modules (SIMs), such as a USIM and/or ISIM, other memory, or any combination thereof. The UICC may for example be an embedded UICC (eUlCC), integrated UICC (iUICC) or a removable UICC commonly known as ‘SIM card.' The memory 1810 may allow the UE 1800 to access instructions, application programs and the like, stored on transitory or non-transitory memory media, to offload data, or to upload data. An article of manufacture, such as one utilizing a communication system may be tangibly embodied as or in the memory 1810, which may be or comprise a device-readable storage medium. [0243] The processing circuitry 1802 may be configured to communicate with an access network or other network using the communication interface 1812. The communication interface 1812 may comprise one or more communication subsystems and may include or be communicatively coupled to an antenna 1822. The communication interface 1812 may include one or more transceivers used to communicate, such as by communicating with one or more remote transceivers of another device capable of wireless communication (e.g., another UE or a network node in an access network). Each transceiver may include a transmitter 1818 and/or a receiver 1820 appropriate to provide network communications (e.g., optical, electrical, frequency allocations, and so forth). Moreover, the transmitter 1818 and receiver 1820 may be coupled to one or more antennas (e.g., antenna 1822) and may share circuit components, software or firmware, or alternatively be implemented separately.
[0244] In some embodiments, communication functions of the communication interface 1812 may include cellular communication, Wi-Fi communication, LPWAN communication, data communication, voice communication, multimedia communication, short-range communications such as Bluetooth, near-field communication, location-based communication such as the use of the global positioning system (GPS) to determine a location, another like communication function, or any combination thereof. Communications may be implemented in according to one or more communication protocols and/or standards, such as IEEE 802.11 , Code Division Multiplexing Access (CDMA), Wideband Code Division Multiple Access (WCDMA), GSM, LTE, New Radio (NR), UMTS, WiMax, Ethernet, transmission control protocol/internet protocol (TCP/IP), synchronous optical networking (SONET), Asynchronous Transfer Mode (ATM), QUIC, Hypertext Transfer Protocol (HTTP), and so forth.
[0245] Regardless of the type of sensor, a UE may provide an output of data captured by its sensors, through its communication interface 1812, via a wireless connection to a network node. Data captured by sensors of a UE can be communicated through a wireless connection to a network node via another UE. The output may be periodic (e.g., once every 15 minutes if it reports the sensed temperature), random (e.g., to even out the load from reporting from several sensors), in response to a triggering event (e.g., when moisture is detected an alert is sent), in response to a request (e.g., a user initiated request), or a continuous stream (e.g., a live video feed of a patient). [0246] As another example, a UE comprises an actuator, a motor, or a switch, related to a communication interface configured to receive wireless input from a network node via a wireless connection. In response to the received wireless input the states of the actuator, the motor, or the switch may change. For example, the UE may comprise a motor that adjusts the control surfaces or rotors of a drone in flight according to the received input or controls a robotic arm performing a medical procedure according to the received input.
[0247] A UE, when in the form of an Internet of Things (loT) device, may be a device for use in one or more application domains, these domains comprising, but not limited to, city wearable technology, extended industrial application and healthcare. Non-limiting examples of such an loT device are devices which are or which are embedded in: a connected refrigerator or freezer, a TV, a connected lighting device, an electricity meter, a robot vacuum cleaner, a voice controlled smart speaker, a home security camera, a motion detector, a thermostat, a smoke detector, a door/window sensor, a flood/moisture sensor, an electrical door lock, a connected doorbell, an air conditioning system like a heat pump, an autonomous vehicle, a surveillance system, a weather monitoring device, a vehicle parking monitoring device, an electric vehicle charging station, a smart watch, a fitness tracker, a head-mounted display for Augmented Reality (AR) or Virtual Reality (VR), a wearable for tactile augmentation or sensory enhancement, a water sprinkler, an animal- or item-tracking device, a sensor for monitoring a plant or animal, an industrial robot, an Unmanned Aerial Vehicle (UAV), and any kind of medical device, like a heart rate monitor or a remote controlled surgical robot. A UE in the form of an loT device comprises circuitry and/or software in dependence on the intended application of the loT device in addition to other components as described in relation to the UE 1800 shown in Figure 18.
[0248] As yet another specific example, in an loT scenario, a UE may represent a machine or other device that performs monitoring and/or measurements, and transmits the results of such monitoring and/or measurements to another UE and/or a network node. The UE may in this case be an M2M device, which may in a 3GPP context be referred to as an MTC device. As one particular example, the UE may implement the 3GPP NB- loT standard. In other scenarios, a UE may represent a vehicle, such as a car, a bus, a truck, a ship and an airplane, or other equipment that is capable of monitoring and/or reporting on its operational status or other functions associated with its operation.
[0249] In practice, any number of UEs may be used together with respect to a single use case. For example, a first UE might be or be integrated in a drone and provide the drone's speed information (obtained through a speed sensor) to a second UE that is a remote controller operating the drone. When the user makes changes from the remote controller, the first UE may adjust the throttle on the drone (e.g. by controlling an actuator) to increase or decrease the drone's speed. The first and/or the second UE can also include more than one of the functionalities described above. For example, a UE might comprise the sensor and the actuator, and handle communication of data for both the speed sensor and the actuators.
[0250] Figure 19 shows a network node 1900 in accordance with some embodiments. As used herein, network node refers to equipment capable, configured, arranged and/or operable to communicate directly or indirectly with a UE and/or with other network nodes or equipment, in a telecommunication network. Examples of network nodes include, but are not limited to, access points (APs) (e.g., radio access points), base stations (BSs) (e.g., radio base stations, Node Bs, evolved Node Bs (eNBs) and NR NodeBs (gNBs)).
[0251] Base stations may be categorized based on the amount of coverage they provide (or, stated differently, their transmit power level) and so, depending on the provided amount of coverage, may be referred to as femto base stations, pico base stations, micro base stations, or macro base stations. A base station may be a relay node or a relay donor node controlling a relay. A network node may also include one or more (or all) parts of a distributed radio base station such as centralized digital units and/or remote radio units (RRUs), sometimes referred to as Remote Radio Heads (RRHs). Such remote radio units may or may not be integrated with an antenna as an antenna integrated radio. Parts of a distributed radio base station may also be referred to as nodes in a distributed antenna system (DAS).
[0252] Other examples of network nodes include multiple transmission point (multi-TRP) 5G access nodes, multistandard radio (MSR) equipment such as MSR BSs, network controllers such as radio network controllers (RNCs) or base station controllers (BSCs), base transceiver stations (BTSs), transmission points, transmission nodes, multi-cell/multicast coordination entities (MCEs), Operation and Maintenance (O&M) nodes, Operations Support System (OSS) nodes, Self-Organizing Network (SON) nodes, positioning nodes (e.g., Evolved Serving Mobile Location Centers (E-SMLCs)), and/or Minimization of Drive Tests (MDTs).
[0253] The network node 1900 includes processing circuitry 1902, a memory 1904, a communication interface 1906, and a power source 1908, and/or any other component, or any combination thereof. The network node 1900 may be composed of multiple physically separate components (e.g., a NodeB component and a RNC component, or a BTS component and a BSC component, etc.), which may each have their own respective components. In certain scenarios in which the network node 1900 comprises multiple separate components (e.g., BTS and BSC components), one or more of the separate components may be shared among several network nodes. For example, a single RNC may control multiple NodeBs. In such a scenario, each unique NodeB and RNC pair, may in some instances be considered a single separate network node. In some embodiments, the network node 1900 may be configured to support multiple radio access technologies (RATs). In such embodiments, some components may be duplicated (e.g., separate memory 1904 for different RATs) and some components may be reused (e.g., a same antenna 1910 may be shared by different RATs). The network node 1900 may also include multiple sets of the various illustrated components for different wireless technologies integrated into network node 1900, for example GSM, WCDMA, LTE, NR, WiFi, Zigbee, Z-wave, LoRaWAN, Radio Frequency Identification (RFID) or Bluetooth wireless technologies. These wireless technologies may be integrated into the same or different chip or set of chips and other components within network node 1900.
[0254] The processing circuitry 1902 may comprise a combination of one or more of a microprocessor, controller, microcontroller, central processing unit, digital signal processor, application-specific integrated circuit, field programmable gate array, or any other suitable computing device, resource, or combination of hardware, software and/or encoded logic operable to provide, either alone or in conjunction with other network node 1900 components, such as the memory 1904, network node 1900 functionality. For example, the processing circuitry 1902 may be configured to cause the network node to perform the methods as described with reference to Figure 4-7.
[0255] In some embodiments, the processing circuitry 1902 includes a system on a chip (SOC). In some embodiments, the processing circuitry 1902 includes one or more of radio frequency (RF) transceiver circuitry 1912 and baseband processing circuitry 1914. In some embodiments, the radio frequency (RF) transceiver circuitry 1912 and the baseband processing circuitry 1914 may be on separate chips (or sets of chips), boards, or units, such as radio units and digital units. In alternative embodiments, part or all of RF transceiver circuitry 1912 and baseband processing circuitry 1914 may be on the same chip or set of chips, boards, or units.
[0256] The memory 1904 may comprise any form of volatile or non-volatile computer-readable memory including, without limitation, persistent storage, solid-state memory, remotely mounted memory, magnetic media, optical media, random access memory (RAM), read-only memory (ROM), mass storage media (for example, a hard disk), removable storage media (for example, a flash drive, a Compact Disk (CD) or a Digital Video Disk (DVD)), and/or any other volatile or non-volatile, non-transitory device-readable and/or computer-executable memory devices that store information, data, and/or instructions that may be used by the processing circuitry 1902. The memory 1904 may store any suitable instructions, data, or information, including a computer program, software, an application including one or more of logic, rules, code, tables, and/or other instructions capable of being executed by the processing circuitry 1902 and utilized by the network node 1900. The memory 1904 may be used to store any calculations made by the processing circuitry 1902 and/or any data received via the communication interface 1906. In some embodiments, the processing circuitry 1902 and memory 1904 is integrated.
[0257] The communication interface 1906 is used in wired or wireless communication of signaling and/or data between a network node, access network, and/or UE. As illustrated, the communication interface 1906 comprises port(s)/terminal(s) 1916 to send and receive data, for example to and from a network over a wired connection. The communication interface 1906 also includes radio front-end circuitry 1918 that may be coupled to, or in certain embodiments a part of, the antenna 1910. Radio front-end circuitry 1918 comprises filters 1920 and amplifiers 1922. The radio front-end circuitry 1918 may be connected to an antenna 1910 and processing circuitry 1902. The radio front-end circuitry may be configured to condition signals communicated between antenna 1910 and processing circuitry 1902. The radio front-end circuitry 1918 may receive digital data that is to be sent out to other network nodes or UEs via a wireless connection. The radio front-end circuitry 1918 may convert the digital data into a radio signal having the appropriate channel and bandwidth parameters using a combination of filters 1920 and/or amplifiers 1922. The radio signal may then be transmitted via the antenna 1910. Similarly, when receiving data, the antenna 1910 may collect radio signals which are then converted into digital data by the radio front-end circuitry 1918. The digital data may be passed to the processing circuitry 1902. In other embodiments, the communication interface may comprise different components and/or different combinations of components. [0258] In certain alternative embodiments, the network node 1900 does not include separate radio front-end circuitry 1918, instead, the processing circuitry 1902 includes radio front-end circuitry and is connected to the antenna 1910. Similarly, in some embodiments, all or some of the RF transceiver circuitry 1912 is part of the communication interface 1906. In still other embodiments, the communication interface 1906 includes one or more ports or terminals 1916, the radio front-end circuitry 1918, and the RF transceiver circuitry 1912, as part of a radio unit (not shown), and the communication interface 1906 communicates with the baseband processing circuitry 1914, which is part of a digital unit (not shown).
[0259] The antenna 1910 may include one or more antennas, or antenna arrays, configured to send and/or receive wireless signals. The antenna 1910 may be coupled to the radio front-end circuitry 1918 and may be any type of antenna capable of transmitting and receiving data and/or signals wirelessly. In certain embodiments, the antenna 1910 is separate from the network node 1900 and connectable to the network node 1900 through an interface or port.
[0260] The antenna 1910, communication interface 1906, and/or the processing circuitry 1902 may be configured to perform any receiving operations and/or certain obtaining operations described herein as being performed by the network node. Any information, data and/or signals may be received from a UE, another network node and/or any other network equipment. Similarly, the antenna 1910, the communication interface 1906, and/or the processing circuitry 1902 may be configured to perform any transmitting operations described herein as being performed by the network node. Any information, data and/or signals may be transmitted to a UE, another network node and/or any other network equipment.
[0261] The power source 1908 provides power to the various components of network node 1900 in a form suitable for the respective components (e.g., at a voltage and current level needed for each respective component). The power source 1908 may further comprise, or be coupled to, power management circuitry to supply the components of the network node 1900 with power for performing the functionality described herein. For example, the network node 1900 may be connectable to an external power source (e.g., the power grid, an electricity outlet) via an input circuitry or interface such as an electrical cable, whereby the external power source supplies power to power circuitry of the power source 1908. As a further example, the power source 1908 may comprise a source of power in the form of a battery or battery pack which is connected to, or integrated in, power circuitry. The battery may provide backup power could the external power source fail.
[0262] Embodiments of the network node 1900 may include additional components beyond those shown in Figure 19 for providing certain aspects of the network node's functionality, including any of the functionality described herein and/or any functionality necessary to support the subject matter described herein. For example, the network node 1900 may include user interface equipment to allow input of information into the network node 1900 and to allow output of information from the network node 1900. This may allow a user to perform diagnostic, maintenance, repair, and other administrative functions for the network node 1900.
[0263] Figure 20 is a block diagram of a host 2000, which may be an embodiment of the host 1716 of Figure 17, in accordance with various aspects described herein. As used herein, the host 2000 may be or comprise various combinations hardware and/or software, including a standalone server, a blade server, a cloud- implemented server, a distributed server, a virtual machine, container, or processing resources in a server farm. The host 2000 may provide one or more services to one or more UEs.
[0264] The host 2000 includes processing circuitry 2002 that is operatively coupled via a bus 2004 to an input/output interface 2006, a network interface 2008, a power source 2010, and a memory 2012. Other components may be included in other embodiments. Features of these components may be substantially similar to those described with respect to the devices of previous figures, such as Figures 18 and 19, such that the descriptions thereof are generally applicable to the corresponding components of host 2000.
[0265] The memory 2012 may include one or more computer programs including one or more host application programs 2014 and data 2016, which may include user data, e.g., data generated by a UE for the host 2000 or data generated by the host 2000 for a UE. Embodiments of the host 2000 may utilize only a subset or all of the components shown. The host application programs 2014 may be implemented in a container-based architecture and may provide support for video codecs (e.g., Versatile Video Coding (VVC), High Efficiency Video Coding (HEVC), Advanced Video Coding (AVC), MPEG, VP9) and audio codecs (e.g., FLAG, Advanced Audio Coding (AAC), MPEG, G.711), including transcoding for multiple different classes, types, or implementations of UEs (e.g., handsets, desktop computers, wearable display systems, heads-up display systems). The host application programs 2014 may also provide for user authentication and licensing checks and may periodically report health, routes, and content availability to a central node, such as a device in or on the edge of a core network. Accordingly, the host 2000 may select and/or indicate a different host for over- the-top services for a UE. The host application programs 2014 may support various protocols, such as the HTTP Live Streaming (HLS) protocol, Real-Time Messaging Protocol (RTMP), Real-Time Streaming Protocol (RTSP), Dynamic Adaptive Streaming over HTTP (MPEG-DASH), etc.
[0266] Figure 21 is a block diagram illustrating a virtualization environment 2100 in which functions implemented by some embodiments may be virtualized. In the present context, virtualizing means creating virtual versions of apparatuses or devices which may include virtualizing hardware platforms, storage devices and networking resources. As used herein, virtualization can be applied to any device described herein, or components thereof, and relates to an implementation in which at least a portion of the functionality is implemented as one or more virtual components. Some or all of the functions described herein may be implemented as virtual components executed by one or more virtual machines (VMs) implemented in one or more virtual environments 2100 hosted by one or more of hardware nodes, such as a hardware computing device that operates as a network node, UE, core network node, or host. Further, in embodiments in which the virtual node does not require radio connectivity (e.g., a core network node or host), then the node may be entirely virtualized.
[0267] Applications 2102 (which may alternatively be called software instances, virtual appliances, network functions, virtual nodes, virtual network functions, etc.) are run in the virtualization environment 0400 to implement some of the features, functions, and/or benefits of some of the embodiments disclosed herein. [0268] Hardware 2104 includes processing circuitry, memory that stores software and/or instructions executable by hardware processing circuitry, and/or other hardware devices as described herein, such as a network interface, input/output interface, and so forth. Software may be executed by the processing circuitry to instantiate one or more virtualization layers 2106 (also referred to as hypervisors or virtual machine monitors (VMMs)), provide VMs 2108a and 2108b (one or more of which may be generally referred to as VMs 2108), and/or perform any of the functions, features and/or benefits described in relation with some embodiments described herein. The virtualization layer 2106 may present a virtual operating platform that appears like networking hardware to the VMs 2108.
[0269] The VMs 2108 comprise virtual processing, virtual memory, virtual networking or interface and virtual storage, and may be run by a corresponding virtualization layer 2106. Different embodiments of the instance of a virtual appliance 2102 may be implemented on one or more of VMs 2108, and the implementations may be made in different ways. Virtualization of the hardware is in some contexts referred to as network function virtualization (NFV). NFV may be used to consolidate many network equipment types onto industry standard high volume server hardware, physical switches, and physical storage, which can be located in data centers, and customer premise equipment.
[0270] In the context of NFV, a VM 2108 may be a software implementation of a physical machine that runs programs as if they were executing on a physical, non-virtualized machine. Each of the VMs 2108, and that part of hardware 2104 that executes that VM, be it hardware dedicated to that VM and/or hardware shared by that VM with others of the VMs, forms separate virtual network elements. Still in the context of NFV, a virtual network function is responsible for handling specific network functions that run in one or more VMs 2108 on top of the hardware 2104 and corresponds to the application 2102.
[0271] Hardware 2104 may be implemented in a standalone network node with generic or specific components. Hardware 2104 may implement some functions via virtualization. Alternatively, hardware 2104 may be part of a larger cluster of hardware (e.g. such as in a data center or CPE) where many hardware nodes work together and are managed via management and orchestration 2110, which, among others, oversees lifecycle management of applications 2102. In some embodiments, hardware 2104 is coupled to one or more radio units that each include one or more transmitters and one or more receivers that may be coupled to one or more antennas. Radio units may communicate directly with other hardware nodes via one or more appropriate network interfaces and may be used in combination with the virtual components to provide a virtual node with radio capabilities, such as a radio access node or a base station. In some embodiments, some signaling can be provided with the use of a control system 2112 which may alternatively be used for communication between hardware nodes and radio units.
[0272] Figure 22 shows a communication diagram of a host 2202 communicating via a network node 2204 with a UE 2206 over a partially wireless connection in accordance with some embodiments. Example implementations, in accordance with various embodiments, of the UE (such as a UE 1712a of Figure 17 and/or UE 1800 of Figure 18), network node (such as network node 1710a of Figure 17 and/or network node 1900 of Figure 19), and host (such as host 1716 of Figure 17 and/or host 2000 of Figure 20) discussed in the preceding paragraphs will now be described with reference to Figure 22.
[0273] Like host 2000, embodiments of host 2202 include hardware, such as a communication interface, processing circuitry, and memory. The host 2202 also includes software, which is stored in or accessible by the host 2202 and executable by the processing circuitry. The software includes a host application that may be operable to provide a service to a remote user, such as the UE 2206 connecting via an over-the-top (OTT) connection 2250 extending between the UE 2206 and host 2202. In providing the service to the remote user, a host application may provide user data which is transmitted using the OTT connection 2250.
[0274] The network node 2204 includes hardware enabling it to communicate with the host 2202 and UE 2206. The connection 2260 may be direct or pass through a core network (like core network 1706 of Figure 17) and/or one or more other intermediate networks, such as one or more public, private, or hosted networks. For example, an intermediate network may be a backbone network or the Internet.
[0275] The UE 2206 includes hardware and software, which is stored in or accessible by UE 2206 and executable by the UE's processing circuitry. The software includes a client application, such as a web browser or operatorspecific "app” that may be operable to provide a service to a human or non-human user via UE 2206 with the support of the host 2202. In the host 2202, an executing host application may communicate with the executing client application via the OTT connection 2250 terminating at the UE 2206 and host 2202. In providing the service to the user, the UE's client application may receive request data from the host's host application and provide user data in response to the request data. The OTT connection 2250 may transfer both the request data and the user data. The UE's client application may interact with the user to generate the user data that it provides to the host application through the OTT connection 2250.
[0276] The OTT connection 2250 may extend via a connection 2260 between the host 2202 and the network node 2204 and via a wireless connection 2270 between the network node 2204 and the UE 2206 to provide the connection between the host 2202 and the UE 2206. The connection 2260 and wireless connection 2270, over which the OTT connection 2250 may be provided, have been drawn abstractly to illustrate the communication between the host 2202 and the UE 2206 via the network node 2204, without explicit reference to any intermediary devices and the precise routing of messages via these devices.
[0277] As an example of transmitting data via the OTT connection 2250, in step 2208, the host 2202 provides user data, which may be performed by executing a host application. In some embodiments, the user data is associated with a particular human user interacting with the UE 2206. In other embodiments, the user data is associated with a UE 2206 that shares data with the host 2202 without explicit human interaction. In step 2210, the host 2202 initiates a transmission carrying the user data towards the UE 2206. The host 2202 may initiate the transmission responsive to a request transmitted by the UE 2206. The request may be caused by human interaction with the UE 2206 or by operation of the client application executing on the UE 2206. The transmission may pass via the network node 2204, in accordance with the teachings of the embodiments described throughout this disclosure. Accordingly, in step 2212, the network node 2204 transmits to the UE 2206 the user data that was carried in the transmission that the host 2202 initiated, in accordance with the teachings of the embodiments described throughout this disclosure. In step 2214, the UE 2206 receives the user data carried in the transmission, which may be performed by a client application executed on the UE 2206 associated with the host application executed by the host 2202.
[0278] In some examples, the UE 2206 executes a client application which provides user data to the host 2202. The user data may be provided in reaction or response to the data received from the host 2202. Accordingly, in step 2216, the UE 2206 may provide user data, which may be performed by executing the client application. In providing the user data, the client application may further consider user input received from the user via an input/output interface of the UE 2206. Regardless of the specific manner in which the user data was provided, the UE 2206 initiates, in step 2218, transmission of the user data towards the host 2202 via the network node 2204. In step 2220, in accordance with the teachings of the embodiments described throughout this disclosure, the network node 2204 receives user data from the UE 2206 and initiates transmission of the received user data towards the host 2202. In step 2222, the host 2202 receives the user data carried in the transmission initiated by the UE 2206.
[0279] One or more of the various embodiments improve the performance of OTT services provided to the UE 2206 using the OTT connection 2250, in which the wireless connection 2270 forms the last segment. More precisely, the teachings of these embodiments may improve the amount and type of data available to a network to determine actions to be taken and the consequences thereof, and thereby provide benefits such as a more efficient control operation.
[0280] In an example scenario, factory status information may be collected and analyzed by the host 2202. As another example, the host 2202 may process audio and video data which may have been retrieved from a UE for use in creating maps. As another example, the host 2202 may collect and analyze real-time data to assist in controlling vehicle congestion (e.g., controlling traffic lights). As another example, the host 2202 may store surveillance video uploaded by a UE. As another example, the host 2202 may store or control access to media content such as video, audio, VR or AR which it can broadcast, multicast or unicast to UEs. As other examples, the host 2202 may be used for energy pricing, remote control of non-time critical electrical load to balance power generation needs, location services, presentation services (such as compiling diagrams etc. from data collected from remote devices), or any other function of collecting, retrieving, storing, analyzing and/or transmitting data.
[0281] In some examples, a measurement procedure may be provided for the purpose of monitoring data rate, latency and other factors on which the one or more embodiments improve. There may further be an optional network functionality for reconfiguring the OTT connection 2250 between the host 2202 and UE 2206, in response to variations in the measurement results. The measurement procedure and/or the network functionality for reconfiguring the OTT connection may be implemented in software and hardware of the host 2202 and/or UE 2206. In some embodiments, sensors (not shown) may be deployed in or in association with other devices through which the OTT connection 2250 passes; the sensors may participate in the measurement procedure by supplying values of the monitored quantities exemplified above, or supplying values of other physical quantities from which software may compute or estimate the monitored quantities. The reconfiguring of the OTT connection 2250 may include message format, retransmission settings, preferred routing etc.; the reconfiguring need not directly alter the operation of the network node 2204. Such procedures and functionalities may be known and practiced in the art. In certain embodiments, measurements may involve proprietary UE signaling that facilitates measurements of throughput, propagation times, latency and the like, by the host 2202. The measurements may be implemented in that software causes messages to be transmitted, in particular empty or 'dummy' messages, using the OTT connection 2250 while monitoring propagation times, errors, etc.
[0282] Although the computing devices described herein (e.g., UEs, network nodes, hosts) may include the illustrated combination of hardware components, other embodiments may comprise computing devices with different combinations of components. It is to be understood that these computing devices may comprise any suitable combination of hardware and/or software needed to perform the tasks, features, functions and methods disclosed herein. Determining, calculating, obtaining or similar operations described herein may be performed by processing circuitry, which may process information by, for example, converting the obtained information into other information, comparing the obtained information or converted information to information stored in the network node, and/or performing one or more operations based on the obtained information or converted information, and as a result of said processing making a determination. Moreover, while components are depicted as single boxes located within a larger box, or nested within multiple boxes, in practice, computing devices may comprise multiple different physical components that make up a single illustrated component, and functionality may be partitioned between separate components. For example, a communication interface may be configured to include any of the components described herein, and/or the functionality of the components may be partitioned between the processing circuitry and the communication interface. In another example, non-computationally intensive functions of any of such components may be implemented in software or firmware and computationally intensive functions may be implemented in hardware.
[0283] In certain embodiments, some or all of the functionality described herein may be provided by processing circuitry executing instructions stored on in memory, which in certain embodiments may be a computer program product in the form of a non-transitory computer-readable storage medium. In alternative embodiments, some or all of the functionality may be provided by the processing circuitry without executing instructions stored on a separate or discrete device-readable storage medium, such as in a hard-wired manner. In any of those particular embodiments, whether executing instructions stored on a non-transitory computer-readable storage medium or not, the processing circuitry can be configured to perform the described functionality. The benefits provided by such functionality are not limited to the processing circuitry alone or to other components of the computing device, but are enjoyed by the computing device as a whole, and/or by end users and a wireless network generally. The following numbered embodiments provide additional information on the disclosure.
1 . A method performed by a controller module for processing data, the method comprising: sending a request for metadata relating to a network to a proxy module; responsive to receiving from the proxy module metadata generated by a generative model, using the received metadata as input to a trained model; and determining an action to take based on the output of the trained model.
2. The method of any previous embodiment, wherein the output of the trained model comprises at least one of: an environmental state; a reward; an observation.
3. The method of any previous embodiment , further comprising: sending the determined action to at least one of a UE, a network node.
4. The method of any previous embodiment , further comprising: receiving a dataset comprising at least one of measured and simulated data, and wherein the dataset is additionally used as input to the trained model.
5. The method of embodiment 4, wherein the metadata is given a different weight to a weight given to the dataset.
6. The method of any previous embodiment, further comprising: sending at least one of: information indicating the type of metadata to be generated; information indicating the range within which the metadata is to be generated; at least one condition to be used for the generation of the metadata; information indicating at least one generative model to be used to generate the metadata; information indicating how the generative model is to be trained prior to generating the metadata; a forecasting time interval for generation of the metadata.
7. The method of any previous embodiment, further comprising: receiving at least one of: an indication of the type of data received; for each type of data received, receiving an indication of whether the data comprises generated metadata; an indication of which part of the data comprises generated metadata.
8. The method of any previous embodiment, further comprising: receiving at least one of: information indicating that the metadata is generated data; information indicating the type of metadata generated; information indicating the model used to generate the metadata; information indicating the conditions used to generate the metadata; information indicating how conditions were used to generate the metadata; information on the generative model; metadata values; information indicating the accuracy of the metadata.
9. The method of embodiment 7 or 8, wherein the method further comprises using the received information as input to the trained model.
10. The method of any previous embodiment, wherein the method is performed by a model inference function.
11. A method performed by a proxy module for generating data, the method comprising: receiving, from a controller module, a request for metadata relating to a network; generating metadata using a generative model; and sending the generated metadata to the controller module.
12. The method of embodiment 11, wherein at least one of a condition and a dataset is input to the generative model to generate metadata.
13. The method of embodiment 11 or 12, wherein the dataset comprises at least one of measured and simulated data.
14. The method of embodiment 12 or 13, wherein a condition comprises information relating to a contextual scenario.
15. The method of embodiment 11 to 14, the method further comprising: sending at least one of: information indicating that the metadata is conditionally generated; information indicating the type of metadata generated; information indicating the model used to generate the metadata; information indicating the conditions used to generate the metadata; information indicating how conditions used to generate the metadata were used; information on the generative model; metadata values; information indicating the accuracy of the metadata.
16. The method of embodiment 11 to 15, the method further comprising: receiving at least one of: information indicating the type of metadata to be generated; information indicating the range within which the metadata is to be generated; at least one condition to be used for the generation of the metadata; information indicating at least one generative model to be used to generate the metadata; information indicating how the generative model is to be trained prior to generating the metadata, and wherein the method further comprises using the received information as input to the generative model.
17. The method of embodiment 11 to 16, the method further comprising receiving feedback on the result of an action instructed by the controller module to at least one of a UE, a network node.
18. The method of embodiment 11 to 17, wherein the generated metadata comprises at least one of: information relating to data missing from a dataset; forecast metadata.
19. The method of any of embodiment 11 to 18, wherein the method is performed by a model inference function.
20. A method in a system, the system comprising a controller module configured to perform the method according to any of embodiment 1 to 10 and a proxy module configured to perform the method according to any of embodiment 11 to 19.
21. The method of embodiment 11 to 20, wherein the method further comprises receiving at the proxy module a dataset.
22. The method of embodiment 21, wherein method further comprises updating the dataset based on the metadata output from the generative model.
23. The method of embodiment 22 wherein the system further comprises a validation module, and the method further comprises processing the metadata by the validation module before updating the dataset.
24. The method of embodiment 23, wherein the validation module updates the dataset if the metadata fulfils an event criteria.
25. The method of embodiment 24, wherein an event criteria comprises at least one of: a rare event criteria; a quality wise criteria; extra desired data; metadata is derived using a new experiment.
26. The method of embodiment 25, wherein an experiment comprises at least one of: altering network configuration; adding or removing base stations; increasing or reducing coverage; altering radio environment; introducing coverage holes; changing propagation characteristics; adding or removing frequency layers; new channel ranks; user dynamics; more throttling at the cells; dynamic/constant type of traffic; measurement campaign that contains new margins of KPIs; shorter or longer margin of latency; shorter or longer margin of throughput; shorter or longer margin of reliability; shorter or longer margin of SI NR; shorter or longer margin of bandwidth; shorter or longer margin of center frequency.
27. A method performed by a controller module for training a model, the method comprising: sending a request for metadata relating to a network to a proxy module; responsive to receiving metadata from the proxy module generated by a generative model, using the metadata as training data to train a model.
28. The method of embodiment 27, wherein the model is trained to output information usable to determine an action to take based on input data.
29. The method of embodiment 27 or 28, wherein the training of the model is further based on data comprising at least one of measured and simulated data.
30. The method of embodiment 27 to 29, wherein the steps of the method are repeated to retrain the model in response to receiving new metadata.
31. The method of embodiment 27 to 30, wherein the trained model is used in the method according to embodiments 1 to 10, and wherein feedback on the performance of the model is used to retrain the model.
32. The method of embodiment 27 to 31, wherein the model is initially trained using a measured or simulated dataset.
33. The method of embodiment 27 to 32, wherein the method further comprises: sending at least one of: a request for a type of training data; an indication of the range of values within which each training data type is required; an indication that the training data can further comprise generated metadata; information indicating the conditions for metadata generation; information related to the generative model to be used to generate the metadata.
34. The method of embodiment 27 to 33, wherein the method further comprises: receiving at least one of: an indication of the type of data received; for each type of data received, receiving an indication of whether the data comprises generated metadata; an indication of which part of the data comprises generated metadata; an indication of the accuracy of the metadata; information on the generative model used to generate the metadata; and wherein the method further comprises using the received information to train the model.
35. The method of embodiment 27 to 34, further comprising sending information on the trained model.
36. The method of embodiment 27 to 35, wherein the method is performed by a model training function.
37. The method of embodiment 27 to 36, wherein the controller module is comprised in an Operation, Administration, and Management entity, OAM.
38. A method performed by a proxy module for training a generative model, the method comprising: receiving a dataset; receiving metadata; using the dataset and metadata to train a generative model to generate metadata based on an input dataset.
39. The method of embodiment 38, wherein the received metadata is at least one of: received explicitly; received implicitly; architectural based metadata.
40. The method of embodiment 38 or 39, wherein the training is further performed based on a condition comprising information relating to a contextual scenario.
41 . The method of embodiment 38 to 40, wherein the method further comprises receiving characteristics of an experiment and using the characteristics to train the generative model.
42. The method of embodiment 38 to 41 , wherein the method further comprises receiving an action and using the action to train the generative model.
43. The method of embodiment 38 to 42, wherein the steps of the method are repeated to retrain the generative model in response to receiving at least one of new metadata, a new dataset.
44. The method of any preceding embodiment, wherein the metadata comprises at least one of: an observation; a state; a reward
45. The method of any preceding embodiment, wherein the generative model comprises at least one of: a general adversarial network, GAN; a generative minimization network; a variational auto-encoder; a conditional GAN; an InfoGAN; a Wasserstein GAN; a neural network; a deep neural network. 46. The method of any preceding embodiment, wherein the method is implemented in a RAN framework.
47. The method of any preceding embodiment, wherein the method relates to at least one of: handover; energy consumption; traffic characteristics measurement; core network measurements; network load measurements; network performance measurements; slice related measurements; UE related analytics; UE congestions related measurements; QoS sustainability measurements.
48. A method in a system, the system comprising a controller module and a proxy module, the system to perform the method according to any preceding embodiment.
49. A method in a network node comprising at least one of a proxy module and a controller module, the network node to perform any of embodiments 1 to 48.
50. A method in a UE comprising at least one of a proxy module and a controller module, the UE to perform any of embodiments 1 to 48.
51 . A controller module for processing data and/or for training a model, comprising: processing circuitry configured to cause the controller module to perform any of the steps of any of embodiments 1 to 48; power supply circuitry configured to supply power to the processing circuitry.
52. A proxy module for generating data and/or for training a generative model, comprising: processing circuitry configured to cause the proxy module to perform any of the steps of any of embodiments 1 to 48; power supply circuitry configured to supply power to the processing circuitry.
53. A network node, the network node comprising: at least one of a controller module and a proxy module; processing circuitry configured to cause the network node to perform any of the steps of any of embodiments 1 to 48; power supply circuitry configured to supply power to the processing circuitry.
54. A user equipment, comprising: at least one of a controller module and a proxy module; processing circuitry configured to cause the user equipment to perform any of the steps of any of embodiments 1 to 48; and power supply circuitry configured to supply power to the processing circuitry.
55. A user equipment (UE), the UE comprising: an antenna configured to send and receive wireless signals; radio front-end circuitry connected to the antenna and to processing circuitry, and configured to condition signals communicated between the antenna and the processing circuitry; the processing circuitry being configured to perform any of the steps of any of embodiments 1 to 48; an input interface connected to the processing circuitry and configured to allow input of information into the UE to be processed by the processing circuitry; an output interface connected to the processing circuitry and configured to output information from the UE that has been processed by the processing circuitry; and a battery connected to the processing circuitry and configured to supply power to the UE.
56. A host configured to operate in a communication system to provide an over-the-top (OTT) service, the host comprising: processing circuitry configured to provide user data; and a network interface configured to initiate transmission of the user data to a cellular network for transmission to a user equipment (UE), wherein the UE comprises a communication interface and processing circuitry, the communication interface and processing circuitry of the UE being configured to perform any of the steps of any of the embodiments 1 to 48 to receive the user data from the host.
57. The host of the previous embodiment, wherein the cellular network further includes a network node configured to communicate with the UE to transmit the user data to the UE from the host.
58. The host of any of embodiments 56 and 57, wherein: the processing circuitry of the host is configured to execute a host application, thereby providing the user data; and the host application is configured to interact with a client application executing on the UE, the client application being associated with the host application. 59. A method implemented by a host operating in a communication system that further includes a network node and a user equipment (UE), the method comprising: providing user data for the UE; and initiating a transmission carrying the user data to the UE via a cellular network comprising the network node, wherein the UE performs any of the operations of any of the Group A embodiments to receive the user data from the host.
60. The method of embodiment 59, further comprising: at the host, executing a host application associated with a client application executing on the UE to receive the user data from the UE.
61 . The method of embodiment 60, further comprising: at the host, transmitting input data to the client application executing on the UE, the input data being provided by executing the host application, wherein the user data is provided by the client application in response to the input data from the host application.
62. A host configured to operate in a communication system to provide an over-the-top (OTT) service, the host comprising: processing circuitry configured to provide user data; and a network interface configured to initiate transmission of the user data to a cellular network for transmission to a user equipment (UE), wherein the UE comprises a communication interface and processing circuitry, the communication interface and processing circuitry of the UE being configured to perform any of the steps of any of embodiments 1 to 48 to transmit the user data to the host.
63. The host of embodiment 62, wherein the cellular network further includes a network node configured to communicate with the UE to transmit the user data from the UE to the host.
64. The host of any of embodiments 62 and 63, wherein: the processing circuitry of the host is configured to execute a host application, thereby providing the user data; and the host application is configured to interact with a client application executing on the UE, the client application being associated with the host application.
65. A method implemented by a host configured to operate in a communication system that further includes a network node and a user equipment (UE), the method comprising: at the host, receiving user data transmitted to the host via the network node by the UE, wherein the UE performs any of the steps of any of embodiments 1 to 48 to transmit the user data to the host.
66. The method of embodiment 65, further comprising: at the host, executing a host application associated with a client application executing on the UE to receive the user data from the UE.
67. The method of embodiment 66, further comprising: at the host, transmitting input data to the client application executing on the UE, the input data being provided by executing the host application, wherein the user data is provided by the client application in response to the input data from the host application.
68. A host configured to operate in a communication system to provide an over-the-top (OTT) service, the host comprising: processing circuitry configured to provide user data; and a network interface configured to initiate transmission of the user data to a network node in a cellular network for transmission to a user equipment (UE), the network node having a communication interface and processing circuitry, the processing circuitry of the network node configured to perform any of the operations of any of embodiments 1 to 48 to transmit the user data from the host to the UE.
69. The host of embodiment 68, wherein: the processing circuitry of the host is configured to execute a host application that provides the user data; and the UE comprises processing circuitry configured to execute a client application associated with the host application to receive the transmission of user data from the host.
70. A method implemented in a host configured to operate in a communication system that further includes a network node and a user equipment (UE), the method comprising: providing user data for the UE; and initiating a transmission carrying the user data to the UE via a cellular network comprising the network node, wherein the network node performs any of the operations of any of embodiments 1 to 48 to transmit the user data from the host to the UE.
71. The method of embodiment 70, further comprising, at the network node, transmitting the user data provided by the host for the UE.
72. The method of any embodiments 70 and 71 , wherein the user data is provided at the host by executing a host application that interacts with a client application executing on the UE, the client application being associated with the host application.
73. A communication system comprising any of a proxy module, a controller module, a validation module, the communication system configured to perform any of the steps of embodiments 1 to 48.
74. A communication system configured to provide an over-the-top service, the communication system comprising: a host comprising: processing circuitry configured to provide user data for a user equipment (UE), the user data being associated with the over-the-top service; and a network interface configured to initiate transmission of the user data toward a cellular network node for transmission to the UE, the network node having a communication interface and processing circuitry, the processing circuitry of the network node configured to perform any of the operations of any of embodiments 1 to 48 to transmit the user data from the host to the UE.
75. The communication system of embodiment 74, further comprising: the network node; and/or the user equipment.
76. A host configured to operate in a communication system to provide an over-the-top (OTT) service, the host comprising: processing circuitry configured to initiate receipt of user data; and a network interface configured to receive the user data from a network node in a cellular network, the network node having a communication interface and processing circuitry, the processing circuitry of the network node configured to perform any of the operations of any of embodiments 1 to 48 to receive the user data from a user equipment (UE) for the host.
77. The host of embodiment 76, wherein: the processing circuitry of the host is configured to execute a host application, thereby providing the user data; and the host application is configured to interact with a client application executing on the UE, the client application being associated with the host application.
78. The host of the any of embodiments 76 and 77, wherein the initiating receipt of the user data comprises requesting the user data.
79. A method implemented by a host configured to operate in a communication system that further includes a network node and a user equipment (UE), the method comprising: at the host, initiating receipt of user data from the UE, the user data originating from a transmission which the network node has received from the UE, wherein the network node performs any of the steps of any of embodiments 1 to 48 to receive the user data from the UE for the host.
80. The method of embodiment 79, further comprising at the network node, transmitting the received user data to the host.

Claims

Claims
1 . A method performed by a controller module for processing data, the method comprising: sending (402) a request for metadata relating to a network to a proxy module; responsive to receiving from the proxy module metadata generated by a generative model, using (404) the received metadata as input to a trained model; and determining (406) an action to take based on the output of the trained model.
2. The method of claim 1 , wherein the output of the trained model comprises at least one of: an environmental state; a reward; an observation.
3. The method of any previous claim , further comprising: sending the determined action to at least one of a UE (1712), a network node (1710).
4. The method of any previous claim , further comprising: receiving a dataset comprising at least one of measured and simulated data, and wherein the dataset is additionally used as input to the trained model, optionally wherein the metadata is given a different weight to a weight given to the dataset.
5. The method of any previous claim, further comprising: sending at least one of: information indicating the type of metadata to be generated; information indicating the range within which the metadata is to be generated; at least one condition to be used for the generation of the metadata; information indicating at least one generative model to be used to generate the metadata; information indicating how the generative model is to be trained prior to generating the metadata; a forecasting time interval for generation of the metadata.
6. The method of any previous claim, further comprising: receiving at least one of: an indication of the type of data received; for each type of data received, receiving an indication of whether the data comprises generated metadata; an indication of which part of the data comprises generated metadata, and/or receiving at least one of: information indicating that the metadata is generated data; information indicating the type of metadata generated; information indicating the model used to generate the metadata; information indicating the conditions used to generate the metadata; information indicating how conditions were used to generate the metadata; information on the generative model; metadata values; information indicating the accuracy of the metadata.
7. The method of claim 6, wherein the method further comprises using the received information as input to the trained model.
8. The method of any previous claim, wherein the method is performed by a model inference function.
9. A method performed by a proxy module for generating data, the method comprising:
Receiving (502), from a controller module, a request for metadata relating to a network; generating (504) metadata using a generative model; and sending (506) the generated metadata to the controller module.
10. The method of claim 9, wherein at least one of a condition and a dataset is input to the generative model to generate metadata, and/or wherein the dataset comprises at least one of measured and simulated data.
11 . The method of claim 10, wherein a condition comprises information relating to a contextual scenario.
12. The method of claim 9 to 11, the method further comprising: sending at least one of: information indicating that the metadata is conditionally generated; information indicating the type of metadata generated; information indicating the model used to generate the metadata; information indicating the conditions used to generate the metadata; information indicating how conditions used to generate the metadata were used; information on the generative model; metadata values; information indicating the accuracy of the metadata, and/or the method further comprising: receiving at least one of: information indicating the type of metadata to be generated; information indicating the range within which the metadata is to be generated; at least one condition to be used for the generation of the metadata; information indicating at least one generative model to be used to generate the metadata; information indicating how the generative model is to be trained prior to generating the metadata, and wherein the method further comprises using the received information as input to the generative model.
13. The method of claim 9 to 12, the method further comprising: receiving feedback on the result of an action instructed by the controller module to at least one of a UE (1712), a network node (1710); and/or wherein the generated metadata comprises at least one of: information relating to data missing from a dataset; forecast metadata; and/or wherein the method is performed by a model inference function.
. A method in a system, the system comprising a controller module configured to perform the method according to any of claim 1 to 8 and a proxy module configured to perform the method according to any of claim 9 to 13. . The method of claim 9 to 14, wherein the method further comprises receiving at the proxy module a dataset. . The method of claim 15, wherein method further comprises updating the dataset based on the metadata output from the generative model. . The method of claim 16 wherein the system further comprises a validation module, and the method further comprises processing the metadata by the validation module before updating the dataset, optionally wherein the validation module updates the dataset if the metadata fulfils an event criteria. . The method of claim 17, wherein an event criteria comprises at least one of: a rare event criteria; a quality wise criteria; extra desired data; metadata is derived using a new experiment. . The method of claim 18, wherein an experiment comprises at least one of: altering network configuration; adding or removing base stations; increasing or reducing coverage; altering radio environment; introducing coverage holes; changing propagation characteristics; adding or removing frequency layers; new channel ranks; user dynamics; more throttling at the cells; dynamic/constant type of traffic; measurement campaign that contains new margins of KPIs; shorter or longer margin of latency; shorter or longer margin of throughput; shorter or longer margin of reliability; shorter or longer margin of SINR; shorter or longer margin of bandwidth; shorter or longer margin of center frequency. . A method performed by a controller module for training a model, the method comprising: sending (602) a request for metadata relating to a network to a proxy module; responsive to receiving metadata from the proxy module generated by a generative model, using (604) the metadata as training data to train a model. . The method of claim 20, wherein the model is trained to output information usable to determine an action to take based on input data. . The method of claim 20 or 21 , wherein the training of the model is further based on data comprising at least one of measured and simulated data, and/or wherein the steps of the method are repeated to retrain the model in response to receiving new metadata.
23. The method of claim 20 to 22, wherein the trained model is used in the method according to claims 1 to 8, and wherein feedback on the performance of the model is used to retrain the model.
24. The method of claim 20 to 23, wherein the model is initially trained using a measured or simulated dataset.
25. The method of claim 20 to 24, wherein the method further comprises: sending at least one of: a request for a type of training data; an indication of the range of values within which each training data type is required; an indication that the training data can further comprise generated metadata; information indicating the conditions for metadata generation; information related to the generative model to be used to generate the metadata, and/or wherein the method further comprises: receiving at least one of: an indication of the type of data received; for each type of data received, receiving an indication of whether the data comprises generated metadata; an indication of which part of the data comprises generated metadata; an indication of the accuracy of the metadata; information on the generative model used to generate the metadata; and wherein the method further comprises using the received information to train the model.
26. The method of claim 20 to 25, further comprising sending information on the trained model, and/or wherein the method is performed by a model training function, and/or wherein the controller module is comprised in an Operation, Administration, and Management entity, OAM.
27. A method performed by a proxy module for training a generative model, the method comprising:
Receiving (702) a dataset; receiving (704) metadata; using (706) the dataset and metadata to train a generative model to generate metadata based on an input dataset.
28. The method of claim 27, wherein the received metadata is at least one of: received explicitly; received implicitly; architectural based metadata.
29. The method of claim 27 or 28, wherein the training is further performed based on a condition comprising information relating to a contextual scenario.
30. The method of claim 27 to 29, wherein the method further comprises receiving characteristics of an experiment and using the characteristics to train the generative model, and/or wherein the method further comprises receiving an action and using the action to train the generative model.
31 . The method of claim 27 to 30, wherein the steps of the method are repeated to retrain the generative model in response to receiving at least one of new metadata, a new dataset.
32. The method of any preceding claim, wherein the metadata comprises at least one of: an observation; a state; a reward, and/or wherein the generative model comprises at least one of: a general adversarial network, GAN; a generative minimization network; a variational auto-encoder; a conditional GAN; an InfoGAN; a Wasserstein GAN; a neural network; a deep neural network.
33. The method of any preceding claim, wherein the method is implemented in a RAN framework.
34. The method of any preceding claim, wherein the method relates to at least one of: handover; energy consumption; traffic characteristics measurement; core network measurements; network load measurements; network performance measurements; slice related measurements; UE related analytics; UE congestions related measurements; QoS sustainability measurements.
35. A controller module for processing data and/or for training a model, comprising: processing circuitry configured to cause the controller module to: send a request for metadata relating to a network to a proxy module; responsive to receiving from the proxy module metadata generated by a generative model, use the received metadata as input to a trained model; and determine an action to take based on the output of the trained model, and ; power supply circuitry configured to supply power to the processing circuitry.
36. A proxy module for generating data and/or for training a generative model, comprising: processing circuitry configured to cause the proxy module to: receive a dataset; receive metadata; use the dataset and metadata to train a generative model to generate metadata based on an input dataset, and ; power supply circuitry configured to supply power to the processing circuitry.
37. A network node (1710), the network node (1710) comprising: at least one of the controller module of claim 35 and the proxy module of claim 36. A user equipment (1712), comprising: at least one of the controller module of claim 35 and the proxy module of claim 36. A communication system comprising at least one of: the network node (1710) of claim 37; and/or the user equipment (1712) of claim 38.
PCT/EP2023/052345 2022-02-10 2023-01-31 Incorporating conditions into data-collection & ai/ml operations WO2023151989A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US202263308755P 2022-02-10 2022-02-10
US63/308,755 2022-02-10

Publications (1)

Publication Number Publication Date
WO2023151989A1 true WO2023151989A1 (en) 2023-08-17

Family

ID=85505581

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/EP2023/052345 WO2023151989A1 (en) 2022-02-10 2023-01-31 Incorporating conditions into data-collection & ai/ml operations

Country Status (1)

Country Link
WO (1) WO2023151989A1 (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3618077A1 (en) * 2018-08-27 2020-03-04 Koninklijke Philips N.V. Generating metadata for trained model
US20210042645A1 (en) * 2019-08-06 2021-02-11 doc.ai, Inc. Tensor Exchange for Federated Cloud Learning

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3618077A1 (en) * 2018-08-27 2020-03-04 Koninklijke Philips N.V. Generating metadata for trained model
US20210042645A1 (en) * 2019-08-06 2021-02-11 doc.ai, Inc. Tensor Exchange for Federated Cloud Learning

Non-Patent Citations (6)

* Cited by examiner, † Cited by third party
Title
3GPP DOCUMENT R3-221014, 18 January 2023 (2023-01-18), Retrieved from the Internet <URL:https://www.3gpp.org/ftp/TSGRAN/WG3_Iu/TSGR3_114bis-e/Docs/?sortby=sizerev>
3GPP FRAMEWORK IN TR 37.817, 18 January 2023 (2023-01-18), Retrieved from the Internet <URL:https://portal.3gpp.org/desktopmodules/Specifications/SpecificationDetails.aspx?specificationld=3817>
3GPP TR 37.817
3GPP TS 23.501, 30 January 2023 (2023-01-30), Retrieved from the Internet <URL:https://portal.3gpp.org/desktopmodules/Specifications/SpecificationDetails.aspx?specificationld=3144>
3GPP TS 33.501, 30 January 2023 (2023-01-30), Retrieved from the Internet <URL:https://portal.3gpp.org/desktopmodules/Specifications/SpecificationDetails.aspx?specificationld=3169>
3GPP TS 38.300, 18 January 2023 (2023-01-18)

Similar Documents

Publication Publication Date Title
WO2023022642A1 (en) Reporting of predicted ue overheating
WO2023151989A1 (en) Incorporating conditions into data-collection &amp; ai/ml operations
WO2023206238A1 (en) Method and apparatus for dynamically configuring slice in communication network
WO2024094176A1 (en) L1 data collection
WO2023191682A1 (en) Artificial intelligence/machine learning model management between wireless radio nodes
WO2023113678A1 (en) Reporting performance impacts associated to load
WO2024096796A1 (en) Flexible qoe configuration for qoe handling
WO2023239287A1 (en) Machine learning for radio access network optimization
WO2023186724A1 (en) Radio access network (ran) analytics exposure mechanism
WO2023209557A1 (en) Intent based automation for predictive route performance
WO2023012351A1 (en) Controlling and ensuring uncertainty reporting from ml models
WO2024075130A1 (en) Optimizing user equipment service level agreement violations for network slice allocation
WO2023218271A1 (en) Adjusting a physical route based on real-time connectivity data
WO2023069001A1 (en) Reconfiguration procedure in a wireless communication network
WO2023187678A1 (en) Network assisted user equipment machine model handling
EP4364377A1 (en) Boost enhanced active measurement
WO2024096803A1 (en) Enhanced mobility optimization using ue trajectory prediction
WO2023217557A1 (en) Artificial intelligence/machine learning (ai/ml) translator for 5g core network (5gc)
WO2024035305A1 (en) Successful pscell change or addition report
WO2023140768A1 (en) Opportunistic alignment of mdt and qoe
WO2024096805A1 (en) Communicating based on network configuration identifier sharing
EP4378202A1 (en) Methods and apparatuses for controlling load reporting
WO2024030059A1 (en) Quality of experience measurement
WO2023148010A1 (en) Network-centric life cycle management of ai/ml models deployed in a user equipment (ue)
WO2023131845A1 (en) Network based handling of quality of experience session status indication during conditional handover

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23709320

Country of ref document: EP

Kind code of ref document: A1