WO2023143572A1 - 基于人工智能ai模型的定位方法及通信设备 - Google Patents

基于人工智能ai模型的定位方法及通信设备 Download PDF

Info

Publication number
WO2023143572A1
WO2023143572A1 PCT/CN2023/073715 CN2023073715W WO2023143572A1 WO 2023143572 A1 WO2023143572 A1 WO 2023143572A1 CN 2023073715 W CN2023073715 W CN 2023073715W WO 2023143572 A1 WO2023143572 A1 WO 2023143572A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
model
measurement
method based
positioning method
Prior art date
Application number
PCT/CN2023/073715
Other languages
English (en)
French (fr)
Inventor
王园园
邬华明
庄子荀
Original Assignee
维沃移动通信有限公司
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 维沃移动通信有限公司 filed Critical 维沃移动通信有限公司
Publication of WO2023143572A1 publication Critical patent/WO2023143572A1/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W64/00Locating users or terminals or network equipment for network management purposes, e.g. mobility management
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00Machine learning
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14Network analysis or design
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14Network analysis or design
    • H04L41/145Network analysis or design involving simulating, designing, planning or modelling of a network

Definitions

  • the present application belongs to the technical field of communication, and in particular relates to a positioning method and communication equipment based on an artificial intelligence AI model.
  • AI artificial intelligence
  • a positioning method based on an artificial intelligence AI model is provided, which is applied to a first communication device, and the method includes:
  • the first communication device acquires first information associated with information related to the AI model
  • the first communication device determines target information according to the first information, and the target information includes at least one of the following: a target AI model, validity information of information related to the AI model, or positioning based on the target AI model received feedback;
  • the first information is used to indicate the effective scope of application of the AI model-related information
  • the AI model-related information includes at least one of the following: AI model, AI model parameters, AI model input, and AI model output.
  • a positioning method based on an AI model is provided, which is applied to a second communication device, and the method includes:
  • the second communication device receives the target information sent by the first communication device, and the target information includes at least one of the following: a target AI model, validity information of information related to the AI model, or feedback obtained from positioning based on the target AI model information;
  • the target information is determined according to the first information associated with the AI model-related information, the first information is used to indicate the effective scope of application of the AI model-related information, and the AI model-related information includes at least one of the following: AI Model, AI model parameters, AI model input, AI model output.
  • an AI model-based positioning device including:
  • An acquisition module configured to acquire first information associated with information related to the AI model
  • a processing module configured to determine target information according to the first information, where the target information includes at least one of the following: a target AI model, validity information of information related to the AI model, or positioning based on the target AI model. feedback information;
  • the first information is used to indicate the effective scope of application of the AI model-related information
  • the AI model-related information includes at least one of the following: AI model, AI model parameters, AI model input, and AI model output.
  • an AI model-based positioning device including:
  • a receiving module configured to receive target information sent by the first communication device, where the target information includes at least one of the following: a target AI model, validity information of information related to the AI model, or information obtained by positioning based on the target AI model Feedback;
  • the target information is determined according to the first information associated with the AI model-related information, the first information is used to indicate the effective scope of application of the AI model-related information, and the AI model-related information includes at least one of the following: AI Model, AI model parameters, AI model input, AI model output.
  • a first communication device in a fifth aspect, includes a processor and a memory, the memory stores programs or instructions that can run on the processor, and the programs or instructions are processed by the The steps of the method described in the first aspect are realized when the controller is executed.
  • a first communication device including a processor and a communication interface, wherein the communication interface is used to obtain first information associated with information related to the AI model; information to determine the target information, the target information includes at least one of the following: the target AI model, the validity information of the AI model-related information, or the feedback information obtained by positioning based on the target AI model; the first information uses
  • the AI model-related information includes at least one of the following: AI model, AI model parameters, AI model input, and AI model output.
  • a second communication device in a seventh aspect, includes a processor and a memory, the memory stores programs or instructions that can run on the processor, and the programs or instructions are processed by the implement the steps of the method as described in the second aspect when the controller is executed.
  • a second communication device including a processor and a communication interface, wherein the communication interface is used to receive target information sent by the first communication device, and the target information includes at least one of the following: The target AI model, the validity information of the AI model-related information, or the feedback information obtained by positioning based on the target AI model; the target information is determined according to the first information associated with the AI model-related information, and the first The information is used to indicate the effective scope of application of the AI model-related information, and the AI model-related information includes at least one of the following: AI model, AI model parameters, AI model input, and AI model output.
  • a communication system including: a first communication device and a second communication device, the first communication device can be used to perform the steps of the positioning method based on the AI model as described in the first aspect, the The second communication device can be used to execute the steps of the positioning method based on the AI model as described in the second aspect.
  • a readable storage medium is provided, and a program or an instruction is stored on the readable storage medium, and when the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented, or the steps of the method as described in the first aspect are implemented, or the The steps of the method described in the second aspect.
  • a chip in an eleventh aspect, includes a processor and a communication interface, the communication interface is coupled to the processor, and the processor is used to run a program or an instruction to implement the method described in the first aspect. method, or implement the method as described in the second aspect.
  • a twelfth aspect provides a computer program/program product, the computer program/program product is stored in a storage medium, and the computer program/program product is executed by at least one processor to implement the The steps of the positioning method based on the AI model described in the second aspect.
  • the first communication device acquires first information associated with information related to the AI model; the first communication device determines target information according to the first information, and the target information includes at least one of the following: target AI model, AI model The validity information of relevant information or the feedback information obtained by positioning based on the target AI model. Since the first information is used to indicate the effective scope of application of the relevant information of the AI model, determining the target information based on the first information can make the positioned AI model more in line with According to actual scene requirements, the AI model and/or AI model parameters can also be updated based on the determined target information, so that the accuracy of the positioning result obtained based on the updated AI model is higher.
  • FIG. 1 is a structural diagram of a wireless communication system applicable to an embodiment of the present application
  • Fig. 2 is one of the functional block diagrams of the AI model of the positioning method based on the AI model provided by the embodiment of the present application;
  • Fig. 3 is the second functional block diagram of the AI model of the positioning method based on the AI model provided by the embodiment of the present application;
  • Fig. 4 is one of the schematic flow charts of the positioning method based on the AI model provided by the embodiment of the present application;
  • Fig. 5 is one of the interactive flowcharts of the positioning method based on the AI model provided by the embodiment of the present application;
  • Fig. 6 is the second schematic diagram of the interaction process of the positioning method based on the AI model provided by the embodiment of the present application;
  • Fig. 7 is the third schematic diagram of the interaction process of the positioning method based on the AI model provided by the embodiment of the present application.
  • Fig. 8 is one of the schematic diagrams of the parameter range of the positioning method based on the AI model provided by the embodiment of the present application;
  • Fig. 9 is the second schematic diagram of the parameter range of the positioning method based on the AI model provided by the embodiment of the present application.
  • Fig. 10 is the fourth schematic diagram of the interaction process of the positioning method based on the AI model provided by the embodiment of the present application.
  • Fig. 11 is the fifth schematic diagram of the interaction process of the positioning method based on the AI model provided by the embodiment of the present application.
  • Figure 12 is the sixth schematic diagram of the interaction process of the positioning method based on the AI model provided by the embodiment of the present application
  • Figure 13 is one of the structural schematic diagrams of the positioning device based on the AI model provided by the embodiment of the present application
  • Fig. 14 is the second structural schematic diagram of the positioning device based on the AI model provided by the embodiment of the present application.
  • FIG. 15 is a schematic structural diagram of a communication device provided by an embodiment of the present application.
  • FIG. 16 is a schematic diagram of a hardware structure of a terminal provided by an embodiment of the present application.
  • FIG. 17 is a schematic structural diagram of a network-side device according to an embodiment of the present application.
  • FIG. 18 is another schematic structural diagram of a network-side device according to an embodiment of the present application.
  • first, second and the like in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific sequence or sequence. It is to be understood that the terms so used are interchangeable under appropriate circumstances such that the embodiments of the application are capable of operation in sequences other than those illustrated or described herein and that "first" and “second” distinguish objects. It is usually one category, and the number of objects is not limited. For example, there may be one or more first objects.
  • “and/or” in the description and claims means at least one of the connected objects, and the character “/” generally means that the related objects are an "or” relationship.
  • LTE Long Term Evolution
  • LTE-Advanced LTE-Advanced
  • LTE-A Long Term Evolution-Advanced
  • CDMA Code Division Multiple Access
  • TDMA Time Division Multiple Access
  • FDMA Frequency Division Multiple Access
  • OFDMA Orthogonal Frequency Division Multiple Access
  • SC-FDMA Single-carrier Frequency Division Multiple Access
  • system and “network” in the embodiments of the present application are often used interchangeably, and the described technology can be used for the above-mentioned system and radio technology, and can also be used for other systems and radio technologies.
  • NR New Radio
  • the following description describes the New Radio (NR) system for illustrative purposes, and uses NR terminology in most of the following descriptions, but these techniques can also be applied to applications other than NR system applications, such as the 6th generation (6 th Generation, 6G) communication system.
  • 6G 6th Generation
  • Fig. 1 shows a block diagram of a wireless communication system to which the embodiment of the present application is applicable.
  • Wireless communication systems include Terminal 11 and network side device 12.
  • the terminal 11 can be a mobile phone, a tablet computer (Tablet Personal Computer), a laptop computer (Laptop Computer) or a notebook computer, a personal digital assistant (Personal Digital Assistant, PDA), a palmtop computer, a netbook, a super mobile personal computer (ultra-mobile personal computer, UMPC), mobile Internet device (Mobile Internet Device, MID), augmented reality (augmented reality, AR) / virtual reality (virtual reality, VR) equipment, robot, wearable device (Wearable Device) , vehicle equipment (VUE), pedestrian terminal (PUE), smart home (home equipment with wireless communication functions, such as refrigerators, TVs, washing machines or furniture, etc.), game consoles, personal computers (personal computers, PCs), teller machines or self-service Wearable devices include: smart watches, smart bracelets, smart headphones, smart glasses, smart jewelry (smart bracelets, smart
  • the network side device 12 may include an access network device or a core network device, wherein the access network device 12 may also be called a radio access network device, a radio access network (Radio Access Network, RAN), a radio access network function, or Wireless access network unit.
  • RAN Radio Access Network
  • RAN Radio Access Network
  • Wireless access network unit Wireless access network unit
  • the access network device 12 may include a base station, a WLAN access point, or a WiFi node, etc., and the base station may be called a node B, an evolved node B (eNB), an access point, a base transceiver station (Base Transceiver Station, BTS), a radio Base station, radio transceiver, Basic Service Set (BSS), Extended Service Set (ESS), Home Node B, Home Evolved Node B, Transmitting Receiving Point (TRP) or all As long as the same technical effect is achieved, the base station is not limited to a specific technical vocabulary. It should be noted that in this embodiment of the application, only the base station in the NR system is used as an example for introduction, and The specific type of the base station is not limited.
  • the core network equipment may include but not limited to at least one of the following: core network node, core network function, mobility management entity (Mobility Management Entity, MME), access mobility management function (Access and Mobility Management Function, AMF), session management function (Session Management Function, SMF), user plane function (User Plane Function, UPF), policy control function (Policy Control Function, PCF), policy and charging rules function unit (Policy and Charging Rules Function, PCRF), edge application service Discovery function (Edge Application Server Discovery Function, EASDF), unified data management (Unified Data Management, UDM), unified data storage (Unified Data Repository, UDR), home subscriber server (Home Subscriber Server, HSS), centralized network configuration ( Centralized network configuration, CNC), network storage function (Network Repository Function, NRF), network exposure function (Network Exposure Function, NEF), local NEF (Local NEF, or L-NEF), binding support function (Binding Support Function, BSF), application function (Application Function, AF), location management function L
  • AI is currently widely used in various fields.
  • the embodiment of the present application uses a neural network as an example for illustration, but does not limit the specific type of the AI model.
  • the neural network is composed of neurons, where a 1 , a 2 ,...a K is the input, w is the weight (multiplicative coefficient), b is the bias (additive coefficient), and ⁇ (.) is the activation function.
  • Common activation functions include Sigmoid, tanh, Rectified Linear Unit (ReLU), etc.
  • the parameters of the neural network are optimized by an optimization algorithm.
  • An optimization algorithm is a class of algorithms that can help us minimize or maximize an objective function (sometimes called a loss function).
  • the objective function is often a mathematical combination of model parameters and data. For example, given the data X and its corresponding label Y, we construct a neural network model f(.), with the model, the predicted output f(x) can be obtained according to the input x, and the predicted value and the real value can be calculated The gap between (f(x)-y), this is the loss function.
  • Our purpose is to find the appropriate w, b to minimize the value of the above loss function, the smaller the loss value, the closer our model is to the real situation.
  • the current common optimization algorithms are basically based on the error back propagation (error Back Propagation, BP) algorithm.
  • BP error Back Propagation
  • the basic idea of the BP algorithm is that the learning process consists of two processes: the forward propagation of the signal and the back propagation of the error.
  • the input samples are passed in from the input layer, processed layer by layer by each hidden layer, and passed to the output layer. If the actual output of the output layer does not match the expected output, it will enter the error backpropagation stage.
  • Error backpropagation is to transmit the output error layer by layer through the hidden layer to the input layer in some form, and distribute the error to all the units of each layer, so as to obtain the error signal of each layer unit, and this error signal is used as the correction unit Basis for weight.
  • This weight adjustment process of each layer of signal forward propagation and error back propagation is carried out repeatedly.
  • the process of continuously adjusting the weights is also the learning and training process of the network. This process has been carried out until the error of the network output is reduced to an acceptable level, or until the preset number of learning times.
  • optimization algorithms are based on the error/loss obtained by the loss function when the error is backpropagated, and the derivative/partial derivative of the current neuron is calculated, and the learning rate, the previous gradient/derivative/partial derivative, etc. are added to obtain the gradient. Pass the gradient to the previous layer.
  • the method of the embodiment of this application can be applied to positioning scenarios, such as positioning based on an AI model. Since the measurement environment and requirements will change over time, the positioning results obtained by the AI model may not meet the current requirements as time goes by. Therefore, the method of the embodiment of the present application can determine whether the current AI model is valid based on the actual situation, and update the AI model and/or AI model parameters based on the judgment, so that the AI positioning The results met the performance specifications.
  • FIG. 2 and FIG. 3 are functional block diagrams of AI model applications.
  • AI model application (AI inference) is the process of obtaining output based on AI model, AI model parameters and current input data.
  • Fig. 4 is one of the schematic flowcharts of the AI model-based positioning method provided by the embodiment of the present application. As shown in Figure 4, the method provided in this embodiment includes:
  • Step 101 the first communication device acquires first information associated with information related to the AI model
  • the first information is used to indicate the effective scope of application of the AI model-related information
  • the AI model-related information includes at least one of the following: AI model, AI model parameters, AI model input, and AI model output.
  • the first information is associated with information related to the AI model, and the first information indicates the effective scope of application of the information related to the AI model, such as the valid range of input, valid range of output, validity conditions, etc.;
  • the information related to the AI model only ensures that the selected and specific AI model can be understood and confirmed by the first communication device and other communication devices, and the specific transmission format and transmission method are not limited here.
  • the AI model-related information is also a corresponding feature for protecting the AI model.
  • the first communication device may obtain information related to the AI model and information related to the AI model from the second communication device (for example, a model management device, a network-side device (such as NWDAF), a positioning server (such as LMF, E-SMLC), etc.)
  • the first information associated with related information.
  • the first information may be included in the AI model-related information, or transmitted together with the AI model-related information, or transmitted independently of the AI model-related information, but through the identification information of the AI model and the first information. associated.
  • Step 102 The first communication device determines target information according to the first information, and the target information includes at least one of the following: a target AI model, validity information of information related to the AI model, or feedback information obtained by positioning based on the target AI model.
  • the target information may be determined according to the effective scope of application of the parameters in the first information.
  • the first information associated with the relevant information of the AI model can determine the effective scope of application of the relevant information of the AI model, indicating that the relevant information of the AI model is valid in a specific scene and a specific area.
  • the output of the AI model is in The effective scope of application of the parameters in the first information indicates that the output of the AI model is valid.
  • the input of the AI model is within the effective application range of the parameters in the first information, it means that the input and output of the AI model is valid or the AI model is valid.
  • the several AI models are respectively associated with corresponding first information, and one or more AI models of the multiple AI models are determined as the target AI model according to the first information.
  • the feedback information may include validity information, a target AI model, an input of the AI model, an output of the AI model, measurement information of the terminal, and the like.
  • the feedback information has a corresponding relationship with the first information.
  • the first communication device may send the feedback information to the second communication device, and the second communication device may update the AI model and/or AI model parameters based on the feedback information.
  • the first communication device includes at least one of the following: a location management function (Location Management Function, LMF) network element and an evolution device of the LMF network element;
  • LMF Location Management Function
  • NWDAF Network Data Analytics Function
  • network-side devices include, for example, access network devices, and devices other than the aforementioned core network elements.
  • the first communication device obtains the first information associated with the information related to the AI model; the first communication device determines the target information according to the first information, and the target information includes at least one of the following: target AI model, AI model related The validity information of the information or the feedback information obtained by positioning based on the target AI model. Since the first information is used to indicate the effective scope of application of the relevant information of the AI model, determining the target information based on the first information can make the positioned AI model more realistic According to the scene requirements, the AI model and/or AI model parameters can also be updated based on the determined target information, so that the accuracy of the positioning result obtained based on the updated AI model is higher.
  • the validity information may include at least one of the following:
  • Validity indication information used to indicate whether the relevant information of the AI model is valid
  • Reliability indication information used to indicate whether the positioning results based on the target AI model are reliable
  • the validity indication information may be indicated by at least one bit, for example, 0 indicates failure, and 1 indicates validity.
  • the reliability indication information is similar to the validity indication information.
  • the degree of validity can be represented by at least one bit, for example, the degree of validity between 0-1.
  • the input of the AI model and/or the output of the AI model include at least one of the following:
  • the error information includes at least one of the following: position error value, measurement error value, AI model error value or AI model parameter error value.
  • the method also includes:
  • the first communication device acquires second information of the target terminal; the second information is used to represent the location-related information acquired by the target terminal;
  • Step 102 can be realized in the following manner:
  • the first communication device determines target information according to the first information and the second information.
  • the first information is used to indicate the effective scope of application of the AI model-related information
  • the second information is used to indicate the positioning-related information obtained by the target terminal, such as the measurement information of the terminal, the location information of the terminal, etc.; wherein, the target terminal can be The terminal that needs to be located.
  • the first communication device determines at least one of the target AI model, validity information of AI model related information, or feedback information obtained by positioning based on the target AI model.
  • the first communication device obtains the second information of the target terminal; the first communication device determines the target information according to the first information and the second information, and the target information includes at least one of the following: target AI model, AI model related information The validity information of the target terminal or the feedback information obtained by positioning based on the target AI model, because the first information is used to indicate the effective scope of application of the relevant information of the AI model, and the second information indicates the positioning related information acquired by the target terminal; based on the first information and the second information The second information determines the target information, which can make the positioning AI model more in line with the actual scene requirements, and can also update the AI model and/or AI model parameters based on the determined target information, so that the accuracy of the positioning result based on the updated AI model is higher.
  • target AI model AI model related information
  • the second information determines the target information, which can make the positioning AI model more in line with the actual scene requirements, and can also update the AI model and/or AI model parameters based on the determined target information, so that the accuracy of the positioning result based on
  • the above step "the first communication device determines the target information according to the first information and the second information” may include:
  • the first communication device determines the target information according to the value of the parameter in the second information and the range of the corresponding parameter in the first information.
  • the first communication device determines target information according to the first information, such as determining validity information, and feeds it back to the model management device, and the model management device performs model training based on the information fed back by the first communication device, renew.
  • the second communication device may be a model management device or a data management device or a validity function verification module (such as: Actor);
  • the second communication device is at least one of the following: model management device, LMF, NWDAF, and validity function verification module (such as: Actor).
  • the LMF and the model management device may be one device or different devices;
  • the LMF and model application modules are on one device or on different devices.
  • the terminal performs positioning based on an AI model, and determines validity information.
  • the first information includes at least one of the following: cell information; area information; effective time information; scene information;
  • the cell information includes at least one of the following:
  • TRP Transmission Reception Points
  • Cell frequency domain range information for example, frequency band identification ID
  • the area information includes at least one of the following:
  • the effective time information includes at least one of the following:
  • the scene information includes at least one of the following:
  • Line-of-sight LOS scene Non-line-of-sight NLOS scene; complex scene; indoor scene; outdoor scene.
  • the complex scene is, for example, a mixed scene of various scenes, etc.
  • the NLOS scene may also include a bad NLOS scene.
  • the distance range information is, for example, how much distance is valid from a certain reference point, for example, valid within 500m.
  • the second information includes at least one of the following: location information, cell information, area information, timer information, scene information, and SINR measured by the target terminal of the target terminal;
  • the cell information is at least one of the following information of the target terminal's serving cell, reference cell, or cell with the strongest RSRP, at least one of which includes: identification information and frequency domain information;
  • the area information is area identification information of the target terminal
  • the scene information is the scene information where the target terminal is located.
  • the location information is obtained through at least one of the following methods, and the at least one method includes:
  • OTDOA Observed Time Deviation Of Arrival
  • GNSS Global Navigation Satellite System
  • AoA Uplink Bluetooth Angle Of Arrival
  • AoD Bluetooth Departure Angle
  • Bluetooth Sensor or Wireless Fidelity WIFI.
  • the first communication device receives the first information and AI model-related information, and the first information is used to indicate the valid scope of application of the AI model-related information, such as valid cells, valid areas, valid time, valid scenarios and /or effective SINR range, etc.
  • the second information may be location information, cell information, area information, timer information, field information of the target terminal at least one of scene information and the SINR measured by the target terminal.
  • the target information is determined according to the value of the parameter in the second information and the range of the corresponding parameter in the first information. For example, compare the value of each parameter in the second information with the range of the corresponding parameter in the first information, if the value of the parameter in the second information is within the range of the corresponding parameter in the first information, it means that the AI model related information To be valid, all the parameters in the second information may be within the range of the corresponding parameters in the first information, or the values of most of the parameters may be within the range of the corresponding parameters in the first information, or the values of individual parameters may be If it is outside the range of the corresponding parameter in the first information, but the deviation from the range is not large, it can also be considered that the relevant information of the AI model is valid.
  • determining the validity information includes at least one of the following situations:
  • the AI model related information is valid.
  • the AI model-related information becomes invalid.
  • the relevant information of the AI model becomes invalid or expires.
  • the AI model-related information becomes invalid.
  • the AI model-related information becomes invalid.
  • the AI model-related information is valid, and the value of the parameter in the second information is within the range of the corresponding parameter in the first information. If it falls outside the scope, the relevant information of the AI model will be invalid, including the following situations, for example:
  • the cell information of the target terminal does not belong to the range of the cell information in the first information, and the relevant information of the AI model becomes invalid;
  • the area information of the target terminal does not belong to the area information in the first information, and the relevant information of the AI model becomes invalid;
  • the timer of the target terminal times out, and the relevant information of the AI model becomes invalid or times out;
  • the scene information of the target terminal does not belong to the scope of the scene information in the first information, and the relevant information of the AI model becomes invalid;
  • the location of the target terminal is not within the location range corresponding to the cell information and/or area information in the first information, and the AI model-related information becomes invalid;
  • the SINR measured by the target terminal is not within the SINR range in the first information, and the relevant information of the AI model becomes invalid.
  • the timer of the target terminal satisfies at least one of the following conditions:
  • the duration of the timer is the duration of the timer in the first message
  • the timer starts counting from the timer start time in the first message
  • the timer is re-started.
  • the feedback information is determined according to the validity information, including the following situations:
  • the feedback information includes at least the output of the AI model; or,
  • the feedback information includes at least one of the following: error cause; second information; validity information; AI model request; AI model update request; data collection request.
  • the content included in the feedback information can be determined based on the validity information, for example, when the AI model-related information is valid, the feedback information includes at least the output of the AI model; or, when the AI model-related information is invalid , the feedback information may include at least one of the following: error cause; second information; validity information; AI model request; AI model update request; data collection request, which can be used to train, select, and update the AI model.
  • the feedback information includes at least one of the following situations:
  • the feedback information includes the second information
  • the feedback information includes the second information.
  • the feedback information includes the cell information of the target terminal
  • the feedback information includes the area information of the target terminal.
  • the feedback information includes the scene information of the target terminal
  • the feedback information includes the location information of the target terminal
  • the feedback information includes the SINR measured by the target terminal
  • the feedback information includes timer information of the target terminal, such as timer duration, start time, and the like.
  • the target AI model is the AI model corresponding to the first range of the first information.
  • the feedback information includes: a target AI model.
  • different valid application ranges may correspond to different AI models and/or AI model parameters.
  • the target AI model may be the AI model corresponding to the first range.
  • the method also includes:
  • the first communication device receives multiple preconfigured AI models and/or AI model parameters, and first information corresponding to the AI models and/or AI model parameters.
  • the method also includes:
  • the first communication device acquires the AI model corresponding to the first range of the first information from multiple preconfigured AI models according to the first range of the first information.
  • the feedback information includes at least one of the following: validity information, second information, first measurement information, and an output of the AI model;
  • the first measurement information includes at least one of the following:
  • Signal measurement information position information; error value; channel impulse response (Channel Impulse Response, CIR) information; power delay profile (Power-Delay Profile, PDP) information.
  • CIR Channel Impulse Response
  • PDP Power-Delay Profile
  • the error value includes at least one of the following: a position error value, and a measurement error value.
  • the CIR information includes, for example, time-domain or frequency-domain impulse response information, or processing information (such as truncation information) of the time-domain or frequency-domain impulse response information, and may include single-antenna or multi-antenna CIR information.
  • the first communication device determines the target information according to the first information and the second information, and the target information includes at least one of the following: the target AI model, the validity information of the AI model related information, or the target AI model based on the positioning.
  • the target information can be determined, so that the positioned AI model can be more in line with the actual scene requirements, and the AI model and/or can be updated based on the determined target information Or AI model parameters, so that the accuracy of positioning results based on the updated AI model is higher.
  • the second information includes the second measurement information obtained by the target terminal
  • the step "the first communication device determines the target information according to the first information and the second information" may also be implemented through the following steps:
  • the first communication device determines the target information according to the value of the parameter in the second measurement information and the range of the corresponding parameter in the first information; it is worth noting that, in In one case, the one measurement can be understood as a smoothed result of M (M can be at least one of 1, 2, 3, and 4) measurement instances.
  • the first communication device determines the target information according to the consistency between the distribution of the parameters in the second measurement information and the distribution of the corresponding parameters in the first information.
  • the first communication device receives the first information and AI model-related information, and the first information is used to indicate the effective scope of application of the AI model-related information. In this embodiment, it may specifically indicate where the input data of the AI model-related information is obtained. Effective scope of application, such as SINR range, mean and variance of noise, mean and variance of absolute time, mean and variance of delay extension, mean and variance of angle extension, etc. In other words, AI model-related information is only available to Input data within the valid applicable range above is valid.
  • the relevant information of the AI model can specifically indicate the effective application range of the output data for obtaining the relevant information of the AI model, such as the SINR range, the mean value and variance of noise, the mean value and variance of absolute time (absolute time), and the mean value and variance of delay extension , the mean and variance of the angle expansion, etc., such as measurement information and position
  • the relevant information of the AI model is only valid for the output data within the above effective scope of application. For example, if the output measurement information and location information exceed the valid application range, it will be invalid.
  • the second information may be the second measurement information obtained by the measurement of the target terminal, including the SINR range, noise value, absolute time value introduced by NLOS, delay extension value, angle extension value, SINR mean and variance, noise mean and At least one of the variance, the absolute time mean and variance introduced by NLOS, the time delay spread mean and variance, and the angle spread mean and variance.
  • the first communication device determines the target information according to the value of the parameter in the second measurement information and the range of the corresponding parameter in the first information. For example, compare the value of each parameter in the second measurement information with the range of the corresponding parameter in the first information, if the value of the parameter in the second measurement information is within the range of the corresponding parameter in the first information, it means that the AI model If the relevant information is valid, it may be that all the parameters in the second measurement information are within the range of the corresponding parameters in the first information, or that the values of most of the parameters are within the range of the corresponding parameters in the first information, or that the values of individual parameters are within the range of the corresponding parameters in the first information. Even if the value is outside the range of the corresponding parameter in the first information, but the deviation from the range is not large, it can be considered that the relevant information of the AI model is valid.
  • the absolute time at C1, C2, and C3 is inconsistent with the range of this parameter in the first information, and C2 is closer to the range of this parameter in the first information. Therefore, it can be considered that the AI model corresponding to C2 is related The information is valid, but the corresponding ones of C2 and C3 are invalid. Therefore, the validity of C1, C2 and C3 is different, and the errors of C1, C2 and C3 are inconsistent.
  • the failure can be understood as a description of the degree of validity, for example, the degree of deviation between C2 and the first information can be expressed as validity information. Further, in one embodiment, it can be expressed as C2-average value/validity range.
  • the first communication device determines the target information according to the consistency between the distribution of the parameters in the second measurement information and the distribution of the corresponding parameters in the first information. For example, compare the distribution of each parameter in the second measurement information with the distribution of the corresponding parameter in the first information, if the distribution of the parameter in the second measurement information is consistent with the distribution of the corresponding parameter in the first information, it means that the AI model is related If the information is valid, it may be that the distribution of all parameters in the second measurement information is consistent with the distribution of the corresponding parameters in the first information, or that the distribution of most of the parameters is consistent with the distribution of the corresponding parameters in the first information, or that individual Even if the distribution of the parameters is inconsistent with the distribution of the corresponding parameters in the first information, but the deviation from the distribution of the corresponding parameters in the first information is not large, it can be considered that the relevant information of the AI model is valid.
  • the distribution of a certain parameter in the second measurement information is too different from the distribution of the corresponding parameter in the first information, and only a small part of the intersection, so it can be considered that the relevant information of the AI model is invalid.
  • the failure can be understood as a description of the degree of validity, for example, the degree of deviation between the distribution of a parameter in the second measurement information and the first information can be expressed as validity information.
  • the target information is determined based on the value of the parameter in the second measurement information and the range of the corresponding parameter in the first information, and/or, based on the distribution of the parameter in the second measurement information and the distribution of the corresponding parameter in the first information Determining the target information can make the positioning AI model more in line with the actual scene requirements, and can also update the AI model and/or AI model parameters based on the determined target information, so that the accuracy of the positioning result based on the updated AI model is higher.
  • the first communication device determines target information according to the first information and the second measurement information, such as determining validity information, and feeds it back to the model management device (such as a network side device),
  • the model management device performs model training and update based on the information fed back by the first communication device.
  • the first communication device may be an LMF
  • the model management device and the LMF may be the same device or different devices.
  • the LMF and model application modules can be on one device or on different devices.
  • the first information includes at least one of the following:
  • the range of the delay spread or the mean and/or variance of the delay spread are the range of the delay spread or the mean and/or variance of the delay spread
  • the absolute time introduced by NLOS is the time information obtained by the AI model minus the time information obtained by the non-AI model;
  • the time delay spread, angle spread, and noise distribution are feature information acquired according to the first measurement information or the output of the AI model.
  • the time information includes, for example, RSTD, TOA, Rx-Tx, timer information, and the like.
  • the second measurement information includes at least one of the following:
  • SINR range noise value, absolute time value introduced by NLOS, delay spread value, angle spread value, SINR mean and variance, noise mean and variance, absolute time mean and variance introduced by NLOS, delay spread mean and Variance, Angular Spread Mean and Variance;
  • the SINR range and the SINR mean and variance are obtained according to the SINR of at least one of the following information
  • the noise value and the noise mean and variance are obtained from the noise value of at least one of the following pieces of information; the at least one piece of information includes: measurement channel, measurement signal or first measurement information.
  • the first measurement information includes at least one of the following:
  • Signal measurement information position information; error value; channel impulse response CIR information; power delay profile PDP information.
  • the error value includes at least one of the following: a position error value, and a measurement error value.
  • the signal measurement information includes at least one of the following:
  • RSTD Reference Signal Time Difference
  • RSRP Reference Signal Received Power
  • the multipath measurement information includes at least one of the following:
  • First path power First path delay, first path time of arrival TOA, first path reference signal time difference RSTD, first path antenna subcarrier phase difference, first path antenna subcarrier phase, multipath power, multipath delay, TOA, multipath RSTD, multipath antenna subcarrier phase difference or multipath antenna subcarrier phase.
  • determining the validity information includes at least one of the following situations:
  • the AI model-related information becomes invalid.
  • the relevant information of the AI model is valid
  • the AI model-related information becomes invalid.
  • the value of the parameter in the second measurement information is compared with the range of the corresponding parameter in the first information. If the value of the parameter in the second measurement information is within the range of the corresponding parameter in the first information, it means that the AI model is related. If the information is valid, it may be that all the parameters in the second measurement information are within the range of the corresponding parameters in the first information, or that the values of most of the parameters are within the range of the corresponding parameters in the first information, or that the values of individual parameters are within the range of the corresponding parameters in the first information. Even if the value is outside the range of the corresponding parameter in the first information, but the deviation from the range is not large, the information related to the AI model can also be considered valid. or,
  • the AI model Related information is invalid.
  • the distribution of at least one parameter in the second measurement information is inconsistent with the distribution of the corresponding parameter in the first information, Or the distribution of all parameters in the second measurement information is inconsistent with the distribution of corresponding parameters in the first information, indicating that the relevant information of the AI model is invalid.
  • AI model-related information is invalid.
  • condition threshold information includes at least one of the following:
  • the ⁇ principle the probability that the value is distributed in ( ⁇ - ⁇ , ⁇ + ⁇ ) is 0.6526;
  • the 2 ⁇ principle the probability that the value is distributed in ( ⁇ -2 ⁇ , ⁇ +2 ⁇ ) is 0.9544;
  • the 3 ⁇ principle the value is distributed in The probability in ( ⁇ -3 ⁇ , ⁇ +3 ⁇ ) is 0.9974;
  • stands for the standard deviation in a normal distribution and ⁇ stands for the mean.
  • the feedback information includes at least one of the following: validity information, second measurement information, AI model input, AI model output, AI model identification information, and AI model update request.
  • the AI model update request includes at least one of the following:
  • the first information is obtained according to the test set and verification set of the AI model, including at least one of the following situations:
  • the first information is the characteristic information obtained according to the input data of the test set and the verification set of the AI model
  • the first information is the characteristic information obtained according to the output data of the test set and the verification set of the AI model
  • the first information is feature information obtained from input data and output data of a test set and a verification set of the AI model.
  • test set may be the second information
  • verification set may be the first information
  • the second measurement information is characteristic information obtained according to the first measurement information.
  • the second measurement information is feature information obtained from the output of the AI model.
  • the second measurement information is feature information obtained according to the first measurement information and the output of the AI model.
  • the first communication device may send target information to the second communication device, for example, send feedback information to the second communication device, and the second communication device may analyze the AI model and/or the AI model according to the feedback information. Type parameters are updated.
  • the method also includes:
  • the first communication device receives the updated AI model and/or AI model parameters, and first information corresponding to the AI model and/or AI model parameters.
  • the AI model includes at least one of the following:
  • the AI model type includes, for example: CNN, unsupervised, semi-supervised, supervised, RNN, LSTM, the first communication device performs positioning based on AI, or the first communication device only needs to perform some operations when performing positioning based on AI.
  • the AI model structure for example, includes:
  • a list of neural networks including at least one of the following: neuron types for each neural network, neuron weights and biases for each neural network;
  • Neural networks such as fully connected neural networks, convolutional neural networks, recurrent neural networks, residual networks, etc.;
  • the network structure of the AI model can also be a combination of multiple small networks, such as full connection + convolution, convolution + residual, etc.;
  • the network structure of the AI model can also include: the number of hidden layers; the connection method between the input layer and the hidden layer, the connection method between multiple hidden layers, the connection method between the hidden layer and the output layer, the number of neurons in each layer, etc. .
  • the AI model includes at least one of the following:
  • a list of neural networks including at least one of the following: neuron types for each neural network, neuron weights and biases for each neural network;
  • the hyperparameter information is, for example, the external parameters of the AI model, such as the pooling size (size), Batch size (representing the number of samples selected for one training), iteration mode, learning rate, optimization function selection, loss function selection, etc. .
  • the external parameters of the AI model such as the pooling size (size), Batch size (representing the number of samples selected for one training), iteration mode, learning rate, optimization function selection, loss function selection, etc.
  • the AI parameters include at least one of the following:
  • the initial parameters of the AI model are the initial parameters of the AI model.
  • the AI model describes parameter information such as the input format of the AI model parameters, the output format of the AI model parameters, and the like.
  • the initial parameters of the AI model are the initial parameters used to iterate the relevant information of the target AI model.
  • Fig. 12 is the second schematic diagram of the interaction process of the positioning method based on the AI model provided by the embodiment of the present application. As shown in Figure 12, the method provided in this embodiment includes:
  • Step 201 the second communication device receives the target information sent by the first communication device, and the target information includes at least one of the following: target AI model, validity information of the AI model-related information, or positioning based on the target AI model received feedback;
  • the target information is determined according to the first information associated with the AI model-related information, the first information is used to indicate the effective scope of application of the AI model-related information, and the AI model-related information includes at least one of the following: AI Model, AI model parameters, AI model input, AI model output.
  • the method also includes:
  • the second communication device sends the AI model related information and first information associated with the AI model related information to the first communication device.
  • the method also includes:
  • the second communication device sends multiple pre-configured AI models and/or AI model parameters, and first information corresponding to the AI models and/or AI model parameters to the first communication device.
  • the method also includes:
  • the second communication device sends second information to the first communication device, where the second information is used to represent the location-related information acquired by the terminal.
  • the first information includes at least one of the following: cell information; area information; effective time information; scene information;
  • the cell information includes at least one of the following:
  • the area information includes at least one of the following:
  • the effective time information includes at least one of the following:
  • the scene information includes at least one of the following:
  • Line-of-sight LOS scene Non-line-of-sight NLOS scene; complex scene; indoor scene; outdoor scene.
  • the second information includes at least one of the following: location information, cell information, area information, timer information, scene information, and SINR measured by the target terminal of the target terminal;
  • the cell information is the following at least one item of information of the target terminal's serving cell, reference cell, or cell with the strongest RSRP, the at least one item of information includes: identification information, frequency domain information;
  • the area information is area identification information of the target terminal
  • the scene information is the scene information where the target terminal is located.
  • the AI model related information is valid.
  • the information related to the AI model becomes invalid.
  • the feedback information includes at least one of the following: the validity information, the second information, the first measurement information, and the output of the AI model;
  • the first measurement information includes at least one of the following:
  • Signal measurement information position information; error value; channel impulse response CIR information; power delay profile PDP information.
  • the second information includes second measurement information obtained by the target terminal, and when the value of the parameter in the second measurement information is within the range of the corresponding parameter in the first information, the information about the above AI model is valid; and/or;
  • the AI model-related information becomes invalid.
  • the second information includes second measurement information obtained by the target terminal, and when the distribution of parameters in the second measurement information is consistent with the distribution of corresponding parameters in the first information, the Information about the AI model is valid; and/or;
  • the AI model-related information becomes invalid.
  • the AI model-related information is invalid, including:
  • the AI model-related information becomes invalid.
  • condition threshold information includes at least one of the following:
  • the second measurement information includes at least one of the following:
  • SINR range noise value, absolute time value introduced by NLOS, delay spread value, angle spread value, SINR mean and variance, noise mean and variance, absolute time mean and variance introduced by NLOS, delay spread mean and Variance, Angular Spread Mean and Variance;
  • the SINR range and the SINR mean and variance are obtained according to the SINR of at least one of the following information
  • the noise value and the noise mean and variance are obtained from the noise value of at least one of the following information;
  • the at least one piece of information includes: a measurement channel, a measurement signal, or first measurement information.
  • the first measurement information includes at least one of the following:
  • Signal measurement information position information; error value; channel impulse response CIR information; power delay profile PDP information.
  • the feedback information includes at least one of the following: the validity information, second measurement information, AI model input, AI model output, AI model identification information, and AI model update request.
  • the input of the AI model can be the current input of the AI model, or the input or input distribution determined based on multiple inputs, so it is hoped that the peer end can provide an AI model and parameters that satisfy the input
  • the output of the AI model can be the current output of the AI model, or the output or output distribution determined according to multiple inputs, so it is hoped that the peer end can provide an AI model and parameters that satisfy the output
  • the input and output of the AI model can be the current input and output of the AI model, or the input and output or input and output distribution determined according to multiple inputs and outputs, so it is hoped that the peer end can provide Input and output AI models and parameters
  • the second measurement information may be the second measurement information obtained by one measurement, or may be the distribution of the second measurement information determined by multiple inputs.
  • the first information includes at least one of the following:
  • the range of the delay spread or the mean and/or variance of the delay spread are the range of the delay spread or the mean and/or variance of the delay spread
  • the second communication device includes at least one of the following:
  • the executing subject may be a positioning device based on the AI model.
  • the AI model-based positioning device provided in the embodiment of the present application is described by taking the AI model-based positioning device executing the AI model-based positioning method as an example.
  • Fig. 13 is one of the structural schematic diagrams of the positioning device based on the AI model provided by the present application. As shown in Figure 13, the positioning device based on the AI model provided in this embodiment includes:
  • An acquisition module 210 configured to acquire first information associated with information related to the AI model
  • the processing module 220 is configured to determine target information according to the first information, and the target information includes at least one of the following: a target AI model, validity information of information related to the AI model, or positioning based on the target AI model received feedback;
  • the first information is used to indicate the effective scope of application of the AI model-related information
  • the AI model-related information includes at least one of the following: AI model, AI model parameters, AI model input, and AI model output.
  • the acquisition module 210 is also configured to:
  • the second information is used to represent the location-related information acquired by the target terminal;
  • the processing module 220 is specifically used for:
  • the target information is determined according to the first information and the second information.
  • the first information includes at least one of the following: cell information; area information; effective time information; scene information;
  • the cell information includes at least one of the following:
  • the area information includes at least one of the following:
  • the effective time information includes at least one of the following:
  • the scene information includes at least one of the following:
  • Line-of-sight LOS scene Non-line-of-sight NLOS scene; complex scene; indoor scene; outdoor scene.
  • the second information includes at least one of the following: location information, cell information, area information, timer information, scene information, and SINR measured by the target terminal of the target terminal;
  • the cell information is the following at least one item of information of the target terminal's serving cell, reference cell, or cell with the strongest RSRP, the at least one item of information includes: identification information, frequency domain information;
  • the area information is area identification information of the target terminal
  • the scene information is the scene information where the target terminal is located.
  • processing module 220 is specifically configured to:
  • the first communication device determines the target information according to the value of the parameter in the second information and the range of the corresponding parameter in the first information.
  • the AI model related information is valid.
  • the feedback information includes the second information
  • the feedback information includes the second information.
  • the AI model-related information becomes invalid.
  • the feedback information includes the cell information of the target terminal.
  • the AI model-related information becomes invalid or expires.
  • the timer satisfies at least one of the following conditions:
  • the duration of the timer is the duration of the timer in the first message
  • the timer starts counting from the timer start time in the first message
  • the timer is restarted.
  • the AI model-related information becomes invalid.
  • the AI model-related information becomes invalid.
  • the target AI model is the one corresponding to the first range of the first information. AI model.
  • the obtaining module 310 is also used for:
  • the obtaining module 310 is also used for:
  • the first communication device acquires an AI model corresponding to the first range of the first information from multiple preconfigured AI models according to the first range of the first information.
  • the feedback information includes: the target AI model.
  • the feedback information is determined according to the validity information, including:
  • the feedback information includes at least the output of the AI model
  • the feedback information includes at least one of the following: error cause; second information; validity information; AI model request; AI model update request; data collection request .
  • the feedback information includes at least one of the following: the validity information, the second information, the first measurement information, and the output of the AI model;
  • the first measurement information includes at least one of the following:
  • Signal measurement information position information; error value; channel impulse response CIR information; power delay profile PDP information.
  • the second information includes second measurement information obtained by the target terminal, and the processing module 220 is specifically configured to:
  • the first communication device determines the value of the parameter in the second measurement information and the range of the corresponding parameter in the first information. target information;
  • the first communication device In the case where the second measurement information is measurement information obtained by multiple measurements, the first communication device, according to the consistency between the distribution of parameters in the second measurement information and the distribution of corresponding parameters in the first information, , to determine the target information.
  • the AI model related information is valid
  • the AI model-related information becomes invalid.
  • the AI model related information is valid
  • the distribution of at least one parameter in the second measurement information is the same as the distribution of the corresponding parameter in the first information In the case of inconsistency, the relevant information of the AI model becomes invalid.
  • the AI model-related information is invalid, including:
  • the AI model-related information becomes invalid.
  • condition threshold information includes at least one of the following:
  • the second measurement information includes at least one of the following:
  • SINR range noise value, absolute time value introduced by NLOS, delay spread value, angle spread value, SINR mean and variance, noise mean and variance, absolute time mean and variance introduced by NLOS, delay spread mean and Variance, Angular Spread Mean and Variance;
  • the SINR range and the SINR mean and variance are obtained according to the SINR of at least one of the following information
  • the noise value and the noise mean and variance are obtained from the noise value of at least one of the following information;
  • the at least one piece of information includes: a measurement channel, a measurement signal, or first measurement information.
  • the first measurement information includes at least one of the following:
  • Signal measurement information position information; error value; channel impulse response CIR information; power delay profile PDP information.
  • the signal measurement information includes at least one of the following:
  • Reference signal time difference RSTD measurement results round-trip delay measurement results, angle of arrival AOA measurement results, angle of departure AOD measurement results, reference information received power RSRP, multipath measurement information, line-of-sight LOS indication information;
  • the multipath measurement information includes at least one of the following:
  • First path power First path delay, first path time of arrival TOA, first path reference signal time difference RSTD, first path antenna subcarrier phase difference, first path antenna subcarrier phase, multipath power, multipath delay, TOA, multipath RSTD, multipath antenna subcarrier phase difference or multipath antenna subcarrier phase.
  • the second measurement information further includes the first measurement information.
  • the feedback information includes at least one of the following: the validity information, second measurement information, AI model input, AI model output, AI model identification information, and AI model update request.
  • the first information includes at least one of the following:
  • the range of the delay spread or the mean and/or variance of the delay spread are the range of the delay spread or the mean and/or variance of the delay spread
  • the first information is obtained according to the test set and verification set of the AI model, including at least one of the following situations:
  • the first information is feature information obtained according to the input data of the test set and the verification set of the AI model
  • the first information is feature information obtained according to the output data of the test set and verification set of the AI model
  • the first information is feature information acquired according to input data and output data of a test set and a verification set of the AI model.
  • the second measurement information is characteristic information obtained according to the first measurement information.
  • the second measurement information is feature information obtained from the output of the AI model.
  • the absolute time introduced by the NLOS is the time information obtained by the AI model minus the time information obtained by the non-AI model;
  • the time delay spread, angle spread, and noise distribution are feature information acquired according to the first measurement information or the output of the AI model.
  • the input of the AI model and/or the output of the AI model include at least one of the following:
  • the error information includes at least one of the following: position error value, measurement error value, AI model error value or AI model parameter error value.
  • the AI model update request includes at least one of the following:
  • the AI model includes at least one of the following:
  • the AI model includes at least one of the following:
  • a list of neural networks including at least one of the following: neuron types for each neural network, neuron weights and biases for each neural network;
  • the AI parameters include at least one of the following:
  • the initial parameters of the AI model are the initial parameters of the AI model.
  • the validity information may include at least one of the following:
  • Validity indication information used to indicate whether the relevant information of the AI model is valid
  • Reliability indication information used to indicate whether the positioning result obtained based on the target AI model is reliable
  • the acquiring module 310 is also used for:
  • the location information is obtained through at least one of the following methods, the at least one method including:
  • Time difference of arrival positioning method OTDOA, global navigation satellite system GNSS, downlink time difference of arrival, uplink time difference of arrival, uplink Bluetooth angle of arrival AoA, Bluetooth angle of departure AoD, Bluetooth, sensor or wireless fidelity WIFI.
  • the first communication device includes at least one of the following:
  • a location management function LMF network element and an evolution device of the LMF network element A location management function LMF network element and an evolution device of the LMF network element;
  • NWADF Network data analysis function
  • the device in this embodiment can be used to execute the method in any one of the foregoing terminal-side method embodiments, and its specific implementation process and technical effect are similar to those in the terminal-side method embodiment.
  • the terminal-side method embodiment please refer to the terminal-side method embodiment. A detailed introduction will not be repeated here.
  • FIG. 14 is the second structural schematic diagram of the positioning device based on the AI model provided by the present application. As shown in Figure 14, the positioning device based on the AI model provided in this embodiment includes:
  • the receiving module 310 is configured to receive target information sent by the first communication device, where the target information includes at least one of the following: a target AI model, validity information of information related to the AI model, or positioning based on the target AI model feedback information;
  • the target information is determined according to the first information associated with the AI model-related information, the first information is used to indicate the effective scope of application of the AI model-related information, and the AI model-related information includes at least one of the following: AI Model, AI model parameters, AI model input, AI model output.
  • a sending module 320 configured to send the AI model related information and first information associated with the AI model related information to the first communication device.
  • the sending module 320 is also used for:
  • the sending module 320 is also used for:
  • the first information includes at least one of the following: cell information; area information; effective time information; scene information;
  • the cell information includes at least one of the following:
  • the area information includes at least one of the following:
  • the effective time information includes at least one of the following:
  • the scene information includes at least one of the following:
  • Line-of-sight LOS scene Non-line-of-sight NLOS scene; complex scene; indoor scene; outdoor scene.
  • the second information includes at least one of the following: location information, cell information, area information, timer information, scene information, and SINR measured by the target terminal of the target terminal;
  • the cell information is the following at least one item of information of the target terminal's serving cell, reference cell, or cell with the strongest RSRP, the at least one item of information includes: identification information, frequency domain information;
  • the area information is area identification information of the target terminal
  • the scene information is the scene information where the target terminal is located.
  • the AI model related information is valid.
  • the information related to the AI model becomes invalid.
  • the feedback information includes at least one of the following: the validity information, the second information, the first measurement information, and the output of the AI model;
  • the first measurement information includes at least one of the following:
  • Signal measurement information position information; error value; channel impulse response CIR information; power delay profile PDP information.
  • the second information includes second measurement information obtained by the target terminal, and when the value of the parameter in the second measurement information is within the range of the corresponding parameter in the first information, the information about the above AI model is valid; and/or;
  • the AI model-related information becomes invalid.
  • the second information includes second measurement information obtained by the target terminal, and when the distribution of parameters in the second measurement information is consistent with the distribution of corresponding parameters in the first information, the Information about the AI model is available; and/or;
  • the AI model-related information becomes invalid.
  • the AI model-related information is invalid, including:
  • the AI model-related information becomes invalid.
  • condition threshold information includes at least one of the following:
  • the second measurement information includes at least one of the following:
  • SINR range noise value, absolute time value introduced by NLOS, delay spread value, angle spread value, SINR mean and variance, noise mean and variance, absolute time mean and variance introduced by NLOS, delay spread mean and Variance, Angular Spread Mean and Variance;
  • the SINR range and the SINR mean and variance are obtained according to the SINR of at least one of the following information
  • the noise value and the noise mean and variance are obtained from the noise value of at least one of the following information;
  • the at least one piece of information includes: a measurement channel, a measurement signal, or first measurement information.
  • the first measurement information includes at least one of the following:
  • Signal measurement information position information; error value; channel impulse response CIR information; power delay profile PDP information.
  • the feedback information includes at least one of the following: the validity information, second measurement information, AI model input, AI model output, AI model identification information, and AI model update request.
  • the first information includes at least one of the following:
  • the range of the delay spread or the mean and/or variance of the delay spread are the range of the delay spread or the mean and/or variance of the delay spread
  • the second communication device includes at least one of the following:
  • the device in this embodiment can be used to execute the method in any of the aforementioned network-side method embodiments, and its specific implementation process and technical effects are similar to those in the network-side method embodiments. For details, please refer to the network-side method embodiments. A detailed introduction will not be repeated here.
  • the positioning apparatus based on the AI model in the embodiment of the present application may be an electronic device, such as an electronic device with an operating system, or a component in the electronic device, such as an integrated circuit or a chip.
  • the electronic device may be a terminal, or other devices other than the terminal.
  • the terminal may include, but not limited to, the types of terminal 11 listed above, and other devices may be servers, Network Attached Storage (NAS), etc., which are not specifically limited in this embodiment of the present application.
  • NAS Network Attached Storage
  • the positioning device based on the AI model provided by the embodiment of the present application can realize each process realized by the method embodiments in FIG. 2 to FIG. 12 and achieve the same technical effect. To avoid repetition, details are not repeated here.
  • this embodiment of the present application also provides a communication device 1500, including a processor 1501 and a memory 1502, and the memory 1502 stores programs or instructions that can run on the processor 1501, such as
  • the communication device 1500 is a terminal
  • the program or instruction is executed by the processor 1501
  • each step of the above embodiment of the positioning method based on the AI model can be realized, and the same technical effect can be achieved.
  • the communication device 1500 is a network-side device, when the program or instruction is executed by the processor 1501, each step of the above-mentioned embodiment of the positioning method based on the AI model can be achieved, and the same technical effect can be achieved. To avoid repetition, details are not repeated here. .
  • the embodiment of the present application also provides a terminal, including a processor and a communication interface, where the communication interface is used to obtain first information associated with information related to the AI model; the processor is used to determine target information according to the first information,
  • the target information includes at least one of the following: a target AI model, validity information of information related to the AI model, or feedback information obtained by positioning based on the target AI model; the first information is used to represent the AI model
  • the AI model related information includes at least one of the following: AI model, AI model parameters, AI model input, and AI model output.
  • FIG. 16 is a schematic diagram of a hardware structure of a terminal implementing an embodiment of the present application.
  • the terminal 1000 includes but not limited to: a radio frequency unit 1001, a network module 1002, an audio output unit 1003, an input unit 1004, a sensor 1005, a display unit 1006, a user input unit 1007, an interface unit 1008, a memory 1009, and a processor 1010, etc. at least some of the components.
  • the terminal 1000 can also include a power supply (such as a battery) for supplying power to various components, and the power supply can be logically connected to the processor 1010 through the power management system, so as to manage charging, discharging, and power consumption through the power management system. Management and other functions.
  • a power supply such as a battery
  • the terminal structure shown in FIG. 16 does not constitute a limitation on the terminal.
  • the terminal may include more or less components than shown in the figure, or combine certain components, or arrange different components, which will not be repeated here.
  • the input unit 1004 may include a graphics processing unit (Graphics Processing Unit, GPU) 10041 and a microphone 10042, and the graphics processor 10041 is used in video capture mode or image capture mode by an image capture device (such as the image data of the still picture or video obtained by the camera) for processing.
  • the display unit 1006 may include a display panel 10061, and may adopt a liquid crystal display
  • the display panel 10061 is configured in the form of a device, an organic light emitting diode, or the like.
  • the user input unit 1007 includes at least one of a touch panel 10071 and other input devices 10072 .
  • the touch panel 10071 is also called a touch screen.
  • the touch panel 10071 may include two parts, a touch detection device and a touch controller.
  • Other input devices 10072 may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, switch buttons, etc.), trackballs, mice, and joysticks, which will not be repeated here.
  • the radio frequency unit 1001 can transmit the downlink data received from the network side device to the processor 1010 for processing; in addition, the radio frequency unit 1001 can send the uplink data to the network side device to send the uplink data.
  • the radio frequency unit 1001 includes, but is not limited to, an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier, a duplexer, and the like.
  • the memory 1009 can be used to store software programs or instructions as well as various data.
  • the memory 1009 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage program or instruction area may store an operating system, an application program or instructions required by at least one function (such as a voice playback function, image playback function, etc.), etc.
  • memory 1009 may include volatile memory or nonvolatile memory, or, memory 1009 may include both volatile and nonvolatile memory.
  • Non-volatile memory can also include non-volatile memory, wherein, the non-volatile memory can be read-only memory (Read-Only Memory, ROM), programmable read-only memory (Programmable ROM, PROM), programmable Erasable programmable read-only memory (Erasable PROM, EPROM), electrically erasable programmable read-only memory (Electrically EPROM, EEPROM) or flash memory.
  • ROM Read-Only Memory
  • PROM programmable read-only memory
  • Erasable PROM Erasable PROM
  • EPROM programmable Erasable programmable read-only memory
  • Electrically erasable programmable read-only memory Electrically erasable programmable read-only memory
  • EEPROM electrically erasable programmable read-only memory
  • Volatile memory can be random access memory (Random Access Memory, RAM), static random access memory (Static RAM, SRAM), dynamic random access memory (Dynamic RAM, DRAM), synchronous dynamic random access memory (Synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (Double Data Rate SDRAM, DDRSDRAM), enhanced synchronous dynamic random access memory (Enhanced SDRAM, ESDRAM), synchronous connection dynamic random access memory (Synch link DRAM , SLDRAM) and Direct Memory Bus Random Access Memory (Direct Rambus RAM, DRRAM).
  • RAM Random Access Memory
  • SRAM static random access memory
  • DRAM dynamic random access memory
  • DRAM synchronous dynamic random access memory
  • SDRAM double data rate synchronous dynamic random access memory
  • Double Data Rate SDRAM Double Data Rate SDRAM
  • DDRSDRAM double data rate synchronous dynamic random access memory
  • Enhanced SDRAM, ESDRAM enhanced synchronous dynamic random access memory
  • Synch link DRAM , SLDRAM
  • Direct Memory Bus Random Access Memory Direct Rambus
  • the processor 1010 may include one or more processing units; optionally, the processor 1010 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, application programs or instructions, etc.
  • modem processors mainly process wireless communication signals, such as baseband processors. It can be understood that the foregoing modem processor may not be integrated into the processor 1010 .
  • the radio frequency unit 1001 is configured to obtain first information associated with information related to the AI model
  • the processor 1010 is configured to determine target information according to the first information, where the target information includes at least one of the following: a target AI model, validity information of information related to the AI model, or positioning based on the target AI model received feedback;
  • the first information is used to indicate the effective scope of application of the relevant information of the AI model, and the relevant information of the AI model.
  • the relevant information includes at least one of the following: AI model, AI model parameters, AI model input, and AI model output.
  • the radio frequency unit obtains the first information associated with the AI model-related information; the processor determines the target information according to the first information, and the target information includes at least one of the following: target AI model, validity information of the AI model-related information Or based on the feedback information obtained by positioning based on the target AI model, since the first information is used to indicate the effective scope of application of the relevant information of the AI model, determining the target information based on the first information can make the positioned AI model more in line with the actual scene requirements, and can also The AI model and/or AI model parameters are updated based on the determined target information, so that the accuracy of the positioning result obtained based on the updated AI model is higher.
  • the radio frequency unit 1001 is also used for:
  • the second information is used to represent the location-related information acquired by the target terminal;
  • the processor 1010 is specifically used for:
  • the target information is determined according to the first information and the second information.
  • the first information includes at least one of the following: cell information; area information; effective time information; scene information;
  • the cell information includes at least one of the following:
  • the area information includes at least one of the following:
  • the effective time information includes at least one of the following:
  • the scene information includes at least one of the following:
  • Line-of-sight LOS scene Non-line-of-sight NLOS scene; complex scene; indoor scene; outdoor scene.
  • the second information includes at least one of the following: location information, cell information, area information, timer information, scene information, and SINR measured by the target terminal of the target terminal;
  • the cell information is the following at least one item of information of the target terminal's serving cell, reference cell, or cell with the strongest RSRP, the at least one item of information includes: identification information, frequency domain information;
  • the area information is area identification information of the target terminal
  • the scene information is the scene information where the target terminal is located.
  • processor 1010 is specifically configured to:
  • the first communication device determines the target information according to the value of the parameter in the second information and the range of the corresponding parameter in the first information.
  • the AI model related information is valid.
  • the feedback information includes the second information
  • the feedback information includes the second information.
  • the AI model-related information becomes invalid.
  • the feedback information includes the cell information of the target terminal.
  • the AI model-related information becomes invalid or expires.
  • the timer satisfies at least one of the following conditions:
  • the duration of the timer is the duration of the timer in the first message
  • the timer starts counting from the timer start time in the first message
  • the timer is restarted.
  • the AI model-related information becomes invalid.
  • the AI model-related information becomes invalid.
  • the target AI model is the one corresponding to the first range of the first information. AI model.
  • the radio frequency unit 1001 is also used for:
  • the radio frequency unit 1001 is also used for:
  • the first communication device acquires an AI model corresponding to the first range of the first information from multiple preconfigured AI models according to the first range of the first information.
  • the feedback information includes: the target AI model.
  • the feedback information is determined according to the validity information, including:
  • the feedback information includes at least the output of the AI model
  • the feedback information includes at least one of the following: error cause; second information; validity information; AI model request; AI model update request; data collection request .
  • the feedback information includes at least one of the following: the validity information, the second information, the first measurement information, and the output of the AI model;
  • the first measurement information includes at least one of the following:
  • Signal measurement information position information; error value; channel impulse response CIR information; power delay profile PDP information.
  • the second information includes second measurement information obtained by the target terminal, and the processor 1010 is specifically configured to:
  • the first communication device determines the value of the parameter in the second measurement information and the range of the corresponding parameter in the first information. target information;
  • the first communication device In the case where the second measurement information is measurement information obtained by multiple measurements, the first communication device, according to the consistency between the distribution of parameters in the second measurement information and the distribution of corresponding parameters in the first information, , to determine the target information.
  • the AI model related information is valid
  • the AI model-related information becomes invalid.
  • the AI model related information is valid
  • the AI model-related information becomes invalid.
  • the AI model-related information is invalid, including:
  • the AI model-related information becomes invalid.
  • condition threshold information includes at least one of the following:
  • the second measurement information includes at least one of the following:
  • SINR range noise value, absolute time value introduced by NLOS, delay spread value, angle spread value, SINR mean and variance, noise mean and variance, absolute time mean and variance introduced by NLOS, delay spread mean and Variance, Angular Spread Mean and Variance;
  • the SINR range and the SINR mean and variance are obtained according to the SINR of at least one of the following information
  • the noise value and the noise mean and variance are obtained from the noise value of at least one of the following information;
  • the at least one piece of information includes: a measurement channel, a measurement signal, or first measurement information.
  • the first measurement information includes at least one of the following:
  • Signal measurement information position information; error value; channel impulse response CIR information; power delay profile PDP information.
  • the signal measurement information includes at least one of the following:
  • Reference signal time difference RSTD measurement results round-trip delay measurement results, angle of arrival AOA measurement results, angle of departure AOD measurement results, reference information received power RSRP, multipath measurement information, line-of-sight LOS indication information;
  • the multipath measurement information includes at least one of the following:
  • First path power First path delay, first path time of arrival TOA, first path reference signal time difference RSTD, first path antenna subcarrier phase difference, first path antenna subcarrier phase, multipath power, multipath delay, TOA, multipath RSTD, multipath antenna subcarrier phase difference or multipath antenna subcarrier phase.
  • the second measurement information further includes the first measurement information.
  • the feedback information includes at least one of the following: the validity information, second measurement information, AI model input, AI model output, AI model identification information, and AI model update request.
  • the first information includes at least one of the following:
  • the range of the delay spread or the mean and/or variance of the delay spread are the range of the delay spread or the mean and/or variance of the delay spread
  • the first information is obtained according to the test set and verification set of the AI model, including at least one of the following Cases:
  • the first information is feature information obtained according to the input data of the test set and the verification set of the AI model
  • the first information is feature information obtained according to the output data of the test set and verification set of the AI model
  • the first information is feature information acquired according to input data and output data of a test set and a verification set of the AI model.
  • the second measurement information is characteristic information obtained according to the first measurement information.
  • the second measurement information is feature information obtained from the output of the AI model.
  • the absolute time introduced by the NLOS is the time information obtained by the AI model minus the time information obtained by the non-AI model;
  • the time delay spread, angle spread, and noise distribution are feature information acquired according to the first measurement information or the output of the AI model.
  • the input of the AI model and/or the output of the AI model include at least one of the following:
  • the error information includes at least one of the following: position error value, measurement error value, AI model error value or AI model parameter error value.
  • the AI model update request includes at least one of the following:
  • the AI model includes at least one of the following:
  • the AI model includes at least one of the following:
  • a list of neural networks including at least one of the following: neuron types for each neural network, neuron weights and biases for each neural network;
  • the AI parameters include at least one of the following:
  • the initial parameters of the AI model are the initial parameters of the AI model.
  • the validity information may include at least one of the following:
  • Validity indication information used to indicate whether the relevant information of the AI model is valid
  • Reliability indication information used to indicate whether the positioning result obtained based on the target AI model is reliable
  • the radio frequency unit 1001 is also used for:
  • the location information is obtained through at least one of the following methods, the at least one method including:
  • Time difference of arrival positioning method OTDOA, global navigation satellite system GNSS, downlink time difference of arrival, uplink time difference of arrival, uplink Bluetooth angle of arrival AoA, Bluetooth angle of departure AoD, Bluetooth, sensor or wireless fidelity WIFI.
  • the first communication device includes at least one of the following:
  • a location management function LMF network element and an evolution device of the LMF network element A location management function LMF network element and an evolution device of the LMF network element;
  • NWADF Network data analysis function
  • the embodiment of the present application also provides a network side device, including a processor and a communication interface, the communication interface is used to obtain the first information associated with the information related to the AI model; the processor is used to determine the target information according to the first information , the target information includes at least one of the following: a target AI model, validity information of information related to the AI model, or feedback information obtained by positioning based on the target AI model; the first information is used to represent the AI Effective scope of application of model-related information, where the AI model-related information includes at least one of the following: AI model, AI model parameters, AI model input, and AI model output.
  • This network-side device embodiment corresponds to the above-mentioned first communication device-side or second communication-device-side method embodiment, and each implementation process and implementation method of the above-mentioned method embodiment can be applied to this network-side device embodiment, and can achieve Same technical effect.
  • the embodiment of the present application also provides a network side device.
  • the network side device 700 includes: an antenna 71 , a radio frequency device 72 , a baseband device 73 , a processor 75 and a memory 75 .
  • Antenna 71 with The radio frequency device 72 is connected.
  • the radio frequency device 72 receives information through the antenna 71, and sends the received information to the baseband device 73 for processing.
  • the baseband device 73 processes the information to be sent and sends it to the radio frequency device 72
  • the radio frequency device 72 processes the received information and sends it out through the antenna 71 .
  • the above frequency band processing device may be located in the baseband device 73 , and the method performed by the network side device in the above embodiment may be implemented in the baseband device 73 , and the baseband device 73 includes a baseband processor 75 and a memory 75 .
  • the baseband device 73 can include at least one baseband board, for example, a plurality of chips are arranged on the baseband board, as shown in FIG.
  • the program executes the network device operations shown in the above method embodiments.
  • the network side equipment of the baseband device 73 may also include a network interface 76 for exchanging information with the radio frequency device 72, and the interface is, for example, a common public radio interface (CPRI for short).
  • CPRI common public radio interface
  • the network-side device 700 in the embodiment of the present invention further includes: instructions or programs stored in the memory 75 and operable on the processor 75, and the processor 75 invokes the instructions or programs in the memory 75 to execute FIG. 13 or FIG. 14
  • the methods executed by each module shown in the figure achieve the same technical effect, so in order to avoid repetition, they are not repeated here.
  • the embodiment of the present application also provides a network side device.
  • the network side device 700 includes: a processor 701 , a network interface 702 and a memory 703 .
  • the network interface 702 is, for example, a common public radio interface (common public radio interface, CPRI).
  • the network-side device 700 in this embodiment of the present application further includes: instructions or programs stored in the memory 703 and operable on the processor 701, and the processor 701 calls the instructions or programs in the memory 703 to execute the instructions shown in FIG. 13 or 14.
  • the methods executed by each module are shown to achieve the same technical effect. In order to avoid repetition, the details are not repeated here.
  • first communication device and/or the second communication device may be implemented through the foregoing network side device embodiments.
  • the embodiment of the present application also provides a readable storage medium, where a program or instruction is stored on the readable storage medium, and when the program or instruction is executed by a processor, each process of the above embodiment of the positioning method based on the AI model is implemented, and The same technical effect can be achieved, so in order to avoid repetition, details will not be repeated here.
  • the processor is the processor in the terminal described in the foregoing embodiments.
  • the readable storage medium includes a computer-readable storage medium, such as a computer read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk, and the like.
  • the embodiment of the present application further provides a chip, the chip includes a processor and a communication interface, the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the above-mentioned positioning method based on the AI model
  • the various processes of the embodiment can achieve the same technical effect, so in order to avoid repetition, details are not repeated here.
  • the chip mentioned in the embodiment of the present application may also be called a system-on-chip, a system-on-chip, a system-on-a-chip, or a system-on-a-chip.
  • the embodiment of the present application further provides a computer program/program product, the computer program/program product is stored in a storage medium, and the computer program/program product is executed by at least one processor to realize the above-mentioned positioning based on the AI model
  • Each process of the method embodiment can achieve the same technical effect, and will not be repeated here to avoid repetition.
  • the embodiment of the present application also provides a communication system, including: a terminal and a network-side device, the terminal can be used to perform the steps of the above-mentioned AI model-based positioning method, and the network-side device can be used to perform the steps of the above-mentioned The steps of the AI model-based positioning method.
  • the methods of the above embodiments can be implemented by means of software plus a necessary general-purpose hardware platform, and of course also by hardware, but in many cases the former is better implementation.
  • the technical solution of the present application can be embodied in the form of computer software products, which are stored in a storage medium (such as ROM/RAM, magnetic disk, etc.) , CD-ROM), including several instructions to make a terminal (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) execute the methods described in the various embodiments of the present application.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Artificial Intelligence (AREA)
  • Mobile Radio Communication Systems (AREA)
  • Position Fixing By Use Of Radio Waves (AREA)

Abstract

本申请公开了一种基于人工智能AI模型的定位方法及通信设备,属于通信技术领域,本申请实施例的基于人工智能AI模型的定位方法包括:第一通信设备获取与AI模型相关信息关联的第一信息;所述第一通信设备根据所述第一信息,确定目标信息,所述目标信息包括以下至少一项:目标AI模型、所述AI模型相关信息的有效性信息或者基于所述目标AI模型进行定位得到的反馈信息;所述第一信息用于表示所述AI模型相关信息的有效适用范围,所述AI模型相关信息包括以下至少一项:AI模型、AI模型参数、AI模型的输入、AI模型的输出。

Description

基于人工智能AI模型的定位方法及通信设备
相关申请的交叉引用
本申请要求于2022年1月29日提交的申请号为202210113101.4,发明名称为“基于人工智能AI模型的定位方法及通信设备”的中国专利申请的优先权,其通过引用方式全部并入本申请。
技术领域
本申请属于通信技术领域,具体涉及一种基于人工智能AI模型的定位方法及通信设备。
背景技术
随着定位技术的逐步成熟,带有定位功能的终端设备越来越多。为了提高定位效率,可以采用人工智能(Artificial Intelligence,AI)模型获取定位数据,但是用于定位的测量数据和需求是会随着时间改变的,即随着时间的推移使用的AI模型可能不符合实际场景需求,导致定位结果不满足实际场景需求。因此,对于本领域技术人员来说,亟需实现一种符合实际场景需求的基于AI模型的定位方案。
发明内容
I模型进行定位的结果不满足实际场景需求的问题。
第一方面,提供了一种基于人工智能AI模型的定位方法,应用于第一通信设备,该方法包括:
第一通信设备获取与AI模型相关信息关联的第一信息;
所述第一通信设备根据所述第一信息,确定目标信息,所述目标信息包括以下至少一项:目标AI模型、所述AI模型相关信息的有效性信息或者基于所述目标AI模型进行定位得到的反馈信息;
所述第一信息用于表示所述AI模型相关信息的有效适用范围,所述AI模型相关信息包括以下至少一项:AI模型、AI模型参数、AI模型的输入、AI模型的输出。
第二方面,提供了一种基于AI模型的定位方法,应用于第二通信设备,该方法包括:
第二通信设备接收第一通信设备发送的目标信息,所述目标信息包括以下至少一项:目标AI模型、所述AI模型相关信息的有效性信息或者基于所述目标AI模型进行定位得到的反馈信息;
所述目标信息为根据AI模型相关信息关联的第一信息确定的,所述第一信息用于表示所述AI模型相关信息的有效适用范围,所述AI模型相关信息包括以下至少一项:AI模型、AI模型参数、AI模型的输入、AI模型的输出。
第三方面,提供了一种基于AI模型的定位装置,包括:
获取模块,用于获取与AI模型相关信息关联的第一信息;
处理模块,用于根据所述第一信息,确定目标信息,所述目标信息包括以下至少一项:目标AI模型、所述AI模型相关信息的有效性信息或者基于所述目标AI模型进行定位得到的反馈信息;
所述第一信息用于表示所述AI模型相关信息的有效适用范围,所述AI模型相关信息包括以下至少一项:AI模型、AI模型参数、AI模型的输入、AI模型的输出。
第四方面,提供了一种基于AI模型的定位装置,包括:
接收模块,用于接收第一通信设备发送的目标信息,所述目标信息包括以下至少一项:目标AI模型、所述AI模型相关信息的有效性信息或者基于所述目标AI模型进行定位得到的反馈信息;
所述目标信息为根据AI模型相关信息关联的第一信息确定的,所述第一信息用于表示所述AI模型相关信息的有效适用范围,所述AI模型相关信息包括以下至少一项:AI模型、AI模型参数、AI模型的输入、AI模型的输出。
第五方面,提供了一种第一通信设备,该第一通信设备包括处理器和存储器,所述存储器存储可在所述处理器上运行的程序或指令,所述程序或指令被所述处理器执行时实现如第一方面所述的方法的步骤。
第六方面,提供了一种第一通信设备,包括处理器及通信接口,其中,所述通信接口用于获取与AI模型相关信息关联的第一信息;所述处理器用于根据所述第一信息,确定目标信息,所述目标信息包括以下至少一项:目标AI模型、所述AI模型相关信息的有效性信息或者基于所述目标AI模型进行定位得到的反馈信息;所述第一信息用于表示所述AI模型相关信息的有效适用范围,所述AI模型相关信息包括以下至少一项:AI模型、AI模型参数、AI模型的输入、AI模型的输出。
第七方面,提供了一种第二通信设备,该第二通信设备包括处理器和存储器,所述存储器存储可在所述处理器上运行的程序或指令,所述程序或指令被所述处理器执行时实现如第二方面所述的方法的步骤。
第八方面,提供了一种第二通信设备,包括处理器及通信接口,其中,所述通信接口用于接收第一通信设备发送的目标信息,所述目标信息包括以下至少一项: 目标AI模型、所述AI模型相关信息的有效性信息或者基于所述目标AI模型进行定位得到的反馈信息;所述目标信息为根据AI模型相关信息关联的第一信息确定的,所述第一信息用于表示所述AI模型相关信息的有效适用范围,所述AI模型相关信息包括以下至少一项:AI模型、AI模型参数、AI模型的输入、AI模型的输出。
第九方面,提供了一种通信***,包括:第一通信设备及第二通信设备,所述第一通信设备可用于执行如第一方面所述的基于AI模型的定位方法的步骤,所述第二通信设备可用于执行如第二方面所述的基于AI模型的定位方法的步骤。
第十方面,提供了一种可读存储介质,所述可读存储介质上存储程序或指令,所述程序或指令被处理器执行时实现如第一方面所述的方法的步骤,或者实现如第二方面所述的方法的步骤。
第十一方面,提供了一种芯片,所述芯片包括处理器和通信接口,所述通信接口和所述处理器耦合,所述处理器用于运行程序或指令,实现如第一方面所述的方法,或实现如第二方面所述的方法。
第十二方面,提供了一种计算机程序/程序产品,所述计算机程序/程序产品被存储在存储介质中,所述计算机程序/程序产品被至少一个处理器执行以实现如第一方面或第二方面所述的基于AI模型的定位方法的步骤。
在本申请实施例中,第一通信设备获取与AI模型相关信息关联的第一信息;第一通信设备根据第一信息,确定目标信息,目标信息包括以下至少一项:目标AI模型、AI模型相关信息的有效性信息或者基于目标AI模型进行定位得到的反馈信息,由于第一信息用于表示AI模型相关信息的有效适用范围,基于第一信息确定目标信息,可以使得定位的AI模型更加符合实际场景需求,还可以基于确定的目标信息更新AI模型和/或AI模型参数,使得基于更新后的AI模型得到定位结果准确性更高。
附图说明
图1是本申请实施例可应用的无线通信***的结构图;
图2是本申请实施例提供的基于AI模型的定位方法的AI模型原理框图之一;
图3是本申请实施例提供的基于AI模型的定位方法的AI模型原理框图之二;
图4是本申请实施例提供的基于AI模型的定位方法的流程示意图之一;
图5是本申请实施例提供的基于AI模型的定位方法的交互流程示意图之一;
图6是本申请实施例提供的基于AI模型的定位方法的交互流程示意图之二;
图7是本申请实施例提供的基于AI模型的定位方法的交互流程示意图之三;
图8是本申请实施例提供的基于AI模型的定位方法的参数范围示意图之一;
图9是本申请实施例提供的基于AI模型的定位方法的参数范围示意图之二;
图10是本申请实施例提供的基于AI模型的定位方法的交互流程示意图之四;
图11是本申请实施例提供的基于AI模型的定位方法的交互流程示意图之五;
图12是本申请实施例提供的基于AI模型的定位方法的交互流程示意图之六
图13是本申请实施例提供的基于AI模型的定位装置的结构示意图之一
图14是本申请实施例提供的基于AI模型的定位装置的结构示意图之二;
图15是本申请实施例提供的通信设备的结构示意图;
图16是本申请实施例提供的终端的硬件结构示意图;
图17是本申请实施例的网络侧设备的结构示意图;
图18是本申请实施例的网络侧设备的另一结构示意图。
具体实施方式
下面将结合本申请实施例中的附图,对本申请实施例中的技术方案进行清楚描述,显然,所描述的实施例是本申请一部分实施例,而不是全部的实施例。基于本申请中的实施例,本领域普通技术人员所获得的所有其他实施例,都属于本申请保护的范围。
本申请的说明书和权利要求书中的术语“第一”、“第二”等是用于区别类似的对象,而不用于描述特定的顺序或先后次序。应该理解这样使用的术语在适当情况下可以互换,以便本申请的实施例能够以除了在这里图示或描述的那些以外的顺序实施,且“第一”、“第二”所区别的对象通常为一类,并不限定对象的个数,例如第一对象可以是一个,也可以是多个。此外,说明书以及权利要求中“和/或”表示所连接对象的至少其中之一,字符“/”一般表示前后关联对象是一种“或”的关系。
值得指出的是,本申请实施例所描述的技术不限于长期演进型(Long Term Evolution,LTE)/LTE的演进(LTE-Advanced,LTE-A)***,还可用于其他无线通信***,诸如码分多址(Code Division Multiple Access,CDMA)、时分多址(Time Division Multiple Access,TDMA)、频分多址(Frequency Division Multiple Access,FDMA)、正交频分多址(Orthogonal Frequency Division Multiple Access,OFDMA)、单载波频分多址(Single-carrier Frequency Division Multiple Access,SC-FDMA)和其他***。本申请实施例中的术语“***”和“网络”常被可互换地使用,所描述的技术既可用于以上提及的***和无线电技术,也可用于其他***和无线电技术。以下描述出于示例目的描述了新空口(New Radio,NR)***,并且在以下大部分描述中使用NR术语,但是这些技术也可应用于NR***应用以外的应用,如第6代(6th Generation,6G)通信***。
图1示出本申请实施例可应用的一种无线通信***的框图。无线通信***包括 终端11和网络侧设备12。其中,终端11可以是手机、平板电脑(Tablet Personal Computer)、膝上型电脑(Laptop Computer)或称为笔记本电脑、个人数字助理(Personal Digital Assistant,PDA)、掌上电脑、上网本、超级移动个人计算机(ultra-mobile personal computer,UMPC)、移动上网装置(Mobile Internet Device,MID)、增强现实(augmented reality,AR)/虚拟现实(virtual reality,VR)设备、机器人、可穿戴式设备(Wearable Device)、车载设备(VUE)、行人终端(PUE)、智能家居(具有无线通信功能的家居设备,如冰箱、电视、洗衣机或者家具等)、游戏机、个人计算机(personal computer,PC)、柜员机或者自助机等终端侧设备,可穿戴式设备包括:智能手表、智能手环、智能耳机、智能眼镜、智能首饰(智能手镯、智能手链、智能戒指、智能项链、智能脚镯、智能脚链等)、智能腕带、智能服装等。需要说明的是,在本申请实施例并不限定终端11的具体类型。网络侧设备12可以包括接入网设备或核心网设备,其中,接入网设备12也可以称为无线接入网设备、无线接入网(Radio Access Network,RAN)、无线接入网功能或无线接入网单元。接入网设备12可以包括基站、WLAN接入点或WiFi节点等,基站可被称为节点B、演进节点B(eNB)、接入点、基收发机站(Base Transceiver Station,BTS)、无线电基站、无线电收发机、基本服务集(Basic Service Set,BSS)、扩展服务集(Extended Service Set,ESS)、家用B节点、家用演进型B节点、发送接收点(Transmitting Receiving Point,TRP)或所述领域中其他某个合适的术语,只要达到相同的技术效果,所述基站不限于特定技术词汇,需要说明的是,在本申请实施例中仅以NR***中的基站为例进行介绍,并不限定基站的具体类型。核心网设备可以包含但不限于如下至少之一:核心网节点、核心网功能、移动管理实体(Mobility Management Entity,MME)、接入移动管理功能(Access and Mobility Management Function,AMF)、会话管理功能(Session Management Function,SMF)、用户平面功能(User Plane Function,UPF)、策略控制功能(Policy Control Function,PCF)、策略与计费规则功能单元(Policy and Charging Rules Function,PCRF)、边缘应用服务发现功能(Edge Application Server Discovery Function,EASDF)、统一数据管理(Unified Data Management,UDM),统一数据仓储(Unified Data Repository,UDR)、归属用户服务器(Home Subscriber Server,HSS)、集中式网络配置(Centralized network configuration,CNC)、网络存储功能(Network Repository Function,NRF),网络开放功能(Network Exposure Function,NEF)、本地NEF(Local NEF,或L-NEF)、绑定支持功能(Binding Support Function,BSF)、应用功能(Application Function,AF),位置管理功能LMF(location manage function),增强服务移动定位中心E-SMLC,网络数据分析功能(network data analytics function,NWDAF)等。需要说明的是,在本申请实施例中仅以NR***中的核心网设备为例进行介绍,并不限定核心 网设备的具体类型
AI目前在各个领域获得了广泛的应用。AI模型有多种实现方式,例如神经网络、决策树、支持向量机、贝叶斯分类器等。本申请实施例以神经网络为例进行说明,但是并不限定AI模型的具体类型。神经网络由神经元组成,其中a1,a2,...aK为输入,w为权值(乘性系数),b为偏置(加性系数),σ(.)为激活函数。常见的激活函数包括Sigmoid、tanh、修正线性单元(Rectified Linear Unit,ReLU)等等。神经网络的参数通过优化算法进行优化。优化算法就是一种能够帮我们最小化或者最大化目标函数(有时候也叫损失函数)的一类算法。而目标函数往往是模型参数和数据的数学组合。例如给定数据X和其对应的标签Y,我们构建一个神经网络模型f(.),有了模型后,根据输入x就可以得到预测输出f(x),并且可以计算出预测值和真实值之间的差距(f(x)-y),这个就是损失函数。我们的目的是找到合适的w,b使上述的损失函数的值达到最小,损失值越小,则说明我们的模型越接近于真实情况。
目前常见的优化算法,基本都是基于误差反向传播(error Back Propagation,BP)算法。BP算法的基本思想是,学习过程由信号的正向传播与误差的反向传播两个过程组成。正向传播时,输入样本从输入层传入,经各隐层逐层处理后,传向输出层。若输出层的实际输出与期望的输出不符,则转入误差的反向传播阶段。误差反传是将输出误差以某种形式通过隐层向输入层逐层反传,并将误差分摊给各层的所有单元,从而获得各层单元的误差信号,此误差信号即作为修正各单元权值的依据。这种信号正向传播与误差反向传播的各层权值调整过程,是周而复始地进行的。权值不断调整的过程,也就是网络的学习训练过程。此过程一直进行到网络输出的误差减少到可接受的程度,或进行到预先设定的学习次数为止。
常见的优化算法有梯度下降(Gradient Descent)、随机梯度下降(Stochastic Gradient Descent,SGD)、mini-batch gradient descent(小批量梯度下降)、动量法(Momentum)、Nesterov(发明者的名字,具体为带动量的随机梯度下降)、自适应梯度下降(ADAptive GRADient descent,Adagrad)、Adadelta、均方根误差降速(root mean square prop,RMSprop)、自适应动量估计(Adaptive Moment Estimation,Adam)等。
这些优化算法在误差反向传播时,都是根据损失函数得到的误差/损失,对当前神经元求导数/偏导,加上学习速率、之前的梯度/导数/偏导等影响,得到梯度,将梯度传给上一层。
本申请实施例的方法可以应用于定位场景中,例如基于AI模型进行定位,由于测量环境和需求是会随着时间改变的,因此随着时间的推移AI模型得到的定位结果可能会不满足当前的需求,因此,本申请实施例的方法可以基于实际情况确定当前AI模型是否有效,并基于判定去更新AI模型和/或AI模型参数,来使得AI定位的 结果满足性能指标。
在一实施例中,如图2、图3所示为AI模型应用的原理框图。其中,AI模型应用(AI inference)就是基于AI模型和AI模型参数以及当前的输入数据获取输出的过程。
下面结合附图,通过一些实施例及其应用场景对本申请实施例提供的基于AI模型的定位方法进行详细地说明。
图4是本申请实施例提供的基于AI模型的定位方法的流程示意图之一。如图4所示,本实施例提供的方法,包括:
步骤101、第一通信设备获取与AI模型相关信息关联的第一信息;
其中,第一信息用于表示AI模型相关信息的有效适用范围,AI模型相关信息包括以下至少一项:AI模型、AI模型参数、AI模型的输入、AI模型的输出。
具体地,第一信息与AI模型相关信息关联,第一信息表示AI模型相关信息的有效适用范围,例如输入有效范围、输出有效范围、有效性条件等等;
值得注意的是,AI模型相关信息只是保证选定的、特定的AI模型可以被第一通信设备与其它通信设备所理解和确认,具体的传输格式和传输方式,在此并不做限制。所述AI模型相关信息也是用于保护AI模型的相应特征。
可选地,第一通信设备可以从第二通信设备(例如为模型管理设备、网络侧设备(如NWDAF)、定位服务器(如LMF,E-SMLC)等)获取AI模型相关信息以及与AI模型相关信息关联的第一信息。
值得注意的是,所述第一信息可以包含于AI模型相关信息,也可以与AI模型相关信息一起传输,或者独立于AI模型相关信息传输,但通过AI模型的识别信息和第一信息来进行关联。
步骤102、第一通信设备根据第一信息,确定目标信息,目标信息包括以下至少一项:目标AI模型、AI模型相关信息的有效性信息或者基于目标AI模型进行定位得到的反馈信息。
具体地,可以根据第一信息中参数的有效适用范围,确定目标信息。例如,AI模型的相关信息关联的第一信息可以确定所述AI模型的相关信息的有效适用范围,说明AI模型的相关信息在特定场景,特定区域是有效,又比如,假设AI模型的输出处于第一信息中参数的有效适用范围,说明AI模型的输出有效。再比如,假设AI模型的输入处于第一信息中参数的有效适用范围,说明AI模型的输入输出有效或者AI模型有效。
在另一个实施例中,假设有多个AI模型,若干AI模型分别关联相应的第一信息,根据所述第一信息确定多个AI模型的一个或多个AI模型为目标AI模型。
例如,反馈信息可以包括有效性信息、目标AI模型、AI模型的输入、AI模型的输出以及终端的测量信息等。
可选的,反馈信息与所述第一信息有对应关系。
可选地,第一通信设备可以将反馈信息发送给第二通信设备,第二通信设备基于反馈信息可以更新AI模型和/或AI模型参数。
可选地,第一通信设备包括以下至少一项:位置管理功能(Location Management Function,LMF)网元和LMF网元的演进设备;
定位服务器;
网络数据分析功能(NetWork Data Analytics Function,NWDAF)网元;
其它网络侧设备;
终端;
监控设备Actor。
其中,其它网络侧设备例如包括接入网设备,以及除上述核心网网元之外的设备等。
本实施例的方法,第一通信设备获取与AI模型相关信息关联的第一信息;第一通信设备根据第一信息,确定目标信息,目标信息包括以下至少一项:目标AI模型、AI模型相关信息的有效性信息或者基于目标AI模型进行定位得到的反馈信息,由于第一信息用于表示AI模型相关信息的有效适用范围,基于第一信息确定目标信息,可以使得定位的AI模型更加符合实际场景需求,还可以基于确定的目标信息更新AI模型和/或AI模型参数,使得基于更新后的AI模型得到定位结果准确性更高。
可选地,有效性信息可以包括以下至少一项:
有效性指示信息,用于指示AI模型相关信息是否有效;
有效程度;
有效等级;
失效原因;
可靠性指示信息,用于指示基于目标AI模型进行定位得到的结果是否可靠;
可靠性程度;
可靠性等级。
其中,有效性指示信息可以通过至少一个bit指示,例如0表示失效,1表示有效。
可靠性指示信息与有效性指示信息类似。
有效性程度可以通过至少一个bit表示,例如0-1之间的有效程度。
可选地,AI模型的输入和/或AI模型的输出,包括以下至少一项:
终端的定位信号测量信息;所述终端的位置信息;误差信息;信道冲激响应CIR;首径的功率;首径的时延;首径的到达时间TOA;首径的参考信号时间差RSTD;首径的到达角;首径的天线子载波相位差;多径的功率;多径的时延;多径 的TOA;多径的RSTD;多径的到达角;多径的天线子载波相位差;平均过量时延;均方根时延拓展;相干带宽;
误差信息包括以下至少一项:位置误差值、测量误差值、AI模型误差值或AI模型参数误差值。
在一实施例中,该方法还包括:
第一通信设备获取目标终端的第二信息;第二信息用于表示目标终端获取的定位相关信息;
步骤102可以通过如下方式实现:
第一通信设备根据第一信息和第二信息,确定目标信息。
具体地,第一信息用于表示AI模型相关信息的有效适用范围,第二信息用于表示目标终端获取的定位相关信息,例如终端的测量信息、终端的位置信息等;其中,目标终端可以是需要定位的终端。
第一通信设备根据第一信息和第二信息,确定目标AI模型、AI模型相关信息的有效性信息或者基于目标AI模型进行定位得到的反馈信息中至少一项。
上述实施方式中,第一通信设备获取目标终端的第二信息;第一通信设备根据第一信息和第二信息,确定目标信息,目标信息包括以下至少一项:目标AI模型、AI模型相关信息的有效性信息或者基于目标AI模型进行定位得到的反馈信息,由于第一信息用于表示AI模型相关信息的有效适用范围,第二信息表示目标终端获取的定位相关信息;基于第一信息和第二信息确定目标信息,可以使得定位的AI模型更加符合实际场景需求,还可以基于确定的目标信息更新AI模型和/或AI模型参数,使得基于更新后的AI模型得到定位结果准确性更高。
在一实施例中,上述步骤“第一通信设备根据第一信息和第二信息,确定目标信息”,可以包括:
第一通信设备根据第二信息中参数的取值以及第一信息中对应参数的范围,确定目标信息。
示例性地,如图5所示,第一通信设备根据第一信息,确定目标信息,如确定有效性信息,反馈给模型管理设备,模型管理设备基于第一通信设备反馈的信息进行模型训练、更新。
可选地,在有效性信息用于指示AI模型相关信息失效的情况下,第二通信设备可以为模型管理设备或者数据管理设备或者有效性功能验证模块(如:Actor);
在有效性信息用于指示所述AI模型相关信息有效的情况下,第二通信设备为以下至少一项:模型管理设备、LMF、NWDAF、有效性功能验证模块(如:Actor)。
示例性地,若第一通信设备为LMF,如图6所示,LMF和模型管理设备可以是一个设备或者是不同的设备;
可选地,LMF和模型应用模块在一个设备上或者在不同的设备上。
示例性地,若第一通信设备为终端,如图7所示,终端基于AI模型进行定位,并确定有效性信息。
可选地,第一信息包括以下至少一项:小区信息;区域信息;有效时间信息;场景信息;信干噪比SINR范围;
其中,小区信息包括以下至少一项:
一个或多个小区的识别信息;
一个或多个基站的识别信息;
一个或多个传输接收点(Transmission Reception Point,TRP)的识别信息;
小区列表信息;
小区频域范围信息;例如,频带标识ID;
其中,所述区域信息包括以下至少一项:
区域识别信息;距离范围信息;距离范围对应的参考点信息;
其中,有效时间信息包括以下至少一项:
计时器时长;
计时器起始时间;
其中,场景信息包括以下至少一项:
视距LOS场景;非视距NLOS场景;复杂场景;室内场景;室外场景。
可选地,复杂场景例如为多种场景的混合场景等,NLOS场景还可以包括NLOS恶劣场景。
其中,距离范围信息例如为距某个参考点多少距离内有效,例如500m内有效。
可选地,第二信息包括以下至少一项:目标终端的位置信息、小区信息、区域信息,计时器信息、场景信息和目标终端测量的SINR;
其中,小区信息为目标终端的服务小区、参考小区或者参考信号接收功率RSRP最强小区的以下至少一项信息,至少一项信息包括:识别信息、频域信息;
其中,区域信息为目标终端的区域识别信息;
其中,所场景信息为目标终端所处的场景信息。
可选地,位置信息是通过以下至少一种方式得到的,所至少一种方式包括:
到达时间差定位法(Observed Time Deviation Of Arrival,OTDOA)、全球导航卫星***(Global Navigation Satellite System,GNSS)、下行到达时间差、上行到达时间差、上行蓝牙到达角(Angle of Arrival,AoA)、蓝牙出发角(Angle of Departure,AoD)、蓝牙、传感器或无线保真WIFI。
具体地,第一通信设备接收第一信息和AI模型相关信息,第一信息用于表示AI模型相关信息的有效适用范围,例如AI模型相关信息的有效小区、有效区域、有效时间、有效场景和/或有效SINR范围等。
第二信息可以是目标终端的位置信息、小区信息、区域信息,计时器信息、场 景信息和所述目标终端测量的SINR中至少一项。
根据第二信息中参数的取值以及第一信息中对应参数的范围,确定目标信息。例如,将第二信息中每个参数的取值与第一信息中对应参数的范围进行比较,若第二信息中参数的取值处于第一信息中对应参数的范围内,说明AI模型相关信息有效,可以是第二信息中全部参数均处于第一信息中对应参数的范围内,也可以是绝大部分参数的取值处于第一信息中对应参数的范围内,或者个别参数的取值即使处于第一信息中对应参数的范围外,但是与该范围的偏差不大,也可以认为AI模型相关信息是有效的。
在一实施例中,确定有效性信息,包括以下至少一种情况:
在第二信息中参数的取值处于第一信息中对应参数的范围内的情况下,AI模型相关信息有效。
若第一通信设备确定目标终端的小区信息不属于第一信息中的小区信息的范围内,AI模型相关信息失效。
若第一通信设备确定目标终端的计时器超时,AI模型相关信息失效或超时。
若第一通信设备确定目标终端的场景信息不属于第一信息中的场景信息的范围内,AI模型相关信息失效。
若第一通信设备确定目标终端的位置不处于第一信息中的小区信息和/或区域信息对应的位置范围内,AI模型相关信息失效。
具体地,在第二信息中参数的取值处于第一信息中对应参数的范围内的情况下,则AI模型相关信息有效,在第二信息中参数的取值处于第一信息中对应参数的范围外的情况下,则AI模型相关信息失效,例如包括以下几种情况:
目标终端的小区信息不属于第一信息中的小区信息的范围内,AI模型相关信息失效;
目标终端的区域信息不属于第一信息中的区域信息的范围内,AI模型相关信息失效;
目标终端的计时器超时,AI模型相关信息失效或超时;
目标终端的场景信息不属于第一信息中的场景信息的范围内,AI模型相关信息失效;
目标终端的位置不处于第一信息中的小区信息和/或区域信息对应的位置范围内,AI模型相关信息失效;
目标终端测量的SINR不处于第一信息中SINR范围内,AI模型相关信息失效。
可选地,目标终端的计时器满足以下至少一种情况:
计时器的时长为第一信息中的计时器时长;
计时器从第一信息中计时器起始时间开始计时;
在AI模型和/或AI模型参数更新的情况下,计时器重新计时。
可选地,反馈信息为根据有效性信息确定的,包括以下几种情况:
若有效性信息用于指示AI模型相关信息有效,反馈信息至少包括AI模型的输出;或,
若有效性信息用于指示AI模型相关信息失效,反馈信息包括以下至少一项:误差原因;第二信息;有效性信息;AI模型请求;AI模型更新请求;数据收集请求。
具体地,在确定出有效性信息后,可以基于有效性信息确定反馈信息包括的内容,例如在AI模型相关信息有效时,反馈信息至少包括AI模型的输出;或,在AI模型相关信息失效时,反馈信息可以包括以下至少一项:误差原因;第二信息;有效性信息;AI模型请求;AI模型更新请求;数据收集请求,可以用于对AI模型进行训练、选择、更新。
可选地,反馈信息包括以下至少一种情况:
在第二信息中参数的取值处于第一信息中对应参数的范围内的情况下,反馈信息中包括第二信息;或者,
在所述第二信息中参数的取值不处于所述第一信息中对应参数的范围内的情况下,所述反馈信息中包括所述第二信息。
若第一通信设备确定目标终端的小区信息不属于第一信息中的小区信息的范围内,反馈信息包括目标终端的小区信息;
若目标终端的区域信息不属于第一信息中的区域信息的范围内,反馈信息包括目标终端的区域信息。
若第一通信设备确定所述目标终端的场景信息不属于所述第一信息中的场景信息的范围内,反馈信息包括目标终端的场景信息;
若目标终端的位置不处于第一信息中的小区信息和/或区域信息对应的位置范围内,反馈信息包括目标终端的位置信息;
若目标终端测量的SINR不处于第一信息中SINR范围内,反馈信息包括目标终端策测量的SINR;
若第一通信设备确定目标终端的计时器超时,反馈信息包括目标终端的计时器信息,例如包括计时器时长,起始时间等。
可选地,在第二信息中参数的取值处于第一信息中对应参数的第一范围内的情况下,目标AI模型为第一信息的第一范围对应的AI模型。
可选地,反馈信息包括:目标AI模型。
具体地,若第一信息中参数具有多个有效适用范围,不同的有效适用范围可以对应不同的AI模型和/或AI模型参数。
若第二信息中参数的取值处于第一信息中对应参数的某个第一范围内,目标AI模型可以是该第一范围对应的AI模型。
可选地,该方法还包括:
第一通信设备接收多个预配置的AI模型和/或AI模型参数,以及AI模型和/或AI模型参数对应的第一信息。
可选地,所述方法还包括:
第一通信设备根据第一信息的第一范围,从多个预配置的AI模型中,获取第一信息的第一范围对应的AI模型。
可选地,作为另一种实现方式,反馈信息包括以下至少一项:有效性信息、第二信息、第一测量信息和AI模型的输出;
第一测量信息包括以下至少一项:
信号测量信息;位置信息;误差值;信道冲击响应(Channel Impulse Response,CIR)信息;功率时延谱(Power-Delay Profile,PDP)信息。
其中,误差值包括以下至少之一:位置误差值,测量误差值。
其中,CIR信息例如包括时域或频域冲击响应信息,或时域或频域冲击响应信息的处理信息(如截断信息)等,可以包括单天线或者多天线的CIR信息。
上述实施方式中,第一通信设备根据第一信息和第二信息,确定目标信息,目标信息包括以下至少一项:目标AI模型、AI模型相关信息的有效性信息或者基于目标AI模型进行定位得到的反馈信息,基于第二信息中参数的取值和第一信息中对应参数的范围确定目标信息,可以使得定位的AI模型更加符合实际场景需求,还可以基于确定的目标信息更新AI模型和/或AI模型参数,使得基于更新后的AI模型得到定位结果准确性更高。
在另一实施例中,第二信息包括目标终端获得的第二测量信息,步骤“第一通信设备根据第一信息和第二信息,确定目标信息”还可以通过如下步骤实现:
在第二测量信息为一次测量得到的测量信息的情况下,第一通信设备根据第二测量信息中参数的取值以及第一信息中对应参数的范围,确定目标信息;值得注意的是,在一种情况下,所述一次测量可以理解为M(M可以为1,2,3,4至少一个)个测量实例平滑的结果。
在第二测量信息为多次测量得到的测量信息的情况下,第一通信设备根据第二测量信息中参数的分布与第一信息中对应参数的分布的一致性,确定目标信息。
具体地,第一通信设备接收第一信息和AI模型相关信息,第一信息用于表示AI模型相关信息的有效适用范围,本实施例中具体可以表示获得AI模型相关信息的输入数据所处的有效适用范围,例如SINR范围、噪声的均值和方差、绝对时间(absolute time)的均值和方差、时延拓展的均值和方差、角度拓展的均值和方差等等,换言之,AI模型相关信息只对处于上述有效适用范围的输入数据有效。
本实施例中具体可以表示获得AI模型相关信息的输出数据所处的有效适用范围,例如SINR范围、噪声的均值和方差、绝对时间(absolute time)的均值和方差、时延拓展的均值和方差、角度拓展的均值和方差等等,又比如测量信息和位置 信息的范围,换言之,AI模型相关信息只对处于上述有效适用范围的输出数据有效。比如,输出的测量信息和位置信息超出有效适用范围,则无效。
第二信息可以是目标终端测量得到的第二测量信息,包括信干噪比SINR范围、噪声值、NLOS引入的绝对时间值、时延拓展值、角度拓展值、SINR均值和方差、噪声均值和方差、NLOS引入的绝对时间均值和方差、时延拓展均值和方差、角度拓展均值和方差中至少一项。
一种实现方式:
在第二测量信息为一次测量得到的测量信息的情况下,第一通信设备根据第二测量信息中参数的取值以及第一信息中对应参数的范围,确定目标信息。例如,将第二测量信息中每个参数的取值与第一信息中对应参数的范围进行比较,若第二测量信息中参数的取值处于第一信息中对应参数的范围内,说明AI模型相关信息有效,可以是第二测量信息中全部参数均处于第一信息中对应参数的范围内,也可以是绝大部分参数的取值处于第一信息中对应参数的范围内,或者个别参数的取值即使处于第一信息中对应参数的范围外,但是与该范围的偏差不大,也可以认为AI模型相关信息是有效的。
如图8所示,比如C1,C2和C3处的绝对时间,与第一信息中该参数的范围不一致,C2更接近第一信息中该参数的范围,因此,可以认为C2对应的AI模型相关信息有效,而C2和C3对应的失效,因此,C1、C2和C3有效性不一样,C1,C2和C3误差不一致。
值得注意的是,所述失效可以理解为有效性程度的描述,如C2与第一信息的偏离程度可以表示为有效性信息。进一步,在一个实施例下,可以表示为C2-均值/有效性范围。
另一种实现方式:
在第二测量信息为多次测量得到的测量信息的情况下,第一通信设备根据第二测量信息中参数的分布与第一信息中对应参数的分布的一致性,确定目标信息。例如,将第二测量信息中每个参数的分布与第一信息中对应参数的分布进行比较,若第二测量信息中参数的分布与第一信息中对应参数的分布的一致,说明AI模型相关信息有效,可以是第二测量信息中全部参数的分布均与第一信息中对应参数的分布的一致,也可以是绝大部分参数的分布与第一信息中对应参数的分布的一致,或者个别参数的分布即使与第一信息中对应参数的分布的不一致,但是与第一信息中对应参数的分布的偏差不大,也可以认为AI模型相关信息是有效的。
如图9所示,第二测量信息中某参数的分布与第一信息中对应参数的分布相差太大,仅为一少部分交集,因此可以认为AI模型相关信息失效。
得注意的是,所述失效可以理解为有效性程度的描述,如第二测量信息中某参数的分布与第一信息的偏离程度可以表示为有效性信息。
上述实施方式中,基于第二测量信息中参数的取值和第一信息中对应参数的范围确定目标信息,和/或,基于第二测量信息中参数的分布和第一信息中对应参数的分布确定目标信息,可以使得定位的AI模型更加符合实际场景需求,还可以基于确定的目标信息更新AI模型和/或AI模型参数,使得基于更新后的AI模型得到定位结果准确性更高。
示例性地,如图10所示,第一通信设备(例如终端)根据第一信息和第二测量信息,确定目标信息,如确定有效性信息,反馈给模型管理设备(例如网络侧设备),模型管理设备基于第一通信设备反馈的信息进行模型训练、更新。
示例性地,如图11所示,第一通信设备可以为LMF,模型管理设备和LMF可以是一个设备或者是不同的设备。可选地,LMF和模型应用模块可以是在一个设备上或者在不同的设备上。
可选地,第一信息包括以下至少一项:
SINR范围;
噪声的范围;
噪声分布的均值和/或方差;
NLOS引入的绝对时间范围或NLOS引入绝对时间的均值和/或方差;
时延拓展的范围或时延拓展的均值和/或方差;
角度拓展的范围或角度拓展的均值和/或方差;
第一测量信息的范围;
第一测量信息的均值和/或方差。
可选地,NLOS引入的绝对时间为AI模型获取的时间信息减去非AI模型获取的时间信息;
时延拓展、角度拓展、噪声分布是根据第一测量信息或AI模型的输出获取的特征信息。
具体地,时间信息例如包括RSTD、TOA、Rx-Tx、计时器信息等。
可选地,第二测量信息包括以下至少一项:
信干噪比SINR范围、噪声值、NLOS引入的绝对时间值、时延拓展值、角度拓展值、SINR均值和方差、噪声均值和方差、NLOS引入的绝对时间均值和方差、时延拓展均值和方差、角度拓展均值和方差;
其中,SINR范围和所述SINR均值和方差是根据以下至少一项信息的SINR得到的;
噪声值和噪声均值和方差是以下至少一项信息的噪声值得到的;至少一项信息包括:测量信道、测量信号或第一测量信息。
可选地,第一测量信息包括以下至少一项:
信号测量信息;位置信息;误差值;信道冲击响应CIR信息;功率时延谱PDP 信息。
其中,误差值包括以下至少之一:位置误差值,测量误差值。
可选地,信号测量信息包括以下至少一项:
参考信号时间差(Reference Signal Time Difference,RSTD)测量结果、往返时延测量结果、到达角AOA测量结果、离开角AOD测量结果、参考信息接收功率(Reference Signal Received Power,RSRP)、多径测量信息、视距LOS指示信息;
可选地,多径测量信息包括以下至少一项:
首径的功率、首径的时延、首径的到达时间TOA、首径的参考信号时间差RSTD、首径的天线子载波相位差、首径的天线子载波相位、多径的功率、多径的时延、TOA、多径的RSTD、多径的天线子载波相位差或多径的天线子载波相位。
可选地,确定有效性信息,包括以下至少一种情况:
在第二测量信息中参数的取值处于第一信息中对应参数的范围内的情况下,AI模型相关信息有效;
在第二测量信息中至少一个参数的取值处于第一信息中对应参数的范围外的情况下,AI模型相关信息失效。
在第二测量信息中参数的分布与第一信息中对应参数的分布一致的情况下,AI模型相关信息有效;
在第二测量信息中至少一个参数的分布与第一信息中对应参数的分布不一致的情况下,AI模型相关信息失效。
具体地,将第二测量信息中参数的取值与第一信息中对应参数的范围进行比较,若第二测量信息中参数的取值处于第一信息中对应参数的范围内,说明AI模型相关信息有效,可以是第二测量信息中全部参数均处于第一信息中对应参数的范围内,也可以是绝大部分参数的取值处于第一信息中对应参数的范围内,或者个别参数的取值即使处于第一信息中对应参数的范围外,但是与该范围的偏差不大,也可以认为AI模型相关信息是有效的。或,
若第二测量信息中至少一个参数的取值处于第一信息中对应参数的范围外,或第二测量信息中全部参数的取值均处于第一信息中对应参数的范围外,则说明AI模型相关信息失效。
将第二测量信息中每个参数的分布与第一信息中对应参数的分布进行比较,若第二测量信息中参数的分布与第一信息中对应参数的分布的一致,说明AI模型相关信息有效,可以是第二测量信息中全部参数的分布均与第一信息中对应参数的分布的一致,也可以是绝大部分参数的分布与第一信息中对应参数的分布的一致,或者个别参数的分布即使与第一信息中对应参数的分布的不一致,但是与第一信息中对应参数的分布的偏差不大,也可以认为AI模型相关信息是有效的。或,
若第二测量信息中至少一个参数的分布与第一信息中对应参数的分布不一致, 或第二测量信息中全部参数的分布均与第一信息中对应参数的分布不一致,说明AI模型相关信息失效。
进一步地,在第二测量信息中至少一个参数的分布与第一信息中对应参数的分布不一致的情况下,若第二测量信息中参数的分布与第一信息中对应参数的分布不满足于条件阈值信息,AI模型相关信息失效。
可选地,条件阈值信息包括以下至少一项:
正态分布中σ原则、2σ原则或3σ原则对应的绝对时间;
σ原则、2σ原则或3σ原则对应的时延拓展值;
σ原则、2σ原则或3σ原则对应的角度拓展值;
第二测量信息中参数的均值和第一信息中对应参数的均值的最大差;
第二测量信息中参数的均值和第一信息中对应参数的均值的最大方差;
第二测量信息中参数的均值和第一信息中对应参数的方差的最大差;
第二测量信息中参数的均值和第一信息中对应参数的方差的最大方差。
其中,σ原则:数值分布在(μ-σ,μ+σ)中的概率为0.6526;2σ原则:数值分布在(μ-2σ,μ+2σ)中的概率为0.9544;3σ原则:数值分布在(μ-3σ,μ+3σ)中的概率为0.9974;
其中在正态分布中σ代表标准差,μ代表均值。
可选地,反馈信息包括以下至少一项:有效性信息、第二测量信息、AI模型的输入、AI模型的输出、AI模型识别信息、AI模型更新请求。
可选地,AI模型更新请求,包括以下至少一项:
AI模型的标识ID、满足第二测量信息分布的AI模型和/或AI模型参数、AI模型和/或AI模型参数。
在一实施例中,第一信息为根据AI模型的测试集和验证集获取到的,包括以下至少一种情况:
第一信息是根据AI模型的测试集和验证集的输入数据获取的特征信息;
第一信息是根据AI模型的测试集和验证集的输出数据获取的特征信息;
第一信息是根据AI模型的测试集和验证集的输入数据和输出数据获取的特征信息。
其中,测试集可以是第二信息,验证集可以是第一信息。
在一实施例中,第二测量信息是根据第一测量信息获得的特征信息;或,
第二测量信息是根据AI模型的输出获得的特征信息。
在一实施例中,第二测量信息是根据第一测量信息和AI模型的输出获得的特征信息。
在一实施例中,第一通信设备可以将目标信息发送给第二通信设备,例如将反馈信息发送给第二通信设备,第二通信设备根据反馈信息可以对AI模型和/或AI模 型参数进行更新。
可选地,该方法还包括:
第一通信设备接收更新的AI模型和/或AI模型参数,以及与AI模型和/或AI模型参数对应的第一信息。
在一实施例中,AI模型包括以下至少一项:
AI模型类型;
AI模型结构。
其中,AI模型类型例如包括:CNN、非监督、半监督、监督、RNN、LSTM、第一通信设备基于AI进行定位或第一通信设备在基于AI进行定位时只需要执行部分操作等。
其中,AI模型结构,例如包括:
神经网络列表,包括以下至少一项:每个神经网络的神经元类型、每个神经网络的神经元权重和偏差;
损失函数、优化函数类型;
池化层和卷积层位置和大小;
神经网络,例如为全连接神经网络,卷积神经网络,循环神经网络,残差网络等;
AI模型的网络结构还可以是多个小网络的组合方式,例如全连接+卷积,卷积+残差等等;
AI模型的网络结构还可以包括:隐藏层的层数;输入层与隐藏层的连接方式、多个隐藏层之间的连接方式、隐藏层与输出层的连接方式、每层神经元的数目等。
在另一实施例中,AI模型包括以下至少一项:
神经网络列表,包括以下至少一项:每个神经网络的神经元类型、每个神经网络的神经元权重和偏差;
超参数信息;
损失函数信息。
其中,超参数信息例如为AI模型外部的参数,例如池化Pooling大小(size)、Batch size(表示一次训练所选取的样本数)、迭代方式、学习速率,优化函数选择,损失函数选择等信息。
可选地,AI参数包括以下至少一项:
超参数信息;
AI模型描述参数信息;
AI模型的权值信息;
AI模型的初始参数。
其中,AI模型描述参数信息如AI模型参数的输入格式、AI模型参数的输出格式等。
AI模型的初始参数为用于迭代目标AI模型相关信息的初始参数。
图12是本申请实施例提供的基于AI模型的定位方法的交互流程示意图之二。如图12所示,本实施例提供的方法,包括:
步骤201、第二通信设备接收第一通信设备发送的目标信息,所述目标信息包括以下至少一项:目标AI模型、所述AI模型相关信息的有效性信息或者基于所述目标AI模型进行定位得到的反馈信息;
所述目标信息为根据AI模型相关信息关联的第一信息确定的,所述第一信息用于表示所述AI模型相关信息的有效适用范围,所述AI模型相关信息包括以下至少一项:AI模型、AI模型参数、AI模型的输入、AI模型的输出。
可选地,所述方法还包括:
所述第二通信设备向所述第一通信设备发送所述AI模型相关信息和所述AI模型相关信息关联的第一信息。
可选地,所述方法还包括:
所述第二通信设备向第一通信设备发送多个预配置的AI模型和/或AI模型参数,以及所述AI模型和/或AI模型参数对应的第一信息。
可选地,所述方法还包括:
所述第二通信设备向所述第一通信设备发送第二信息,所述第二信息用于表示所述终端获取的定位相关信息。
可选地,所述第一信息包括以下至少一项:小区信息;区域信息;有效时间信息;场景信息;信干噪比SINR范围;
其中,所述小区信息包括以下至少一项:
一个或多个小区的识别信息;
一个或多个基站的识别信息;
一个或多个传输接收点TRP的识别信息;
小区列表信息;
小区频域范围信息;
其中,所述区域信息包括以下至少一项:
区域识别信息;距离范围信息;距离范围对应的参考点信息;
其中,所述有效时间信息包括以下至少一项:
计时器时长;
计时器起始时间;
其中,所述场景信息包括以下至少一项:
视距LOS场景;非视距NLOS场景;复杂场景;室内场景;室外场景。
可选地,所述第二信息包括以下至少一项:所述目标终端的位置信息、小区信息、区域信息,计时器信息、场景信息和所述目标终端测量的SINR;
其中,所述小区信息为所述目标终端的服务小区、参考小区或者参考信号接收功率RSRP最强小区的以下至少一项信息,所述至少一项信息包括:识别信息、频域信息;
其中,所述区域信息为所述目标终端的区域识别信息;
其中,所述场景信息为所述目标终端所处的场景信息。
可选地,在所述第二信息中参数的取值处于所述第一信息中对应参数的范围内的情况下,所述AI模型相关信息有效。
可选地,若第二信息中参数的取值处于所述第一信息中对应参数的范围外的情况下,所述AI模型相关信息失效。
可选地,所述反馈信息包括以下至少一项:所述有效性信息、第二信息、第一测量信息和所述AI模型的输出;
第一测量信息包括以下至少一项:
信号测量信息;位置信息;误差值;信道冲击响应CIR信息;功率时延谱PDP信息。
可选地,所述第二信息包括所述目标终端获得的第二测量信息,在所述第二测量信息中参数的取值处于所述第一信息中对应参数的范围内的情况下,所述AI模型相关信息有效;和/或;
在所述第二测量信息中至少一个参数的取值处于所述第一信息中对应参数的范围外的情况下,所述AI模型相关信息失效。
可选地,所述第二信息包括所述目标终端获得的第二测量信息,在所述第二测量信息中参数的分布与所述第一信息中对应参数的分布一致的情况下,所述AI模型相关信息有效;和/或;
在所述第二测量信息中至少一个参数的分布与所述第一信息中对应参数的分布不一致的情况下,所述AI模型相关信息失效。
可选地,在所述第二测量信息中至少一个参数的分布与所述第一信息中对应参数的分布不一致的情况下,所述AI模型相关信息失效,包括:
在所述第二测量信息中参数的分布与所述第一信息中对应参数的分布不满足于条件阈值信息的情况下,所述AI模型相关信息失效。
可选地,所述条件阈值信息包括以下至少一项:
正态分布中σ原则、2σ原则或3σ原则对应的绝对时间;
σ原则、2σ原则或3σ原则对应的时延拓展值;
σ原则、2σ原则或3σ原则对应的角度拓展值;
所述第二测量信息中参数的均值和所述第一信息中对应参数的均值的最大差;
所述第二测量信息中参数的均值和所述第一信息中对应参数的均值的最大方差;
所述第二测量信息中参数的均值和所述第一信息中对应参数的方差的最大差;
所述第二测量信息中参数的均值和所述第一信息中对应参数的方差的最大方差。
可选地,所述第二测量信息包括以下至少一项:
信干噪比SINR范围、噪声值、NLOS引入的绝对时间值、时延拓展值、角度拓展值、SINR均值和方差、噪声均值和方差、NLOS引入的绝对时间均值和方差、时延拓展均值和方差、角度拓展均值和方差;
其中,所述SINR范围和所述SINR均值和方差是根据以下至少一项信息的SINR得到的;
噪声值和噪声均值和方差是以下至少一项信息的噪声值得到的;
所述至少一项信息包括:测量信道、测量信号或第一测量信息。
可选地,所述第一测量信息包括以下至少一项:
信号测量信息;位置信息;误差值;信道冲击响应CIR信息;功率时延谱PDP信息。
可选地,所述反馈信息包括以下至少一项:所述有效性信息、第二测量信息、AI模型的输入、AI模型的输出、AI模型识别信息、AI模型更新请求。
值得注意的是,所述AI模型的输入,可以AI模型的当前输入,也可以是根据多次输入确定的输入或输入分布,从而希望对端提供满足该输入AI模型和参数
值得注意的是,所述AI模型的输出,可以AI模型的当前输出,也可以是根据多次输入确定的输出或输出分布,从而希望对端提供满足该输出AI模型和参数
值得注意的是,所述AI模型的输入和输出,可以AI模型的当前输入和输出,也可以是根据多次输入和输出确定的输入和输出或输入和输出分布,从而希望对端提供满足该输入和输出的AI模型和参数
值得注意的是,所述第二测量信息,可以一次测量获得的第二测量信息,也可以是多次输入确定的第二测量信息的分布。
可选地,所述第一信息包括以下至少一项:
SINR范围;
噪声的范围;
噪声分布的均值和/或方差;
NLOS引入的绝对时间范围或NLOS引入绝对时间的均值和/或方差;
时延拓展的范围或时延拓展的均值和/或方差;
角度拓展的范围或角度拓展的均值和/或方差;
第一测量信息的范围;
第一测量信息的均值和/或方差。
可选地,所述第二通信设备包括以下至少一项:
终端;
模型管理设备;
网络侧设备。
本实施例的方法,其具体实现过程与技术效果与第一通信设备侧方法实施例中类似,具体可以参见第一通信设备侧方法实施例中的详细介绍,此处不再赘述。
本申请实施例提供的基于AI模型的定位方法,执行主体可以为基于AI模型的定位装置。本申请实施例中以基于AI模型的定位装置执行基于AI模型的定位方法为例,说明本申请实施例提供的基于AI模型的定位装置。
图13是本申请提供的基于AI模型的定位装置的结构示意图之一。如图13所示,本实施例提供的基于AI模型的定位装置,包括:
获取模块210,用于获取与AI模型相关信息关联的第一信息;
处理模块220,用于根据所述第一信息,确定目标信息,所述目标信息包括以下至少一项:目标AI模型、所述AI模型相关信息的有效性信息或者基于所述目标AI模型进行定位得到的反馈信息;
所述第一信息用于表示所述AI模型相关信息的有效适用范围,所述AI模型相关信息包括以下至少一项:AI模型、AI模型参数、AI模型的输入、AI模型的输出。
可选地,所述获取模块210,还用于:
获取目标终端的第二信息;所述第二信息用于表示所述目标终端获取的定位相关信息;
所述处理模块220,具体用于:
根据所述第一信息和所述第二信息,确定所述目标信息。
可选地,所述第一信息包括以下至少一项:小区信息;区域信息;有效时间信息;场景信息;信干噪比SINR范围;
其中,所述小区信息包括以下至少一项:
一个或多个小区的识别信息;
一个或多个基站的识别信息;
一个或多个传输接收点TRP的识别信息;
小区列表信息;
小区频域范围信息;
其中,所述区域信息包括以下至少一项:
区域识别信息;距离范围信息;距离范围对应的参考点信息;
其中,所述有效时间信息包括以下至少一项:
计时器时长;
计时器起始时间;
其中,所述场景信息包括以下至少一项:
视距LOS场景;非视距NLOS场景;复杂场景;室内场景;室外场景。
可选地,所述第二信息包括以下至少一项:所述目标终端的位置信息、小区信息、区域信息,计时器信息、场景信息和所述目标终端测量的SINR;
其中,所述小区信息为所述目标终端的服务小区、参考小区或者参考信号接收功率RSRP最强小区的以下至少一项信息,所述至少一项信息包括:识别信息、频域信息;
其中,所述区域信息为所述目标终端的区域识别信息;
其中,所述场景信息为所述目标终端所处的场景信息。
可选地,所述处理模块220,具体用于:
所述第一通信设备根据所述第二信息中参数的取值以及所述第一信息中对应参数的范围,确定所述目标信息。
可选地,在所述第二信息中参数的取值处于所述第一信息中对应参数的范围内的情况下,所述AI模型相关信息有效。
可选地,在所述第二信息中参数的取值处于所述第一信息中对应参数的范围内的情况下,所述反馈信息中包括所述第二信息;或者,
在所述第二信息中参数的取值不处于所述第一信息中对应参数的范围内的情况下,所述反馈信息中包括所述第二信息。
可选地,若所述第一通信设备确定所述目标终端的小区信息不属于所述第一信息中的小区信息的范围内,所述AI模型相关信息失效。
可选地,若所述第一通信设备确定所述目标终端的小区信息不属于所述第一信息中的小区信息的范围内,所述反馈信息包括所述目标终端的小区信息。
可选地,若所述第一通信设备确定所述目标终端的计时器超时,所述AI模型相关信息失效或超时。
可选地,所述计时器满足以下至少一种情况:
所述计时器的时长为第一信息中的计时器时长;
所述计时器从所述第一信息中计时器起始时间开始计时;
在所述AI模型和/或AI模型参数更新的情况下,所述计时器重新计时。
可选地,若所述第一通信设备确定所述目标终端的场景信息不属于所述第一信息中的场景信息的范围内,所述AI模型相关信息失效。
可选地,若所述第一通信设备确定所述目标终端的位置不处于所述第一信息中的小区信息和/或区域信息对应的位置范围内,所述AI模型相关信息失效。
可选地,在所述第二信息中参数的取值处于所述第一信息中对应参数的第一范围内的情况下,所述目标AI模型为所述第一信息的第一范围对应的AI模型。
可选地,获取模块310还用于:
接收多个预配置的AI模型和/或AI模型参数,以及所述AI模型和/或AI模型参数对应的第一信息。
可选地,获取模块310还用于:
所述第一通信设备根据所述第一信息的第一范围,从多个所述预配置的AI模型中,获取所述第一信息的第一范围对应的AI模型。
可选地,所述反馈信息包括:所述目标AI模型。
可选地,所述反馈信息为根据所述有效性信息确定的,包括:
若所述有效性信息用于指示所述AI模型相关信息有效,所述反馈信息至少包括所述AI模型的输出;
若所述有效性信息用于指示所述AI模型相关信息失效,所述反馈信息包括以下至少一项:误差原因;第二信息;有效性信息;AI模型请求;AI模型更新请求;数据收集请求。
可选地,所述反馈信息包括以下至少一项:所述有效性信息、第二信息、第一测量信息和所述AI模型的输出;
第一测量信息包括以下至少一项:
信号测量信息;位置信息;误差值;信道冲击响应CIR信息;功率时延谱PDP信息。
可选地,所述第二信息包括所述目标终端获得的第二测量信息,所述处理模块220,具体用于:
在所述第二测量信息为一次测量得到的测量信息的情况下,所述第一通信设备根据所述第二测量信息中参数的取值以及所述第一信息中对应参数的范围,确定所述目标信息;
在所述第二测量信息为多次测量得到的测量信息的情况下,所述第一通信设备根据所述第二测量信息中参数的分布与所述第一信息中对应参数的分布的一致性,确定所述目标信息。
可选地,在所述第二测量信息中参数的取值处于所述第一信息中对应参数的范围内的情况下,所述AI模型相关信息有效;
在所述第二测量信息中至少一个参数的取值处于所述第一信息中对应参数的范围外的情况下,所述AI模型相关信息失效。
可选地,在所述第二测量信息中参数的分布与所述第一信息中对应参数的分布一致的情况下,所述AI模型相关信息有效;
在所述第二测量信息中至少一个参数的分布与所述第一信息中对应参数的分布 不一致的情况下,所述AI模型相关信息失效。
可选地,在所述第二测量信息中至少一个参数的分布与所述第一信息中对应参数的分布不一致的情况下,所述AI模型相关信息失效,包括:
在所述第二测量信息中参数的分布与所述第一信息中对应参数的分布不满足于条件阈值信息的情况下,所述AI模型相关信息失效。
可选地,所述条件阈值信息包括以下至少一项:
正态分布中σ原则、2σ原则或3σ原则对应的绝对时间;
σ原则、2σ原则或3σ原则对应的时延拓展值;
σ原则、2σ原则或3σ原则对应的角度拓展值;
所述第二测量信息中参数的均值和所述第一信息中对应参数的均值的最大差;
所述第二测量信息中参数的均值和所述第一信息中对应参数的均值的最大方差;
所述第二测量信息中参数的均值和所述第一信息中对应参数的方差的最大差;
所述第二测量信息中参数的均值和所述第一信息中对应参数的方差的最大方差。
可选地,所述第二测量信息包括以下至少一项:
信干噪比SINR范围、噪声值、NLOS引入的绝对时间值、时延拓展值、角度拓展值、SINR均值和方差、噪声均值和方差、NLOS引入的绝对时间均值和方差、时延拓展均值和方差、角度拓展均值和方差;
其中,所述SINR范围和所述SINR均值和方差是根据以下至少一项信息的SINR得到的;
噪声值和噪声均值和方差是以下至少一项信息的噪声值得到的;
所述至少一项信息包括:测量信道、测量信号或第一测量信息。
可选地,所述第一测量信息包括以下至少一项:
信号测量信息;位置信息;误差值;信道冲击响应CIR信息;功率时延谱PDP信息。
可选地,信号测量信息包括以下至少一项:
参考信号时间差RSTD测量结果、往返时延测量结果、到达角AOA测量结果、离开角AOD测量结果、参考信息接收功率RSRP、多径测量信息、视距LOS指示信息;
所述多径测量信息包括以下至少一项:
首径的功率、首径的时延、首径的到达时间TOA、首径的参考信号时间差RSTD、首径的天线子载波相位差、首径的天线子载波相位、多径的功率、多径的时延、TOA、多径的RSTD、多径的天线子载波相位差或多径的天线子载波相位。
可选地,所述第二测量信息还包括所述第一测量信息。
可选地,所述反馈信息包括以下至少一项:所述有效性信息、第二测量信息、AI模型的输入、AI模型的输出、AI模型识别信息、AI模型更新请求。
可选地,所述第一信息包括以下至少一项:
SINR范围;
噪声的范围;
噪声分布的均值和/或方差;
NLOS引入的绝对时间范围或NLOS引入绝对时间的均值和/或方差;
时延拓展的范围或时延拓展的均值和/或方差;
角度拓展的范围或角度拓展的均值和/或方差;
第一测量信息的范围;
第一测量信息的均值和/或方差。
可选地,第一信息为根据AI模型的测试集和验证集获取到的,包括以下至少一种情况:
所述第一信息是根据AI模型的测试集和验证集的输入数据获取的特征信息;
所述第一信息是根据AI模型的测试集和验证集的输出数据获取的特征信息;
所述第一信息是根据AI模型的测试集和验证集的输入数据和输出数据获取的特征信息。
可选地,所述第二测量信息是根据第一测量信息获得的特征信息;和/或,
所述第二测量信息是根据所述AI模型的输出获得的特征信息。
可选地,所述NLOS引入的绝对时间为所述AI模型获取的时间信息减去非AI模型获取的时间信息;
所述时延拓展、角度拓展、噪声分布是根据第一测量信息或所述AI模型的输出获取的特征信息。
可选地,所述AI模型的输入和/或AI模型的输出,包括以下至少一项:
所述目标终端的定位信号测量信息;所述目标终端的位置信息;误差信息;信道冲激响应CIR;首径的功率;首径的时延;首径的到达时间TOA;首径的参考信号时间差RSTD;首径的到达角;首径的天线子载波相位差;多径的功率;多径的时延;多径的TOA;多径的RSTD;多径的到达角;多径的天线子载波相位差;平均过量时延;均方根时延拓展;相干带宽;
所述误差信息包括以下至少一项:位置误差值、测量误差值、AI模型误差值或AI模型参数误差值。
可选地,所述AI模型更新请求,包括以下至少一项:
AI模型的标识ID、满足第二测量信息分布的AI模型和/或AI模型参数、AI模型和/或AI模型参数。
可选地,所述AI模型包括以下至少一项:
AI模型类型;
AI模型结构。
可选地,所述AI模型包括以下至少一项:
神经网络列表,包括以下至少一项:每个神经网络的神经元类型、每个神经网络的神经元权重和偏差;
超参数信息;
损失函数信息。
可选地,所述AI参数包括以下至少一项:
超参数信息;
AI模型描述参数信息;
AI模型的权值信息;
AI模型的初始参数。
可选地,有效性信息可以包括以下至少一项:
有效性指示信息,用于指示所述AI模型相关信息是否有效;
有效程度;
有效等级;
失效原因;
可靠性指示信息,用于指示基于所述目标AI模型进行定位得到的结果是否可靠;
可靠性程度;
可靠性等级。
可选地,所述获取模块310还用于:
接收更新的AI模型和/或AI模型参数,以及与所述AI模型和/或AI模型参数对应的第一信息。
可选地,所述位置信息是通过以下至少一种方式得到的,所述至少一种方式包括:
到达时间差定位法OTDOA、全球导航卫星***GNSS、下行到达时间差、上行到达时间差、上行蓝牙到达角AoA、蓝牙出发角AoD、蓝牙、传感器或无线保真WIFI。
可选地,所述第一通信设备包括以下至少一项:
位置管理功能LMF网元和所述LMF网元的演进设备;
定位服务器;
网络数据分析功能NWADF网元;
网络侧设备;
终端;
监控设备Actor。
本实施例的装置,可以用于执行前述终端侧方法实施例中任一实施例的方法,其具体实现过程与技术效果与终端侧方法实施例中类似,具体可以参见终端侧方法实施例中的详细介绍,此处不再赘述。
图14是本申请提供的基于AI模型的定位装置的结构示意图之二。如图14所示,本实施例提供的基于AI模型的定位装置,包括:
接收模块310,用于接收第一通信设备发送的目标信息,所述目标信息包括以下至少一项:目标AI模型、所述AI模型相关信息的有效性信息或者基于所述目标AI模型进行定位得到的反馈信息;
所述目标信息为根据AI模型相关信息关联的第一信息确定的,所述第一信息用于表示所述AI模型相关信息的有效适用范围,所述AI模型相关信息包括以下至少一项:AI模型、AI模型参数、AI模型的输入、AI模型的输出。
可选地,还包括:
发送模块320,用于向所述第一通信设备发送所述AI模型相关信息和所述AI模型相关信息关联的第一信息。
可选地,所述发送模块320,还用于:
向第一通信设备发送多个预配置的AI模型和/或AI模型参数,以及所述AI模型和/或AI模型参数对应的第一信息。
可选地,所述发送模块320,还用于:
向所述第一通信设备发送第二信息,所述第二信息用于表示所述终端获取的定位相关信息。
可选地,所述第一信息包括以下至少一项:小区信息;区域信息;有效时间信息;场景信息;信干噪比SINR范围;
其中,所述小区信息包括以下至少一项:
一个或多个小区的识别信息;
一个或多个基站的识别信息;
一个或多个传输接收点TRP的识别信息;
小区列表信息;
小区频域范围信息;
其中,所述区域信息包括以下至少一项:
区域识别信息;距离范围信息;距离范围对应的参考点信息;
其中,所述有效时间信息包括以下至少一项:
计时器时长;
计时器起始时间;
其中,所述场景信息包括以下至少一项:
视距LOS场景;非视距NLOS场景;复杂场景;室内场景;室外场景。
可选地,所述第二信息包括以下至少一项:所述目标终端的位置信息、小区信息、区域信息,计时器信息、场景信息和所述目标终端测量的SINR;
其中,所述小区信息为所述目标终端的服务小区、参考小区或者参考信号接收功率RSRP最强小区的以下至少一项信息,所述至少一项信息包括:识别信息、频域信息;
其中,所述区域信息为所述目标终端的区域识别信息;
其中,所述场景信息为所述目标终端所处的场景信息。
可选地,在所述第二信息中参数的取值处于所述第一信息中对应参数的范围内的情况下,所述AI模型相关信息有效。
可选地,若第二信息中参数的取值处于所述第一信息中对应参数的范围外的情况下,所述AI模型相关信息失效。
可选地,所述反馈信息包括以下至少一项:所述有效性信息、第二信息、第一测量信息和所述AI模型的输出;
第一测量信息包括以下至少一项:
信号测量信息;位置信息;误差值;信道冲击响应CIR信息;功率时延谱PDP信息。
可选地,所述第二信息包括所述目标终端获得的第二测量信息,在所述第二测量信息中参数的取值处于所述第一信息中对应参数的范围内的情况下,所述AI模型相关信息有效;和/或;
在所述第二测量信息中至少一个参数的取值处于所述第一信息中对应参数的范围外的情况下,所述AI模型相关信息失效。
可选地,所述第二信息包括所述目标终端获得的第二测量信息,在所述第二测量信息中参数的分布与所述第一信息中对应参数的分布一致的情况下,所述AI模型相关信息有效;和/或;
在所述第二测量信息中至少一个参数的分布与所述第一信息中对应参数的分布不一致的情况下,所述AI模型相关信息失效。
可选地,在所述第二测量信息中至少一个参数的分布与所述第一信息中对应参数的分布不一致的情况下,所述AI模型相关信息失效,包括:
在所述第二测量信息中参数的分布与所述第一信息中对应参数的分布不满足于条件阈值信息的情况下,所述AI模型相关信息失效。
可选地,所述条件阈值信息包括以下至少一项:
正态分布中σ原则、2σ原则或3σ原则对应的绝对时间;
σ原则、2σ原则或3σ原则对应的时延拓展值;
σ原则、2σ原则或3σ原则对应的角度拓展值;
所述第二测量信息中参数的均值和所述第一信息中对应参数的均值的最大差;
所述第二测量信息中参数的均值和所述第一信息中对应参数的均值的最大方差;
所述第二测量信息中参数的均值和所述第一信息中对应参数的方差的最大差;
所述第二测量信息中参数的均值和所述第一信息中对应参数的方差的最大方差。
可选地,所述第二测量信息包括以下至少一项:
信干噪比SINR范围、噪声值、NLOS引入的绝对时间值、时延拓展值、角度拓展值、SINR均值和方差、噪声均值和方差、NLOS引入的绝对时间均值和方差、时延拓展均值和方差、角度拓展均值和方差;
其中,所述SINR范围和所述SINR均值和方差是根据以下至少一项信息的SINR得到的;
噪声值和噪声均值和方差是以下至少一项信息的噪声值得到的;
所述至少一项信息包括:测量信道、测量信号或第一测量信息。
可选地,所述第一测量信息包括以下至少一项:
信号测量信息;位置信息;误差值;信道冲击响应CIR信息;功率时延谱PDP信息。
可选地,所述反馈信息包括以下至少一项:所述有效性信息、第二测量信息、AI模型的输入、AI模型的输出、AI模型识别信息、AI模型更新请求。
可选地,所述第一信息包括以下至少一项:
SINR范围;
噪声的范围;
噪声分布的均值和/或方差;
NLOS引入的绝对时间范围或NLOS引入绝对时间的均值和/或方差;
时延拓展的范围或时延拓展的均值和/或方差;
角度拓展的范围或角度拓展的均值和/或方差;
第一测量信息的范围;
第一测量信息的均值和/或方差。
可选地,所述第二通信设备包括以下至少一项:
终端;
模型管理设备;
网络侧设备。
本实施例的装置,可以用于执行前述网络侧方法实施例中任一实施例的方法,其具体实现过程与技术效果与网络侧方法实施例中类似,具体可以参见网络侧方法实施例中的详细介绍,此处不再赘述。
本申请实施例中的基于AI模型的定位装置可以是电子设备,例如具有操作***的电子设备,也可以是电子设备中的部件,例如集成电路或芯片。该电子设备可以是终端,也可以为除终端之外的其他设备。示例性的,终端可以包括但不限于上述所列举的终端11的类型,其他设备可以为服务器、网络附属存储器(Network Attached Storage,NAS)等,本申请实施例不作具体限定。
本申请实施例提供的基于AI模型的定位装置能够实现图2至图12的方法实施例实现的各个过程,并达到相同的技术效果,为避免重复,这里不再赘述。
可选的,如图15所示,本申请实施例还提供一种通信设备1500,包括处理器1501和存储器1502,存储器1502上存储有可在所述处理器1501上运行的程序或指令,例如,该通信设备1500为终端时,该程序或指令被处理器1501执行时实现上述基于AI模型的定位方法实施例的各个步骤,且能达到相同的技术效果。该通信设备1500为网络侧设备时,该程序或指令被处理器1501执行时实现上述基于AI模型的定位方法实施例的各个步骤,且能达到相同的技术效果,为避免重复,这里不再赘述。
本申请实施例还提供一种终端,包括处理器和通信接口,述通信接口用于获取与AI模型相关信息关联的第一信息;所述处理器用于根据所述第一信息,确定目标信息,所述目标信息包括以下至少一项:目标AI模型、所述AI模型相关信息的有效性信息或者基于所述目标AI模型进行定位得到的反馈信息;所述第一信息用于表示所述AI模型相关信息的有效适用范围,所述AI模型相关信息包括以下至少一项:AI模型、AI模型参数、AI模型的输入、AI模型的输出。该终端实施例与上述终端侧方法实施例对应,上述方法实施例的各个实施过程和实现方式均可适用于该终端实施例中,且能达到相同的技术效果。具体地,图16为实现本申请实施例的一种终端的硬件结构示意图。
该终端1000包括但不限于:射频单元1001、网络模块1002、音频输出单元1003、输入单元1004、传感器1005、显示单元1006、用户输入单元1007、接口单元1008、存储器1009、以及处理器1010等中的至少部分部件。
本领域技术人员可以理解,终端1000还可以包括给各个部件供电的电源(比如电池),电源可以通过电源管理***与处理器1010逻辑相连,从而通过电源管理***实现管理充电、放电、以及功耗管理等功能。图16中示出的终端结构并不构成对终端的限定,终端可以包括比图示更多或更少的部件,或者组合某些部件,或者不同的部件布置,在此不再赘述。
应理解的是,本申请实施例中,输入单元1004可以包括图形处理单元(Graphics Processing Unit,GPU)10041和麦克风10042,图形处理器10041对在视频捕获模式或图像捕获模式中由图像捕获装置(如摄像头)获得的静态图片或视频的图像数据进行处理。显示单元1006可包括显示面板10061,可以采用液晶显示 器、有机发光二极管等形式来配置显示面板10061。用户输入单元1007包括触控面板10071以及其它输入设备10072中的至少一种。触控面板10071,也称为触摸屏。触控面板10071可包括触摸检测装置和触摸控制器两个部分。其它输入设备10072可以包括但不限于物理键盘、功能键(比如音量控制按键、开关按键等)、轨迹球、鼠标、操作杆,在此不再赘述。
本申请实施例中,射频单元1001将接收来自网络侧设备的下行数据接收后,可以传输给处理器1010进行处理;另外,射频单元1001可以将上行的数据发送给向网络侧设备发送上行数据。通常,射频单元1001包括但不限于天线、至少一个放大器、收发信机、耦合器、低噪声放大器、双工器等。
存储器1009可用于存储软件程序或指令以及各种数据。存储器1009可主要包括存储程序或指令的第一存储区和存储数据的第二存储区,其中,第一存储程序或指令区可存储操作***、至少一个功能所需的应用程序或指令(比如声音播放功能、图像播放功能等)等。此外,存储器1009可以包括易失性存储器或非易失性存储器,或者,存储器1009可以包括易失性和非易失性存储器两者。包括高速随机存取存储器,还可以包括非易失性存储器,其中,非易失性存储器可以是只读存储器(Read-Only Memory,ROM)、可编程只读存储器(Programmable ROM,PROM)、可擦除可编程只读存储器(Erasable PROM,EPROM)、电可擦除可编程只读存储器(Electrically EPROM,EEPROM)或闪存。易失性存储器可以是随机存取存储器(Random Access Memory,RAM),静态随机存取存储器(Static RAM,SRAM)、动态随机存取存储器(Dynamic RAM,DRAM)、同步动态随机存取存储器(Synchronous DRAM,SDRAM)、双倍数据速率同步动态随机存取存储器(Double Data Rate SDRAM,DDRSDRAM)、增强型同步动态随机存取存储器(Enhanced SDRAM,ESDRAM)、同步连接动态随机存取存储器(Synch link DRAM,SLDRAM)和直接内存总线随机存取存储器(Direct Rambus RAM,DRRAM)。本申请实施例中的存储器1009包括但不限于这些和任意其它适合类型的存储器例如至少一个磁盘存储器件、闪存器件、或其它非易失性固态存储器件。
处理器1010可包括一个或多个处理单元;可选的,处理器1010可集成应用处理器和调制解调处理器,其中,应用处理器主要处理涉及操作***、用户界面和应用程序或指令等的操作,调制解调处理器主要处理无线通信信号,如基带处理器。可以理解的是,上述调制解调处理器也可以不集成到处理器1010中。
其中,射频单元1001,用于获取与AI模型相关信息关联的第一信息;
处理器1010,用于根据所述第一信息,确定目标信息,所述目标信息包括以下至少一项:目标AI模型、所述AI模型相关信息的有效性信息或者基于所述目标AI模型进行定位得到的反馈信息;
所述第一信息用于表示所述AI模型相关信息的有效适用范围,所述AI模型相 关信息包括以下至少一项:AI模型、AI模型参数、AI模型的输入、AI模型的输出。
上述实施方式中,射频单元获取与AI模型相关信息关联的第一信息;处理器根据第一信息,确定目标信息,目标信息包括以下至少一项:目标AI模型、AI模型相关信息的有效性信息或者基于目标AI模型进行定位得到的反馈信息,由于第一信息用于表示AI模型相关信息的有效适用范围,基于第一信息确定目标信息,可以使得定位的AI模型更加符合实际场景需求,还可以基于确定的目标信息更新AI模型和/或AI模型参数,使得基于更新后的AI模型得到定位结果准确性更高。
可选地,所述射频单元1001,还用于:
获取目标终端的第二信息;所述第二信息用于表示所述目标终端获取的定位相关信息;
所述处理器1010,具体用于:
根据所述第一信息和所述第二信息,确定所述目标信息。
可选地,所述第一信息包括以下至少一项:小区信息;区域信息;有效时间信息;场景信息;信干噪比SINR范围;
其中,所述小区信息包括以下至少一项:
一个或多个小区的识别信息;
一个或多个基站的识别信息;
一个或多个传输接收点TRP的识别信息;
小区列表信息;
小区频域范围信息;
其中,所述区域信息包括以下至少一项:
区域识别信息;距离范围信息;距离范围对应的参考点信息;
其中,所述有效时间信息包括以下至少一项:
计时器时长;
计时器起始时间;
其中,所述场景信息包括以下至少一项:
视距LOS场景;非视距NLOS场景;复杂场景;室内场景;室外场景。
可选地,所述第二信息包括以下至少一项:所述目标终端的位置信息、小区信息、区域信息,计时器信息、场景信息和所述目标终端测量的SINR;
其中,所述小区信息为所述目标终端的服务小区、参考小区或者参考信号接收功率RSRP最强小区的以下至少一项信息,所述至少一项信息包括:识别信息、频域信息;
其中,所述区域信息为所述目标终端的区域识别信息;
其中,所述场景信息为所述目标终端所处的场景信息。
可选地,所述处理器1010,具体用于:
所述第一通信设备根据所述第二信息中参数的取值以及所述第一信息中对应参数的范围,确定所述目标信息。
可选地,在所述第二信息中参数的取值处于所述第一信息中对应参数的范围内的情况下,所述AI模型相关信息有效。
可选地,在所述第二信息中参数的取值处于所述第一信息中对应参数的范围内的情况下,所述反馈信息中包括所述第二信息;或者,
在所述第二信息中参数的取值不处于所述第一信息中对应参数的范围内的情况下,所述反馈信息中包括所述第二信息。
可选地,若所述第一通信设备确定所述目标终端的小区信息不属于所述第一信息中的小区信息的范围内,所述AI模型相关信息失效。
可选地,若所述第一通信设备确定所述目标终端的小区信息不属于所述第一信息中的小区信息的范围内,所述反馈信息包括所述目标终端的小区信息。
可选地,若所述第一通信设备确定所述目标终端的计时器超时,所述AI模型相关信息失效或超时。
可选地,所述计时器满足以下至少一种情况:
所述计时器的时长为第一信息中的计时器时长;
所述计时器从所述第一信息中计时器起始时间开始计时;
在所述AI模型和/或AI模型参数更新的情况下,所述计时器重新计时。
可选地,若所述第一通信设备确定所述目标终端的场景信息不属于所述第一信息中的场景信息的范围内,所述AI模型相关信息失效。
可选地,若所述第一通信设备确定所述目标终端的位置不处于所述第一信息中的小区信息和/或区域信息对应的位置范围内,所述AI模型相关信息失效。
可选地,在所述第二信息中参数的取值处于所述第一信息中对应参数的第一范围内的情况下,所述目标AI模型为所述第一信息的第一范围对应的AI模型。
可选地,射频单元1001还用于:
接收多个预配置的AI模型和/或AI模型参数,以及所述AI模型和/或AI模型参数对应的第一信息。
可选地,射频单元1001还用于:
所述第一通信设备根据所述第一信息的第一范围,从多个所述预配置的AI模型中,获取所述第一信息的第一范围对应的AI模型。
可选地,所述反馈信息包括:所述目标AI模型。
可选地,所述反馈信息为根据所述有效性信息确定的,包括:
若所述有效性信息用于指示所述AI模型相关信息有效,所述反馈信息至少包括所述AI模型的输出;
若所述有效性信息用于指示所述AI模型相关信息失效,所述反馈信息包括以下至少一项:误差原因;第二信息;有效性信息;AI模型请求;AI模型更新请求;数据收集请求。
可选地,所述反馈信息包括以下至少一项:所述有效性信息、第二信息、第一测量信息和所述AI模型的输出;
第一测量信息包括以下至少一项:
信号测量信息;位置信息;误差值;信道冲击响应CIR信息;功率时延谱PDP信息。
可选地,所述第二信息包括所述目标终端获得的第二测量信息,所述处理器1010,具体用于:
在所述第二测量信息为一次测量得到的测量信息的情况下,所述第一通信设备根据所述第二测量信息中参数的取值以及所述第一信息中对应参数的范围,确定所述目标信息;
在所述第二测量信息为多次测量得到的测量信息的情况下,所述第一通信设备根据所述第二测量信息中参数的分布与所述第一信息中对应参数的分布的一致性,确定所述目标信息。
可选地,在所述第二测量信息中参数的取值处于所述第一信息中对应参数的范围内的情况下,所述AI模型相关信息有效;
在所述第二测量信息中至少一个参数的取值处于所述第一信息中对应参数的范围外的情况下,所述AI模型相关信息失效。
可选地,在所述第二测量信息中参数的分布与所述第一信息中对应参数的分布一致的情况下,所述AI模型相关信息有效;
在所述第二测量信息中至少一个参数的分布与所述第一信息中对应参数的分布不一致的情况下,所述AI模型相关信息失效。
可选地,在所述第二测量信息中至少一个参数的分布与所述第一信息中对应参数的分布不一致的情况下,所述AI模型相关信息失效,包括:
在所述第二测量信息中参数的分布与所述第一信息中对应参数的分布不满足于条件阈值信息的情况下,所述AI模型相关信息失效。
可选地,所述条件阈值信息包括以下至少一项:
正态分布中σ原则、2σ原则或3σ原则对应的绝对时间;
σ原则、2σ原则或3σ原则对应的时延拓展值;
σ原则、2σ原则或3σ原则对应的角度拓展值;
所述第二测量信息中参数的均值和所述第一信息中对应参数的均值的最大差;
所述第二测量信息中参数的均值和所述第一信息中对应参数的均值的最大方差;
所述第二测量信息中参数的均值和所述第一信息中对应参数的方差的最大差;
所述第二测量信息中参数的均值和所述第一信息中对应参数的方差的最大方差。
可选地,所述第二测量信息包括以下至少一项:
信干噪比SINR范围、噪声值、NLOS引入的绝对时间值、时延拓展值、角度拓展值、SINR均值和方差、噪声均值和方差、NLOS引入的绝对时间均值和方差、时延拓展均值和方差、角度拓展均值和方差;
其中,所述SINR范围和所述SINR均值和方差是根据以下至少一项信息的SINR得到的;
噪声值和噪声均值和方差是以下至少一项信息的噪声值得到的;
所述至少一项信息包括:测量信道、测量信号或第一测量信息。
可选地,所述第一测量信息包括以下至少一项:
信号测量信息;位置信息;误差值;信道冲击响应CIR信息;功率时延谱PDP信息。
可选地,信号测量信息包括以下至少一项:
参考信号时间差RSTD测量结果、往返时延测量结果、到达角AOA测量结果、离开角AOD测量结果、参考信息接收功率RSRP、多径测量信息、视距LOS指示信息;
所述多径测量信息包括以下至少一项:
首径的功率、首径的时延、首径的到达时间TOA、首径的参考信号时间差RSTD、首径的天线子载波相位差、首径的天线子载波相位、多径的功率、多径的时延、TOA、多径的RSTD、多径的天线子载波相位差或多径的天线子载波相位。
可选地,所述第二测量信息还包括所述第一测量信息。
可选地,所述反馈信息包括以下至少一项:所述有效性信息、第二测量信息、AI模型的输入、AI模型的输出、AI模型识别信息、AI模型更新请求。
可选地,所述第一信息包括以下至少一项:
SINR范围;
噪声的范围;
噪声分布的均值和/或方差;
NLOS引入的绝对时间范围或NLOS引入绝对时间的均值和/或方差;
时延拓展的范围或时延拓展的均值和/或方差;
角度拓展的范围或角度拓展的均值和/或方差;
第一测量信息的范围;
第一测量信息的均值和/或方差。
可选地,第一信息为根据AI模型的测试集和验证集获取到的,包括以下至少一 种情况:
所述第一信息是根据AI模型的测试集和验证集的输入数据获取的特征信息;
所述第一信息是根据AI模型的测试集和验证集的输出数据获取的特征信息;
所述第一信息是根据AI模型的测试集和验证集的输入数据和输出数据获取的特征信息。
可选地,所述第二测量信息是根据第一测量信息获得的特征信息;和/或,
所述第二测量信息是根据所述AI模型的输出获得的特征信息。
可选地,所述NLOS引入的绝对时间为所述AI模型获取的时间信息减去非AI模型获取的时间信息;
所述时延拓展、角度拓展、噪声分布是根据第一测量信息或所述AI模型的输出获取的特征信息。
可选地,所述AI模型的输入和/或AI模型的输出,包括以下至少一项:
所述目标终端的定位信号测量信息;所述目标终端的位置信息;误差信息;信道冲激响应CIR;首径的功率;首径的时延;首径的到达时间TOA;首径的参考信号时间差RSTD;首径的到达角;首径的天线子载波相位差;多径的功率;多径的时延;多径的TOA;多径的RSTD;多径的到达角;多径的天线子载波相位差;平均过量时延;均方根时延拓展;相干带宽;
所述误差信息包括以下至少一项:位置误差值、测量误差值、AI模型误差值或AI模型参数误差值。
可选地,所述AI模型更新请求,包括以下至少一项:
AI模型的标识ID、满足第二测量信息分布的AI模型和/或AI模型参数、AI模型和/或AI模型参数。
可选地,所述AI模型包括以下至少一项:
AI模型类型;
AI模型结构。
可选地,所述AI模型包括以下至少一项:
神经网络列表,包括以下至少一项:每个神经网络的神经元类型、每个神经网络的神经元权重和偏差;
超参数信息;
损失函数信息。
可选地,所述AI参数包括以下至少一项:
超参数信息;
AI模型描述参数信息;
AI模型的权值信息;
AI模型的初始参数。
可选地,有效性信息可以包括以下至少一项:
有效性指示信息,用于指示所述AI模型相关信息是否有效;
有效程度;
有效等级;
失效原因;
可靠性指示信息,用于指示基于所述目标AI模型进行定位得到的结果是否可靠;
可靠性程度;
可靠性等级。
可选地,所述射频单元1001还用于:
接收更新的AI模型和/或AI模型参数,以及与所述AI模型和/或AI模型参数对应的第一信息。
可选地,所述位置信息是通过以下至少一种方式得到的,所述至少一种方式包括:
到达时间差定位法OTDOA、全球导航卫星***GNSS、下行到达时间差、上行到达时间差、上行蓝牙到达角AoA、蓝牙出发角AoD、蓝牙、传感器或无线保真WIFI。
可选地,所述第一通信设备包括以下至少一项:
位置管理功能LMF网元和所述LMF网元的演进设备;
定位服务器;
网络数据分析功能NWADF网元;
其它网络侧设备;
终端;
监控设备Actor。
本申请实施例还提供一种网络侧设备,包括处理器和通信接口,通信接口用于获取与AI模型相关信息关联的第一信息;所述处理器用于根据所述第一信息,确定目标信息,所述目标信息包括以下至少一项:目标AI模型、所述AI模型相关信息的有效性信息或者基于所述目标AI模型进行定位得到的反馈信息;所述第一信息用于表示所述AI模型相关信息的有效适用范围,所述AI模型相关信息包括以下至少一项:AI模型、AI模型参数、AI模型的输入、AI模型的输出。该网络侧设备实施例与上述第一通信设备侧或第二通信设备侧方法实施例对应,上述方法实施例的各个实施过程和实现方式均可适用于该网络侧设备实施例中,且能达到相同的技术效果。
具体地,本申请实施例还提供了一种网络侧设备。如图17所示,该网络侧设备700包括:天线71、射频装置72、基带装置73、处理器75和存储器75。天线71与 射频装置72连接。在上行方向上,射频装置72通过天线71接收信息,将接收的信息发送给基带装置73进行处理。在下行方向上,基带装置73对要发送的信息进行处理,并发送给射频装置72,射频装置72对收到的信息进行处理后经过天线71发送出去。
上述频带处理装置可以位于基带装置73中,以上实施例中网络侧设备执行的方法可以在基带装置73中实现,该基带装置73包括基带处理器75和存储器75。
基带装置73例如可以包括至少一个基带板,该基带板上设置有多个芯片,如图17所示,其中一个芯片例如为基带处理器75,通过总线接口与存储器75连接,以调用存储器75中的程序,执行以上方法实施例中所示的网络设备操作。
该基带装置73网络侧设备还可以包括网络接口76,用于与射频装置72交互信息,该接口例如为通用公共无线接口(common public radio interface,简称CPRI)。
具体地,本发明实施例的网络侧设备700还包括:存储在存储器75上并可在处理器75上运行的指令或程序,处理器75调用存储器75中的指令或程序执行图13或图14所示各模块执行的方法,并达到相同的技术效果,为避免重复,故不在此赘述。
本申请实施例还提供了一种网络侧设备。如图18所示,该网络侧设备700包括:处理器701、网络接口702和存储器703。其中,网络接口702例如为通用公共无线接口(common public radio interface,CPRI)。
具体地,本申请实施例的网络侧设备700还包括:存储在存储器703上并可在处理器701上运行的指令或程序,处理器701调用存储器703中的指令或程序执行图13或14所示各模块执行的方法,并达到相同的技术效果,为避免重复,故不在此赘述。
其中,第一通信设备和/或第二通信设备可以通过上述网络侧设备实施例实现。
本申请实施例还提供一种可读存储介质,所述可读存储介质上存储有程序或指令,该程序或指令被处理器执行时实现上述基于AI模型的定位方法实施例的各个过程,且能达到相同的技术效果,为避免重复,这里不再赘述。
其中,所述处理器为上述实施例中所述的终端中的处理器。所述可读存储介质,包括计算机可读存储介质,如计算机只读存储器ROM、随机存取存储器RAM、磁碟或者光盘等。
本申请实施例另提供了一种芯片,所述芯片包括处理器和通信接口,所述通信接口和所述处理器耦合,所述处理器用于运行程序或指令,实现上述基于AI模型的定位方法实施例的各个过程,且能达到相同的技术效果,为避免重复,这里不再赘述。
应理解,本申请实施例提到的芯片还可以称为***级芯片,***芯片,芯片***或片上***芯片等。
本申请实施例另提供了一种计算机程序/程序产品,所述计算机程序/程序产品被存储在存储介质中,所述计算机程序/程序产品被至少一个处理器执行以实现上述基于AI模型的定位方法实施例的各个过程,且能达到相同的技术效果,为避免重复,这里不再赘述。
本申请实施例还提供了一种通信***,包括:终端及网络侧设备,所述终端可用于执行如上所述的基于AI模型的定位方法的步骤,所述网络侧设备可用于执行如上所述的基于AI模型的定位方法的步骤。
需要说明的是,在本文中,术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含,从而使得包括一系列要素的过程、方法、物品或者装置不仅包括那些要素,而且还包括没有明确列出的其他要素,或者是还包括为这种过程、方法、物品或者装置所固有的要素。在没有更多限制的情况下,由语句“包括一个......”限定的要素,并不排除在包括该要素的过程、方法、物品或者装置中还存在另外的相同要素。此外,需要指出的是,本申请实施方式中的方法和装置的范围不限按示出或讨论的顺序来执行功能,还可包括根据所涉及的功能按基本同时的方式或按相反的顺序来执行功能,例如,可以按不同于所描述的次序来执行所描述的方法,并且还可以添加、省去、或组合各种步骤。另外,参照某些示例所描述的特征可在其他示例中被组合。
通过以上的实施方式的描述,本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现,当然也可以通过硬件,但很多情况下前者是更佳的实施方式。基于这样的理解,本申请的技术方案本质上或者说对现有技术做出贡献的部分可以以计算机软件产品的形式体现出来,该计算机软件产品存储在一个存储介质(如ROM/RAM、磁碟、光盘)中,包括若干指令用以使得一台终端(可以是手机,计算机,服务器,空调器,或者网络设备等)执行本申请各个实施例所述的方法。
上面结合附图对本申请的实施例进行了描述,但是本申请并不局限于上述的具体实施方式,上述的具体实施方式仅仅是示意性的,而不是限制性的,本领域的普通技术人员在本申请的启示下,在不脱离本申请宗旨和权利要求所保护的范围情况下,还可做出很多形式,均属于本申请的保护之内。

Claims (66)

  1. 一种基于人工智能AI模型的定位方法,包括:
    第一通信设备获取与AI模型相关信息关联的第一信息;
    所述第一通信设备根据所述第一信息,确定目标信息,所述目标信息包括以下至少一项:目标AI模型、所述AI模型相关信息的有效性信息,或者基于所述目标AI模型进行定位得到的反馈信息;
    所述第一信息用于表示所述AI模型相关信息的有效适用范围,所述AI模型相关信息包括以下至少一项:AI模型、AI模型参数、AI模型的输入、AI模型的输出。
  2. 根据权利要求1所述的基于AI模型的定位方法,其中,所述方法还包括:
    所述第一通信设备获取目标终端的第二信息;所述第二信息用于表示所述目标终端获取的定位相关信息;
    所述第一通信设备根据所述第一信息,确定目标信息,包括:
    所述第一通信设备根据所述第一信息和所述第二信息,确定所述目标信息。
  3. 根据权利要求1或2所述的基于AI模型的定位方法,其中,所述第一信息包括以下至少一项:小区信息;区域信息;有效时间信息;场景信息;信干噪比SINR范围。
  4. 根据权利要求3所述的基于AI模型的定位方法,其中,
    所述小区信息包括以下至少一项:
    一个或多个小区的识别信息;
    一个或多个基站的识别信息;
    一个或多个传输接收点TRP的识别信息;
    小区列表信息;
    小区频域范围信息;
    所述区域信息包括以下至少一项:
    区域识别信息;距离范围信息;距离范围对应的参考点信息;
    所述有效时间信息包括以下至少一项:
    计时器时长;
    计时器起始时间;
    或者,所述场景信息包括以下至少一项:
    视距LOS场景;非视距NLOS场景;复杂场景;室内场景;室外场景。
  5. 根据权利要求2所述的基于AI模型的定位方法,其中,
    所述第二信息包括以下至少一项:所述目标终端的位置信息、小区信息、区域信息,计时器信息、场景信息和所述目标终端测量的SINR;
    其中,所述小区信息为所述目标终端的服务小区、参考小区或者参考信号接收功率RSRP最强小区的以下至少一项信息,所述至少一项信息包括:识别信息、频域信息;
    其中,所述区域信息为所述目标终端的区域识别信息;
    其中,所述场景信息为所述目标终端所处的场景信息。
  6. 根据权利要求2所述的基于AI模型的定位方法,其中,所述第一通信设备根据所述第一信息和所述第二信息,确定所述目标信息,包括:
    所述第一通信设备根据所述第二信息中参数的取值以及所述第一信息中对应参数的范围,确定所述目标信息。
  7. 根据权利要求6所述的基于AI模型的定位方法,其中,
    在所述第二信息中参数的取值处于所述第一信息中对应参数的范围内的情况下,所述AI模型相关信息有效。
  8. 根据权利要求6所述的基于AI模型的定位方法,其中,
    在所述第二信息中参数的取值处于所述第一信息中对应参数的范围内的情况下,所述反馈信息中包括所述第二信息;或者,
    在所述第二信息中参数的取值不处于所述第一信息中对应参数的范围内的情况下,所述反馈信息中包括所述第二信息。
  9. 根据权利要求6所述的基于AI模型的定位方法,其中,若所述第一通信设备确定所述目标终端的小区信息不属于所述第一信息中的小区信息的范围内,所述AI模型相关信息失效。
  10. 根据权利要求6所述的基于AI模型的定位方法,其中,若所述第一通信设备确定所述目标终端的小区信息不属于所述第一信息中的小区信息的范围内,所述反馈信息包括所述目标终端的小区信息。
  11. 根据权利要求6所述的基于AI模型的定位方法,其中,若所述第一通信设备确定所述目标终端的计时器超时,所述AI模型相关信息失效或超时。
  12. 根据权利要求11所述的基于AI模型的定位方法,其中,所述计时器满足以下至少一种情况:
    所述计时器的时长为第一信息中的计时器时长;
    所述计时器从所述第一信息中计时器起始时间开始计时;
    在所述AI模型和/或AI模型参数更新的情况下,所述计时器重新计时。
  13. 根据权利要求6所述的基于AI模型的定位方法,其中,若所述第一通信设备确定所述目标终端的场景信息不属于所述第一信息中的场景信息的范围内,所述AI模型相关信息失效。
  14. 根据权利要求6所述的基于AI模型的定位方法,其中,若所述第一通信设备确定所述目标终端的位置不处于所述第一信息中的小区信息和/或区域信息对应的 位置范围内,所述AI模型相关信息失效。
  15. 根据权利要求6所述的基于AI模型的定位方法,其中,在所述第二信息中参数的取值处于所述第一信息中对应参数的第一范围内的情况下,所述目标AI模型为所述第一信息的第一范围对应的AI模型。
  16. 根据权利要求1或15所述的基于AI模型的定位方法,其中,所述方法还包括:
    所述第一通信设备接收多个预配置的AI模型和/或AI模型参数,以及所述AI模型和/或AI模型参数对应的第一信息。
  17. 根据权利要求16所述的基于AI模型的定位方法,其中,所述方法还包括:
    所述第一通信设备根据所述第一信息的第一范围,从多个所述预配置的AI模型中,获取所述第一信息的第一范围对应的AI模型。
  18. 根据权利要求15所述的基于AI模型的定位方法,其中,
    所述反馈信息包括:所述目标AI模型。
  19. 根据权利要求6、7、8或11-14任一项所述的基于AI模型的定位方法,其中,所述反馈信息为根据所述有效性信息确定的,包括:
    若所述有效性信息用于指示所述AI模型相关信息有效,所述反馈信息至少包括所述AI模型的输出;
    若所述有效性信息用于指示所述AI模型相关信息失效,所述反馈信息包括以下至少一项:误差原因;所述第二信息;有效性信息;AI模型请求;AI模型更新请求;数据收集请求。
  20. 根据权利要求2或15所述的基于AI模型的定位方法,其中,所述反馈信息包括以下至少一项:所述有效性信息、第二信息、第一测量信息和所述AI模型的输出;
    第一测量信息包括以下至少一项:
    信号测量信息;位置信息;误差值;信道冲击响应CIR信息;功率时延谱PDP信息。
  21. 根据权利要求2所述的基于AI模型的定位方法,其中,所述第二信息包括所述目标终端获得的第二测量信息,所述第一通信设备根据所述第一信息和所述第二信息,确定所述目标信息,包括:
    在所述第二测量信息为一次测量得到的测量信息的情况下,所述第一通信设备根据所述第二测量信息中参数的取值以及所述第一信息中对应参数的范围,确定所述目标信息;
    在所述第二测量信息为多次测量得到的测量信息的情况下,所述第一通信设备根据所述第二测量信息中参数的分布与所述第一信息中对应参数的分布的一致性,确定所述目标信息。
  22. 根据权利要求21所述的基于AI模型的定位方法,其中,在所述第二测量信息中参数的取值处于所述第一信息中对应参数的范围内的情况下,所述AI模型相关信息有效;
    在所述第二测量信息中至少一个参数的取值处于所述第一信息中对应参数的范围外的情况下,则所述AI模型相关信息失效。
  23. 根据权利要求21所述的基于AI模型的定位方法,其中,在所述第二测量信息中参数的分布与所述第一信息中对应参数的分布一致的情况下,所述AI模型相关信息有效;
    在所述第二测量信息中至少一个参数的分布与所述第一信息中对应参数的分布不一致的情况下,所述AI模型相关信息失效。
  24. 根据权利要求23所述的基于AI模型的定位方法,其中,在所述第二测量信息中至少一个参数的分布与所述第一信息中对应参数的分布不一致的情况下,所述AI模型相关信息失效,包括:
    在所述第二测量信息中参数的分布与所述第一信息中对应参数的分布不满足于条件阈值信息的情况下,则所述AI模型相关信息失效。
  25. 根据权利要求24所述的基于AI模型的定位方法,其中,
    所述条件阈值信息包括以下至少一项:
    正态分布中σ原则、2σ原则或3σ原则对应的绝对时间;
    σ原则、2σ原则或3σ原则对应的时延拓展值;
    σ原则、2σ原则或3σ原则对应的角度拓展值;
    所述第二测量信息中参数的均值和所述第一信息中对应参数的均值的最大差;
    所述第二测量信息中参数的均值和所述第一信息中对应参数的均值的最大方差;
    所述第二测量信息中参数的均值和所述第一信息中对应参数的方差的最大差;
    所述第二测量信息中参数的均值和所述第一信息中对应参数的方差的最大方差。
  26. 根据权利要求21-25任一项所述的基于AI模型的定位方法,其中,所述第二测量信息包括以下至少一项:
    信干噪比SINR范围、噪声值、NLOS引入的绝对时间值、时延拓展值、角度拓展值、SINR均值和方差、噪声均值和方差、NLOS引入的绝对时间均值和方差、时延拓展均值和方差、角度拓展均值和方差;
    其中,所述SINR范围和所述SINR均值和方差是根据以下至少一项信息的
    SINR得到的;
    噪声值和噪声均值和方差是以下至少一项信息的噪声值得到的;
    所述至少一项信息包括:测量信道、测量信号或第一测量信息。
  27. 根据权利要求26所述的基于AI模型的定位方法,其中,
    所述第一测量信息包括以下至少一项:
    信号测量信息;位置信息;误差值;信道冲击响应CIR信息;功率时延谱PDP信息。
  28. 根据权利要求20或27所述的基于AI模型的定位方法,其中,信号测量信息包括以下至少一项:
    参考信号时间差RSTD测量结果、往返时延测量结果、到达角AOA测量结果、离开角AOD测量结果、参考信息接收功率RSRP、多径测量信息、视距LOS指示信息;
    所述多径测量信息包括以下至少一项:
    首径的功率、首径的时延、首径的到达时间TOA、首径的参考信号时间差
    RSTD、首径的天线子载波相位差、首径的天线子载波相位、多径的功率、多径的时延、TOA、多径的RSTD、多径的天线子载波相位差或多径的天线子载波相位。
  29. 根据权利要求27所述的基于AI模型的定位方法,其中,
    所述第二测量信息还包括所述第一测量信息。
  30. 根据权利要求21-25任一项所述的基于AI模型的定位方法,其中,所述反馈信息包括以下至少一项:所述有效性信息、第二测量信息、AI模型的输入、AI模型的输出、AI模型识别信息、AI模型更新请求。
  31. 根据权利要求21-25任一项所述的基于AI模型的定位方法,其中,
    所述第一信息包括以下至少一项:
    SINR范围;
    噪声的范围;
    噪声分布的均值和/或方差;
    NLOS引入的绝对时间范围或NLOS引入绝对时间的均值和/或方差;
    时延拓展的范围或时延拓展的均值和/或方差;
    角度拓展的范围或角度拓展的均值和/或方差;
    第一测量信息的范围;
    第一测量信息的均值和/或方差。
  32. 根据权利要求1-31任一项所述的基于AI模型的定位方法,其中,第一信息为根据AI模型的测试集和验证集获取到的,包括以下至少一种情况:
    所述第一信息是根据AI模型的测试集和验证集的输入数据获取的特征信息;
    所述第一信息是根据AI模型的测试集和验证集的输出数据获取的特征信息;
    所述第一信息是根据AI模型的测试集和验证集的输入数据和输出数据获取的特征信息。
  33. 根据权利要求21-25任一项所述的基于AI模型的定位方法,其中,
    所述第二测量信息是根据第一测量信息获得的特征信息;和/或,
    所述第二测量信息是根据所述AI模型的输出获得的特征信息。
  34. 根据权利要求31所述的基于AI模型的定位方法,其中,
    所述NLOS引入的绝对时间为所述AI模型获取的时间信息减去非AI模型获取的时间信息;
    所述时延拓展、角度拓展、噪声分布是根据第一测量信息或所述AI模型的输出获取的特征信息。
  35. 根据权利要求1或30所述的基于AI模型的定位方法,其中,所述AI模型的输入和/或AI模型的输出,包括以下至少一项:
    所述目标终端的定位信号测量信息;所述目标终端的位置信息;误差信息;信道冲激响应CIR;首径的功率;首径的时延;首径的到达时间TOA;首径的参考信号时间差RSTD;首径的到达角;首径的天线子载波相位差;多径的功率;多径的时延;多径的TOA;多径的RSTD;多径的到达角;多径的天线子载波相位差;平均过量时延;均方根时延拓展;相干带宽;
    所述误差信息包括以下至少一项:位置误差值、测量误差值、AI模型误差值或AI模型参数误差值。
  36. 根据权利要求30所述的基于AI模型的定位方法,其中,所述AI模型更新请求,包括以下至少一项:
    AI模型的标识ID、满足第二测量信息分布的AI模型和/或AI模型参数、AI模型和/或AI模型参数。
  37. 根据权利要求1-36任一项所述的基于AI模型的定位方法,其中,
    所述AI模型包括以下至少一项:
    AI模型类型;
    AI模型结构。
  38. 根据权利要求1-36任一项所述的基于AI模型的定位方法,其中,所述AI模型包括以下至少一项:
    神经网络列表,包括以下至少一项:每个神经网络的神经元类型、每个神经网络的神经元权重和偏差;
    超参数信息;
    损失函数信息。
  39. 根据权利要求1-36任一项所述的基于AI模型的定位方法,其中,所述AI参数包括以下至少一项:
    超参数信息;
    AI模型描述参数信息;
    AI模型的权值信息;
    AI模型的初始参数。
  40. 根据权利要求1-36任一项所述的基于AI模型的定位方法,其中,有效性信息可以包括以下至少一项:
    有效性指示信息,用于指示所述AI模型相关信息是否有效;
    有效程度;
    有效等级;
    失效原因;
    可靠性指示信息,用于指示基于所述目标AI模型进行定位得到的结果是否可靠;
    可靠性程度;
    可靠性等级。
  41. 根据权利要求1-36任一项所述的基于AI模型的定位方法,其中,所述方法还包括:
    所述第一通信设备接收更新的AI模型和/或AI模型参数,以及与所述AI模型和/或AI模型参数对应的第一信息。
  42. 根据权利要求5、20、27或35任一项所述的基于AI模型的定位方法,其中,所述位置信息是通过以下至少一种方式得到的,所述至少一种方式包括:
    到达时间差定位法OTDOA、全球导航卫星***GNSS、下行到达时间差、上行到达时间差、上行蓝牙到达角AoA、蓝牙出发角AoD、蓝牙、传感器或无线保真WIFI。
  43. 根据权利要求1-36任一项所述的基于AI模型的定位方法,其中,所述第一通信设备包括以下至少一项:
    位置管理功能LMF网元和所述LMF网元的演进设备;
    定位服务器;
    网络数据分析功能NWADF网元;
    其它网络侧设备;
    终端;
    监控设备Actor。
  44. 一种基于AI模型的定位方法,包括:
    第二通信设备接收第一通信设备发送的目标信息,所述目标信息包括以下至少一项:目标AI模型、所述AI模型相关信息的有效性信息或者基于所述目标AI模型进行定位得到的反馈信息;
    所述目标信息为根据AI模型相关信息关联的第一信息确定的,所述第一信息用于表示所述AI模型相关信息的有效适用范围,所述AI模型相关信息包括以下至少一项:AI模型、AI模型参数、AI模型的输入、AI模型的输出。
  45. 根据权利要求44所述的基于AI模型的定位方法,其中,所述方法还包括:
    所述第二通信设备向所述第一通信设备发送所述AI模型相关信息和所述AI模型相关信息关联的第一信息。
  46. 根据权利要求44所述的基于AI模型的定位方法,其中,所述方法还包括:
    所述第二通信设备发送多个预配置的AI模型和/或AI模型参数,以及所述AI模型和/或AI模型参数对应的第一信息。
  47. 根据权利要求44-46任一项所述的基于AI模型的定位方法,其中,所述方法还包括:
    所述第二通信设备向所述第一通信设备发送第二信息,所述第二信息用于表示所述目标终端获取的定位相关信息。
  48. 根据权利要求44-46任一项所述的基于AI模型的定位方法,其中,所述第一信息包括以下至少一项:小区信息;区域信息;有效时间信息;场景信息;信干噪比SINR范围。
  49. 根据权利要求47所述的基于AI模型的定位方法,其中,
    所述第二信息包括以下至少一项:所述目标终端的位置信息、小区信息、区域信息,计时器信息、场景信息和所述目标终端测量的SINR;
    其中,所述小区信息为所述目标终端的服务小区、参考小区或者参考信号接收功率RSRP最强小区的以下至少一项信息,所述至少一项信息包括:识别信息、频域信息;
    其中,所述区域信息为所述目标终端的区域识别信息;
    其中,所述场景信息为所述目标终端所处的场景信息。
  50. 根据权利要求47所述的基于AI模型的定位方法,其中,
    在所述第二信息中参数的取值处于所述第一信息中对应参数的范围内的情况下,所述AI模型相关信息有效。
  51. 根据权利要求47所述的基于AI模型的定位方法,其中,若第二信息中参数的取值处于所述第一信息中对应参数的范围外的情况下,所述AI模型相关信息失效。
  52. 根据权利要求47所述的基于AI模型的定位方法,其中,所述反馈信息包括以下至少一项:所述有效性信息、第二信息、第一测量信息和所述AI模型的输出;
    第一测量信息包括以下至少一项:
    信号测量信息;位置信息;误差值;信道冲击响应CIR信息;功率时延谱PDP信息。
  53. 根据权利要求47所述的基于AI模型的定位方法,其中,所述第二信息包括所述目标终端获得的第二测量信息,在所述第二测量信息中参数的取值处于所述第一信息中对应参数的范围内的情况下,所述AI模型相关信息有效;和/或;
    在所述第二测量信息中至少一个参数的取值处于所述第一信息中对应参数的范围外的情况下,所述AI模型相关信息失效。
  54. 根据权利要求47所述的基于AI模型的定位方法,其中,所述第二信息包括所述目标终端获得的第二测量信息,在所述第二测量信息中参数的分布与所述第一信息中对应参数的分布一致的情况下,所述AI模型相关信息有效;和/或;
    在所述第二测量信息中至少一个参数的分布与所述第一信息中对应参数的分布不一致的情况下,所述AI模型相关信息失效。
  55. 根据权利要求54所述的基于AI模型的定位方法,其中,在所述第二测量信息中至少一个参数的分布与所述第一信息中对应参数的分布不一致的情况下,所述AI模型相关信息失效,包括:
    在所述第二测量信息中参数的分布与所述第一信息中对应参数的分布不满足于条件阈值信息的情况下,所述AI模型相关信息失效。
  56. 根据权利要求55所述的基于AI模型的定位方法,其中,
    所述条件阈值信息包括以下至少一项:
    正态分布中σ原则、2σ原则或3σ原则对应的绝对时间;
    σ原则、2σ原则或3σ原则对应的时延拓展值;
    σ原则、2σ原则或3σ原则对应的角度拓展值;
    所述第二测量信息中参数的均值和所述第一信息中对应参数的均值的最大差;
    所述第二测量信息中参数的均值和所述第一信息中对应参数的均值的最大方差;
    所述第二测量信息中参数的均值和所述第一信息中对应参数的方差的最大差;
    所述第二测量信息中参数的均值和所述第一信息中对应参数的方差的最大方差。
  57. 根据权利要求53-56任一项所述的基于AI模型的定位方法,其中,所述第二测量信息包括以下至少一项:
    信干噪比SINR范围、噪声值、NLOS引入的绝对时间值、时延拓展值、角度拓展值、SINR均值和方差、噪声均值和方差、NLOS引入的绝对时间均值和方差、时延拓展均值和方差、角度拓展均值和方差;
    其中,所述SINR范围和所述SINR均值和方差是根据以下至少一项信息的
    SINR得到的;
    噪声值和噪声均值和方差是以下至少一项信息的噪声值得到的;
    所述至少一项信息包括:测量信道、测量信号或第一测量信息。
  58. 根据权利要求57所述的基于AI模型的定位方法,其中,
    所述第一测量信息包括以下至少一项:
    信号测量信息;位置信息;误差值;信道冲击响应CIR信息;功率时延谱PDP 信息。
  59. 根据权利要求53-56任一项所述的基于AI模型的定位方法,其中,所述反馈信息包括以下至少一项:所述有效性信息、第二测量信息、AI模型的输入、AI模型的输出、AI模型识别信息、AI模型更新请求。
  60. 根据权利要求53-56任一项所述的基于AI模型的定位方法,其中,
    所述第一信息包括以下至少一项:
    SINR范围;
    噪声的范围;
    噪声分布的均值和/或方差;
    NLOS引入的绝对时间范围或NLOS引入绝对时间的均值和/或方差;
    时延拓展的范围或时延拓展的均值和/或方差;
    角度拓展的范围或角度拓展的均值和/或方差;
    第一测量信息的范围;
    第一测量信息的均值和/或方差。
  61. 根据权利要求44-60任一项所述的基于AI模型的定位方法,其中,所述第二通信设备包括以下至少一项:
    终端;
    模型管理设备;
    网络侧设备。
  62. 一种AI模型的定位装置,包括:
    获取模块,用于获取与AI模型相关信息关联的第一信息;
    处理模块,用于根据所述第一信息,确定目标信息,所述目标信息包括以下至少一项:目标AI模型、所述AI模型相关信息的有效性信息或者基于所述目标AI模型进行定位得到的反馈信息;
    所述第一信息用于表示所述AI模型相关信息的有效适用范围,所述AI模型相关信息包括以下至少一项:AI模型、AI模型参数、AI模型的输入、AI模型的输出。
  63. 一种AI模型的定位装置,包括:
    接收模块,用于接收第一通信设备发送的目标信息,所述目标信息包括以下至少一项:目标AI模型、所述AI模型相关信息的有效性信息或者基于所述目标AI模型进行定位得到的反馈信息;
    所述目标信息为根据AI模型相关信息关联的第一信息确定的,所述第一信息用于表示所述AI模型相关信息的有效适用范围,所述AI模型相关信息包括以下至少一项:AI模型、AI模型参数、AI模型的输入、AI模型的输出。
  64. 一种第一通信设备,包括处理器和存储器,所述存储器存储可在所述处理器 上运行的程序或指令,所述程序或指令被所述处理器执行时实现如权利要求1至43任一项所述的基于AI模型的定位方法的步骤。
  65. 一种第二通信设备,包括处理器和存储器,所述存储器存储可在所述处理器上运行的程序或指令,所述程序或指令被所述处理器执行时实现如权利要求44至61任一项所述的基于AI模型的定位方法的步骤。
  66. 一种可读存储介质,所述可读存储介质上存储程序或指令,所述程序或指令被处理器执行时实现如权利要求1-43任一项所述的基于AI模型的定位方法,或者实现如权利要求44至61任一项所述的基于AI模型的定位方法的步骤。
PCT/CN2023/073715 2022-01-29 2023-01-29 基于人工智能ai模型的定位方法及通信设备 WO2023143572A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210113101.4A CN116567806A (zh) 2022-01-29 2022-01-29 基于人工智能ai模型的定位方法及通信设备
CN202210113101.4 2022-01-29

Publications (1)

Publication Number Publication Date
WO2023143572A1 true WO2023143572A1 (zh) 2023-08-03

Family

ID=87470718

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/073715 WO2023143572A1 (zh) 2022-01-29 2023-01-29 基于人工智能ai模型的定位方法及通信设备

Country Status (2)

Country Link
CN (1) CN116567806A (zh)
WO (1) WO2023143572A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117580073B (zh) * 2024-01-12 2024-05-07 北京小米移动软件有限公司 通信方法、设备和存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106658420A (zh) * 2016-11-21 2017-05-10 北京金坤科创技术有限公司 基于环境感知的定位终端设计
US20210136527A1 (en) * 2019-11-06 2021-05-06 Huawei Technologies Co., Ltd. Connectivity-based positioning determination in wireless communication networks
DE102020203703A1 (de) * 2020-03-23 2021-09-23 Robert Bosch Gesellschaft mit beschränkter Haftung Verfahren und Vorrichtung zur Lokalisierung mobiler Teilnehmer eines Drahtlosnetzwerks
WO2021213376A1 (zh) * 2020-04-22 2021-10-28 维沃移动通信有限公司 定位方法、通信设备和网络设备
CN113839797A (zh) * 2020-06-23 2021-12-24 华为技术有限公司 数据处理方法和装置
CN113923730A (zh) * 2020-07-07 2022-01-11 展讯通信(上海)有限公司 小区重选方法及装置

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106658420A (zh) * 2016-11-21 2017-05-10 北京金坤科创技术有限公司 基于环境感知的定位终端设计
US20210136527A1 (en) * 2019-11-06 2021-05-06 Huawei Technologies Co., Ltd. Connectivity-based positioning determination in wireless communication networks
DE102020203703A1 (de) * 2020-03-23 2021-09-23 Robert Bosch Gesellschaft mit beschränkter Haftung Verfahren und Vorrichtung zur Lokalisierung mobiler Teilnehmer eines Drahtlosnetzwerks
WO2021213376A1 (zh) * 2020-04-22 2021-10-28 维沃移动通信有限公司 定位方法、通信设备和网络设备
CN113839797A (zh) * 2020-06-23 2021-12-24 华为技术有限公司 数据处理方法和装置
CN113923730A (zh) * 2020-07-07 2022-01-11 展讯通信(上海)有限公司 小区重选方法及装置

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
PURPLE MOUNTAIN LABORATORIES: "Consideration on AI/ML-based 5G enhancements for R18", 3GPP DRAFT; RP-212934, 3RD GENERATION PARTNERSHIP PROJECT (3GPP), MOBILE COMPETENCE CENTRE ; 650, ROUTE DES LUCIOLES ; F-06921 SOPHIA-ANTIPOLIS CEDEX ; FRANCE, vol. TSG RAN, no. e Meeting; 20211206 - 20211217, 26 November 2021 (2021-11-26), Mobile Competence Centre ; 650, route des Lucioles ; F-06921 Sophia-Antipolis Cedex ; France, XP052081701 *

Also Published As

Publication number Publication date
CN116567806A (zh) 2023-08-08

Similar Documents

Publication Publication Date Title
WO2023143572A1 (zh) 基于人工智能ai模型的定位方法及通信设备
WO2023134650A1 (zh) 信息交互方法、装置及通信设备
WO2023066288A1 (zh) 模型请求方法、模型请求处理方法及相关设备
WO2023098661A1 (zh) 定位方法及通信设备
WO2023098662A1 (zh) 定位方法及通信设备
WO2023151657A1 (zh) 信息处理方法及通信设备
WO2024083004A1 (zh) Ai模型配置方法、终端及网络侧设备
WO2023155839A1 (zh) Ai模型的在线学习方法、装置、通信设备及可读存储介质
WO2023185865A1 (zh) 模型有效性反馈方法、装置、终端及网络侧设备
WO2023179617A1 (zh) 定位方法、装置、终端及网络侧设备
WO2023134653A1 (zh) 通信网络预测方法、终端及网络侧设备
WO2023103911A1 (zh) 测量方法、装置、设备及存储介质
WO2024120445A1 (zh) 模型输入信息的确定方法、装置、设备、***及存储介质
WO2023236979A1 (zh) 定位模型的选择方法、终端及网络侧设备
WO2024027576A1 (zh) 一种ai网络模型的性能监督方法、装置和通信设备
WO2023088423A1 (zh) 定位方法、装置、终端及网络侧设备
WO2023179540A1 (zh) 信道预测方法、装置及无线通信设备
CN118042399A (zh) 模型监督方法、终端及网络侧设备
WO2024120444A1 (zh) 模型监督方法、装置、终端、网络侧设备及可读存储介质
WO2024027580A1 (zh) 事件信息传输方法、lmf、接入网设备及终端
WO2024125525A1 (zh) Ai算力上报方法、终端及网络侧设备
WO2024032695A1 (zh) Csi预测处理方法、装置、通信设备及可读存储介质
WO2024067280A1 (zh) 更新ai模型参数的方法、装置及通信设备
WO2023040885A1 (zh) 参数选择方法、参数配置方法、终端及网络侧设备
CN117411602A (zh) 数据采集方法及装置

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: 23746449

Country of ref document: EP

Kind code of ref document: A1