WO2022050444A1 - Communication method for federated learning and device for performing same - Google Patents

Communication method for federated learning and device for performing same Download PDF

Info

Publication number
WO2022050444A1
WO2022050444A1 PCT/KR2020/011878 KR2020011878W WO2022050444A1 WO 2022050444 A1 WO2022050444 A1 WO 2022050444A1 KR 2020011878 W KR2020011878 W KR 2020011878W WO 2022050444 A1 WO2022050444 A1 WO 2022050444A1
Authority
WO
WIPO (PCT)
Prior art keywords
weight change
learning
change amount
communication
devices
Prior art date
Application number
PCT/KR2020/011878
Other languages
French (fr)
Korean (ko)
Inventor
김일환
이종구
Original Assignee
엘지전자 주식회사
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 엘지전자 주식회사 filed Critical 엘지전자 주식회사
Priority to PCT/KR2020/011878 priority Critical patent/WO2022050444A1/en
Priority to KR1020237007045A priority patent/KR20230060505A/en
Publication of WO2022050444A1 publication Critical patent/WO2022050444A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/098—Distributed learning, e.g. federated learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/0495—Quantised networks; Sparse networks; Compressed networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods

Definitions

  • the present specification relates to a communication method for federated learning and a device for performing the same, and in particular, discloses a communication method and device capable of reducing the size of weight information transmitted from devices to a server.
  • a federated learning method capable of learning a global model of the cloud while maintaining the privacy of each device is used.
  • Federated learning can reduce the amount of data transmission because the data acquired by devices is not transmitted to the server, but devices must transmit weight information to the server. Since the weight information has to be transmitted every round in which learning is performed, data consumption due to this may also be burdened.
  • the present specification is to provide a communication method for federated learning capable of reducing data consumption and a device for performing the same.
  • a communication method for federated learning in which a server derives a final learning result includes the steps of devices receiving a scaling factor based on a global model and a weight change amount from a server; It includes the steps of the devices calculating the weight change amount based on the global model, the devices performing quantization based on the weight change amount, and the devices transmitting the quantized weight change amount to the server.
  • the performing of the quantization may include using the scaling coefficients having the same value in each of the devices.
  • the step in which the devices are provided with the scaling factor may include receiving the scaling factor generated based on the distribution of the weight change in the (i-1)-th round (i is a natural number greater than or equal to 2) in the i-th round. .
  • the generating of the weight change distribution in the (i-1)-th round includes generating a cumulative distribution function accumulating the absolute values of the weight change amount of at least one round among rounds up to the (i-1)-th round to do; and clipping the cumulative distribution of the absolute value of the weight change amount greater than or equal to a preset threshold in the cumulative distribution function.
  • the determining of the scaling factor may include: calculating a change in loss based on a difference between the loss of the (i-1)-th round and the loss of the (i-2)-th round; and determining the magnitude of the scaling factor based on the magnitude of the change in loss.
  • the determining of the scaling factor may include determining within a range smaller than a size of the boundary value with respect to a preset maximum quantization bit.
  • the determining of the scaling factor may include dividing a range smaller than the size of the boundary value into two or more sections, and generating different scaling factors in the divided sections.
  • the performing of the quantization may include calculating the weight change amount as a product of the scaling factor and the variable weight change amount; and calculating a quantization range of the variable weight change amount.
  • Transmitting the quantized weight change amount to the server may further include transmitting variable quantization information changed based on the scaling factor.
  • a device for performing federated learning based on a global model provided from a server includes a transceiver for communication with the server and a processor for performing federated learning based on the global model.
  • the processor receives the global model and a scaling factor based on the weight change amount from the server, calculates the weight change amount based on the global model, performs quantization based on the weight change amount, and transmits the quantized weight change amount to the server.
  • the processor may receive, in the i-th round, the scaling factor generated based on the distribution of the weight change amount in the (i-1)-th round (i is a natural number equal to or greater than 2).
  • the weight change distribution generated in the (i-1)-th round is a cumulative distribution function that accumulates the absolute values of the weight change amount of at least one round among rounds before the (i-1)-th round, It may be generated by clipping the cumulative distribution of the absolute value of the weight change amount equal to or greater than a threshold value.
  • the scaling factor may determine a size based on a difference between the loss of the (i-1)-th round and the loss of the (i-2)-th round.
  • 1 illustrates physical channels and general signal transmission used in a 3GPP system.
  • FIG. 2 is a diagram illustrating an example of a communication structure that can be provided in a 6G system.
  • 5 illustrates a deep neural network structure
  • FIG. 6 illustrates a convolutional neural network structure
  • FIG. 7 illustrates a filter operation in a convolutional neural network.
  • FIG. 8 illustrates a neural network structure in which a cyclic loop exists.
  • FIG. 9 illustrates the operational structure of a recurrent neural network.
  • FIG. 11 shows an example of a THz communication application.
  • FIG. 12 is a diagram illustrating a communication system for federated learning to which an embodiment is applied.
  • FIG. 13 is a diagram illustrating a federated learning protocol according to an embodiment.
  • FIG. 14 is a diagram illustrating a communication method for joint learning according to an embodiment.
  • 15 is a diagram illustrating a method of determining a scaling factor according to an embodiment of the present invention.
  • 16 is a diagram illustrating an example of a weight change amount cumulative distribution function generated by a server.
  • 17 is a diagram showing an example of error evaluation of a loss function obtained using a mean square error.
  • 18 is a diagram for explaining the operation of devices.
  • 21 illustrates a signal processing circuit for a transmission signal.
  • FIG. 22 shows another example of a wireless device to which the present invention is applied.
  • FIG. 23 illustrates a portable device to which the present invention is applied.
  • FIG. 24 illustrates a vehicle or an autonomous driving vehicle to which the present invention is applied.
  • 25 illustrates a vehicle to which the present invention is applied.
  • 26 illustrates an XR device applied to the present invention.
  • CDMA may be implemented with a radio technology such as Universal Terrestrial Radio Access (UTRA) or CDMA2000.
  • TDMA may be implemented with a radio technology such as Global System for Mobile communications (GSM)/General Packet Radio Service (GPRS)/Enhanced Data Rates for GSM Evolution (EDGE).
  • GSM Global System for Mobile communications
  • GPRS General Packet Radio Service
  • EDGE Enhanced Data Rates for GSM Evolution
  • OFDMA may be implemented with a radio technology such as IEEE 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE 802-20, Evolved UTRA (E-UTRA), and the like.
  • UTRA is part of the Universal Mobile Telecommunications System (UMTS).
  • 3GPP 3rd Generation Partnership Project
  • Long Term Evolution is a part of Evolved UMTS (E-UMTS) using E-UTRA and LTE-A (Advanced)/LTE-A pro is an evolved version of 3GPP LTE.
  • 3GPP NR New Radio or New Radio Access Technology
  • 3GPP 6G may be an evolved version of 3GPP NR.
  • LTE refers to technology after 3GPP TS 36.xxx Release 8.
  • LTE technology after 3GPP TS 36.xxx Release 10 is referred to as LTE-A
  • LTE technology after 3GPP TS 36.xxx Release 13 is referred to as LTE-A pro
  • 3GPP NR refers to technology after TS 38.
  • 3GPP 6G may refer to technology after TS Release 17 and/or Release 18.
  • xxx stands for standard document detail number.
  • LTE/NR/6G may be collectively referred to as a 3GPP system.
  • terms, abbreviations, etc. used in the description of the present invention reference may be made to matters described in standard documents published before the present invention. For example, you can refer to the following documents:
  • RRC Radio Resource Control
  • RRC Radio Resource Control
  • a terminal receives information through a downlink (DL) from a base station, and the terminal transmits information through an uplink (UL) to the base station.
  • Information transmitted and received between the base station and the terminal includes data and various control information, and various physical channels exist according to the type/use of the information they transmit and receive.
  • the terminal When the terminal is powered on or newly enters a cell, the terminal performs an initial cell search operation, such as synchronizing with the base station (S11). To this end, the terminal receives a primary synchronization signal (PSS) and a secondary synchronization signal (SSS) from the base station, synchronizes with the base station, and obtains information such as a cell ID. Thereafter, the terminal may receive a physical broadcast channel (PBCH) from the base station to obtain intra-cell broadcast information. On the other hand, the UE may receive a downlink reference signal (DL RS) in the initial cell search step to check the downlink channel state.
  • PSS primary synchronization signal
  • SSS secondary synchronization signal
  • PBCH physical broadcast channel
  • DL RS downlink reference signal
  • the UE After the initial cell search, the UE receives a Physical Downlink Control Channel (PDCCH) and a Physical Downlink Control Channel (PDSCH) according to information carried on the PDCCH to obtain more specific system information. It can be done (S12).
  • PDCH Physical Downlink Control Channel
  • PDSCH Physical Downlink Control Channel
  • the terminal may perform a random access procedure (RACH) for the base station (S13 to S16).
  • RACH Random Access procedure
  • the UE transmits a specific sequence as a preamble through a Physical Random Access Channel (PRACH) (S13 and S15), and a response message to the preamble through the PDCCH and the corresponding PDSCH ((Random Access (RAR)) Response) message)
  • PRACH Physical Random Access Channel
  • RAR Random Access
  • a contention resolution procedure may be additionally performed (S16).
  • the UE After performing the procedure as described above, the UE performs PDCCH/PDSCH reception (S17) and Physical Uplink Shared Channel (PUSCH)/Physical Uplink Control Channel (Physical Uplink) as a general uplink/downlink signal transmission procedure.
  • Control Channel (PUCCH) transmission (S18) may be performed.
  • the UE may receive downlink control information (DCI) through the PDCCH.
  • DCI downlink control information
  • the DCI includes control information such as resource allocation information for the terminal, and different formats may be applied according to the purpose of use.
  • control information that the terminal transmits to the base station through the uplink or the terminal receives from the base station includes a downlink/uplink ACK/NACK signal, a channel quality indicator (CQI), a precoding matrix index (PMI), and a rank indicator (RI). ) and the like.
  • the UE may transmit the above-described control information such as CQI/PMI/RI through PUSCH and/or PUCCH.
  • the base station transmits a related signal to the terminal through a downlink channel to be described later, and the terminal receives the related signal from the base station through a downlink channel to be described later.
  • PDSCH Physical Downlink Shared Channel
  • PDSCH carries downlink data (eg, DL-shared channel transport block, DL-SCH TB), and modulation methods such as Quadrature Phase Shift Keying (QPSK), 16 Quadrature Amplitude Modulation (QAM), 64 QAM, and 256 QAM are available. applies.
  • QPSK Quadrature Phase Shift Keying
  • QAM 16 Quadrature Amplitude Modulation
  • a codeword is generated by encoding the TB.
  • a PDSCH can carry multiple codewords. Scrambling and modulation mapping are performed for each codeword, and modulation symbols generated from each codeword are mapped to one or more layers (Layer mapping). Each layer is mapped to a resource together with a demodulation reference signal (DMRS), is generated as an OFDM symbol signal, and is transmitted through a corresponding antenna port.
  • DMRS demodulation reference signal
  • the PDCCH carries downlink control information (DCI) and a QPSK modulation method is applied.
  • DCI downlink control information
  • One PDCCH is composed of 1, 2, 4, 8, 16 CCEs (Control Channel Elements) according to an Aggregation Level (AL).
  • One CCE consists of six REGs (Resource Element Groups).
  • One REG is defined as one OFDM symbol and one (P)RB.
  • the UE obtains DCI transmitted through the PDCCH by performing decoding (aka, blind decoding) on the set of PDCCH candidates.
  • a set of PDCCH candidates decoded by the UE is defined as a PDCCH search space set.
  • the search space set may be a common search space or a UE-specific search space.
  • the UE may acquire DCI by monitoring PDCCH candidates in one or more search space sets configured by MIB or higher layer signaling.
  • the terminal transmits a related signal to the base station through an uplink channel to be described later, and the base station receives the related signal from the terminal through an uplink channel to be described later.
  • PUSCH Physical Uplink Shared Channel
  • PUSCH carries uplink data (eg, UL-shared channel transport block, UL-SCH TB) and/or uplink control information (UCI), and CP-OFDM (Cyclic Prefix - Orthogonal Frequency Division Multiplexing) waveform (waveform) , DFT-s-OFDM (Discrete Fourier Transform - spread - Orthogonal Frequency Division Multiplexing) is transmitted based on the waveform.
  • the PUSCH is transmitted based on the DFT-s-OFDM waveform
  • the UE transmits the PUSCH by applying transform precoding.
  • the UE when transform precoding is not possible (eg, transform precoding is disabled), the UE transmits a PUSCH based on the CP-OFDM waveform, and when transform precoding is possible (eg, transform precoding is enabled), the UE transmits the CP-OFDM PUSCH may be transmitted based on a waveform or a DFT-s-OFDM waveform.
  • PUSCH transmission is dynamically scheduled by a UL grant in DCI, or based on higher layer (eg, RRC) signaling (and/or Layer 1 (L1) signaling (eg, PDCCH)) semi-statically. Can be scheduled (configured grant).
  • PUSCH transmission may be performed on a codebook-based or non-codebook-based basis.
  • the PUCCH carries uplink control information, HARQ-ACK and/or a scheduling request (SR), and may be divided into a plurality of PUCCHs according to the PUCCH transmission length.
  • SR scheduling request
  • 6G (wireless) systems have (i) very high data rates per device, (ii) very large number of connected devices, (iii) global connectivity, (iv) very low latency, (v) battery- It aims to reduce energy consumption of battery-free IoT devices, (vi) ultra-reliable connections, and (vii) connected intelligence with machine learning capabilities.
  • the vision of the 6G system can be in four aspects: intelligent connectivity, deep connectivity, holographic connectivity, and ubiquitous connectivity, and the 6G system can satisfy the requirements shown in Table 1 below. That is, Table 1 is a table showing an example of the requirements of the 6G system.
  • FIG. 2 is a diagram showing an example of a communication structure that can be provided in a 6G system.
  • eMBB Enhanced mobile broadband
  • URLLC Ultra-reliable low latency communications
  • mMTC massive machine-type communication
  • AI integrated communication Tactile internet, High throughput, High network capacity, High energy efficiency, Low backhaul and It may have key factors such as access network congestion and enhanced data security.
  • FIG. 2 is a diagram showing an example of a communication structure that can be provided in a 6G system.
  • 6G systems are expected to have 50 times higher simultaneous wireless connectivity than 5G wireless communication systems.
  • URLLC a key feature of 5G, will become an even more important technology by providing an end-to-end delay of less than 1ms in 6G communication.
  • 6G systems will have much better volumetric spectral efficiencies as opposed to frequently used areal spectral efficiencies.
  • the 6G system can provide very long battery life and advanced battery technology for energy harvesting, so mobile devices will not need to be charged separately in the 6G system.
  • New network characteristics in 6G may be as follows.
  • 6G is expected to be integrated with satellites to provide a global mobile population.
  • the integration of terrestrial, satellite and public networks into one wireless communication system is very important for 6G.
  • AI may be applied in each step of a communication procedure (or each procedure of signal processing to be described later).
  • the 6G wireless network will deliver power to charge the batteries of devices such as smartphones and sensors. Therefore, wireless information and energy transfer (WIET) will be integrated.
  • WIET wireless information and energy transfer
  • Small cell networks The idea of small cell networks was introduced to improve the received signal quality as a result of improved throughput, energy efficiency and spectral efficiency in cellular systems. As a result, small cell networks are essential characteristics for communication systems beyond 5G and Beyond 5G (5GB). Accordingly, the 6G communication system also adopts the characteristics of the small cell network.
  • Ultra-dense heterogeneous networks will be another important characteristic of 6G communication systems.
  • a multi-tier network composed of heterogeneous networks improves overall QoS and reduces costs.
  • a backhaul connection is characterized as a high-capacity backhaul network to support high-capacity traffic.
  • High-speed fiber optics and free-space optics (FSO) systems may be possible solutions to this problem.
  • High-precision localization (or location-based service) through communication is one of the functions of the 6G wireless communication system. Therefore, the radar system will be integrated with the 6G network.
  • Softening and virtualization are two important features that underlie the design process in 5GB networks to ensure flexibility, reconfigurability and programmability. In addition, billions of devices can be shared in a shared physical infrastructure.
  • AI The most important and newly introduced technology for 6G systems is AI.
  • AI was not involved in the 4G system.
  • 5G systems will support partial or very limited AI.
  • the 6G system will be AI-enabled for full automation.
  • Advances in machine learning will create more intelligent networks for real-time communication in 6G.
  • Incorporating AI into communications can simplify and enhance real-time data transmission.
  • AI can use numerous analytics to determine how complex target tasks are performed. In other words, AI can increase efficiency and reduce processing delays.
  • AI can also play an important role in M2M, machine-to-human and human-to-machine communication.
  • AI can be a rapid communication in BCI (Brain Computer Interface).
  • BCI Brain Computer Interface
  • AI-based communication systems can be supported by metamaterials, intelligent structures, intelligent networks, intelligent devices, intelligent cognitive radios, self-sustaining wireless networks, and machine learning.
  • AI-based physical layer transmission means applying a signal processing and communication mechanism based on an AI driver rather than a traditional communication framework in a fundamental signal processing and communication mechanism.
  • deep learning-based channel coding and decoding, deep learning-based signal estimation and detection, deep learning-based MIMO mechanism, AI-based resource scheduling and It may include an allocation (allocation) and the like.
  • Machine learning may be used for channel estimation and channel tracking, and may be used for power allocation, interference cancellation, and the like in a physical layer of a downlink (DL). In addition, machine learning may be used for antenna selection, power control, symbol detection, and the like in a MIMO system.
  • DL downlink
  • machine learning may be used for antenna selection, power control, symbol detection, and the like in a MIMO system.
  • Deep learning-based AI algorithms require large amounts of training data to optimize training parameters.
  • a lot of training data is used offline. This is because static training on training data in a specific channel environment may cause a contradiction between dynamic characteristics and diversity of a wireless channel.
  • signals of the physical layer of wireless communication are complex signals.
  • further research on a neural network for detecting a complex domain signal is needed.
  • Machine learning refers to a set of actions that trains a machine to create a machine that can perform tasks that humans can or cannot do.
  • Machine learning requires data and a learning model.
  • data learning methods can be roughly divided into three types: supervised learning, unsupervised learning, and reinforcement learning.
  • Neural network learning is to minimize output errors. Neural network learning repeatedly inputs learning data into the neural network, calculates the output and target errors of the neural network for the training data, and backpropagates the neural network error from the output layer of the neural network to the input layer in the direction to reduce the error. ) to update the weight of each node in the neural network.
  • Supervised learning uses training data in which the correct answer is labeled in the training data, and in unsupervised learning, the correct answer may not be labeled in the training data. That is, for example, learning data in the case of supervised learning related to data classification may be data in which categories are labeled for each of the training data.
  • the labeled training data is input to the neural network, and an error can be calculated by comparing the output (category) of the neural network with the label of the training data.
  • the calculated error is back propagated in the reverse direction (ie, from the output layer to the input layer) in the neural network, and the connection weight of each node of each layer of the neural network may be updated according to the back propagation.
  • a change amount of the connection weight of each node to be updated may be determined according to a learning rate.
  • the computation of the neural network on the input data and the backpropagation of errors can constitute a learning cycle (epoch).
  • the learning rate may be applied differently depending on the number of repetitions of the learning cycle of the neural network. For example, in the early stage of learning a neural network, a high learning rate can be used to increase the efficiency by allowing the neural network to quickly obtain a certain level of performance, and in the late learning period, a low learning rate can be used to increase the accuracy.
  • the learning method may vary depending on the characteristics of the data. For example, when the purpose of accurately predicting data transmitted from a transmitter in a communication system is at a receiver, it is preferable to perform learning using supervised learning rather than unsupervised learning or reinforcement learning.
  • the learning model corresponds to the human brain, and the most basic linear model can be considered. ) is called
  • the neural network cord used as a learning method is largely divided into deep neural networks (DNN), convolutional deep neural networks (CNN), and Recurrent Boltzmann Machine (RNN) methods. there is.
  • DNN deep neural networks
  • CNN convolutional deep neural networks
  • RNN Recurrent Boltzmann Machine
  • An artificial neural network is an example of connecting several perceptrons.
  • the huge artificial neural network structure may extend the simplified perceptron structure shown in FIG. 3 to apply input vectors to different multidimensional perceptrons.
  • an input value or an output value is referred to as a node.
  • the perceptron structure shown in FIG. 3 can be described as being composed of a total of three layers based on an input value and an output value.
  • An artificial neural network in which H (d+1)-dimensional perceptrons exist between the 1st layer and the 2nd layer and K (H+1)-dimensional perceptrons exist between the 2nd layer and the 3rd layer can be expressed as shown in FIG. 4 .
  • the layer where the input vector is located is called the input layer
  • the layer where the final output value is located is called the output layer
  • all the layers located between the input layer and the output layer are called hidden layers.
  • three layers are disclosed, but when counting the actual number of artificial neural network layers, the input layer is counted except for the input layer, so it can be viewed as a total of two layers.
  • the artificial neural network is constructed by connecting the perceptrons of the basic blocks in two dimensions.
  • the aforementioned input layer, hidden layer, and output layer can be jointly applied in various artificial neural network structures such as CNN and RNN to be described later as well as multi-layer perceptron.
  • various artificial neural network structures such as CNN and RNN to be described later as well as multi-layer perceptron.
  • the artificial neural network becomes deeper, and a machine learning paradigm that uses a sufficiently deep artificial neural network as a learning model is called deep learning.
  • an artificial neural network used for deep learning is called a deep neural network (DNN).
  • DNN deep neural network
  • the deep neural network shown in FIG. 5 is a multilayer perceptron composed of eight hidden layers + output layers.
  • the multi-layered perceptron structure is referred to as a fully-connected neural network.
  • a connection relationship does not exist between nodes located in the same layer, and a connection relationship exists only between nodes located in adjacent layers.
  • DNN has a fully connected neural network structure and is composed of a combination of multiple hidden layers and activation functions, so it can be usefully applied to identify the correlation characteristics between input and output.
  • the correlation characteristic may mean a joint probability of input/output.
  • nodes located inside one layer are arranged in a one-dimensional vertical direction.
  • the nodes are two-dimensionally arranged with w horizontally and h vertical nodes (convolutional neural network structure of FIG. 6 ).
  • a weight is added per connection in the connection process from one input node to the hidden layer, a total of h ⁇ w weights must be considered.
  • h ⁇ w nodes in the input layer a total of h 2 w 2 weights are needed between two adjacent layers.
  • the convolutional neural network of FIG. 6 has a problem in that the number of weights increases exponentially according to the number of connections, so instead of considering the connection of all modes between adjacent layers, it is assumed that a filter with a small size exists in FIG. 7 As in Fig., the weighted sum and activation function calculations are performed on the overlapping filters.
  • One filter has a weight corresponding to the number corresponding to its size, and weight learning can be performed so that a specific feature on an image can be extracted and output as a factor.
  • a filter with a size of 3 ⁇ 3 is applied to the upper left 3 ⁇ 3 region of the input layer, and an output value obtained by performing weighted sum and activation function operations on the corresponding node is stored in z22.
  • the filter performs weight sum and activation function calculations while moving horizontally and vertically at regular intervals while scanning the input layer, and places the output value at the current filter position.
  • This calculation method is similar to a convolution operation on an image in the field of computer vision, so a deep neural network with such a structure is called a convolutional neural network (CNN), and a hidden layer generated as a result of the convolution operation is called a convolutional layer.
  • a neural network having a plurality of convolutional layers is called a deep convolutional neural network (DCNN).
  • the number of weights can be reduced by calculating the weighted sum by including only nodes located in the region covered by the filter in the node where the filter is currently located. Due to this, one filter can be used to focus on features for a local area. Accordingly, CNN can be effectively applied to image data processing in which physical distance in a two-dimensional domain is an important criterion. Meanwhile, in CNN, a plurality of filters may be applied immediately before the convolution layer, and a plurality of output results may be generated through the convolution operation of each filter.
  • a structure in which this method is applied to an artificial neural network is called a recurrent neural network structure.
  • a recurrent neural network connects elements (x1(t), x2(t), ,..., xd(t)) of a certain gaze t on a data sequence to a fully connected neural network.
  • the previous time point t-1 is weighted by inputting the hidden vectors (z1(t-1), z2(t*?*1),..., zH(t*?*1)) together. and a structure to which an activation function is applied.
  • the reason why the hidden vector is transferred to the next time point in this way is that information in the input vector at previous time points is considered to be accumulated in the hidden vector of the current time point.
  • the recurrent neural network operates in a predetermined time sequence with respect to an input data sequence.
  • the hidden vector (z1(1), z2(1), ... ,zH(1)) is input together with the input vector (x1(2),x2(2),...,xd(2)) of time point 2, and then the vector of the hidden layer (z1(2)) through weighted sum and activation functions ),z2(2) ,...,zH(2)). This process is repeatedly performed until time point 2, time point 3, ,, and time point T.
  • a deep recurrent neural network when a plurality of hidden layers are arranged in a recurrent neural network, this is called a deep recurrent neural network (DRNN).
  • the recurrent neural network is designed to be usefully applied to sequence data (eg, natural language processing).
  • Deep Q-Network As a neural network core used as a learning method, in addition to DNN, CNN, and RNN, Restricted Boltzmann Machine (RBM), deep belief networks (DBN), Deep Q-Network and It includes various deep learning techniques such as, and can be applied to fields such as computer vision, voice recognition, natural language processing, and voice/signal processing.
  • RBM Restricted Boltzmann Machine
  • DNN deep belief networks
  • Deep Q-Network includes various deep learning techniques such as, and can be applied to fields such as computer vision, voice recognition, natural language processing, and voice/signal processing.
  • AI-based physical layer transmission means applying a signal processing and communication mechanism based on an AI driver rather than a traditional communication framework in a fundamental signal processing and communication mechanism.
  • deep learning-based channel coding and decoding, deep learning-based signal estimation and detection, deep learning-based MIMO mechanism, AI-based resource scheduling and It may include an allocation (allocation) and the like.
  • the data rate can be increased by increasing the bandwidth. This can be accomplished by using sub-THz communication with a wide bandwidth and applying advanced large-scale MIMO technology.
  • THz waves also known as sub-millimeter radiation, typically exhibit a frequency band between 0.1 THz and 10 THz with corresponding wavelengths in the range of 0.03 mm-3 mm.
  • the 100GHz-300GHz band range (Sub THz band) is considered a major part of the THz band for cellular communication.
  • Sub-THz band Addition to mmWave band increases 6G cellular communication capacity.
  • 300GHz-3THz is in the far-infrared (IR) frequency band.
  • the 300GHz-3THz band is part of the broadband, but at the edge of the wideband, just behind the RF band. Thus, this 300 GHz-3 THz band shows similarities to RF. 10 shows an example of an electromagnetic spectrum.
  • THz wave is located between RF (Radio Frequency)/millimeter (mm) and infrared band, (i) It transmits non-metal/non-polar material better than visible light/infrared light, and has a shorter wavelength than RF/millimeter wave, so it has high straightness. Beam focusing may be possible. In addition, since the photon energy of the THz wave is only a few meV, it is harmless to the human body.
  • the frequency band expected to be used for THz wireless communication may be a D-band (110 GHz to 170 GHz) or H-band (220 GHz to 325 GHz) band with low propagation loss due to absorption of molecules in the air.
  • THz wireless communication may be applied to wireless recognition, sensing, imaging, wireless communication, THz navigation, and the like.
  • the main characteristics of THz communication include (i) widely available bandwidth to support very high data rates, and (ii) high path loss occurring at high frequencies (high directional antennas are indispensable).
  • the narrow beamwidth produced by the highly directional antenna reduces interference.
  • the small wavelength of the THz signal allows a much larger number of antenna elements to be integrated into devices and BSs operating in this band. This allows the use of advanced adaptive nesting techniques that can overcome range limitations.
  • FIG. 11 shows an example of a THz communication application.
  • a THz wireless communication scenario may be classified into a macro network, a micro network, and a nanoscale network.
  • THz wireless communication can be applied to vehicle-to-vehicle connection and backhaul/fronthaul connection.
  • THz wireless communication in micro networks is applied to indoor small cells, fixed point-to-point or multi-point connections such as wireless connections in data centers, and near-field communication such as kiosk downloading.
  • Table 2 below is a table showing an example of a technique that can be used in the THz wave.
  • OWC technology is envisioned for 6G communications in addition to RF-based communications for all possible device-to-access networks. These networks connect to network-to-backhaul/fronthaul network connections.
  • OWC technology has already been used since the 4G communication system, but will be used more widely to meet the needs of the 6G communication system.
  • OWC technologies such as light fidelity, visible light communication, optical camera communication, and FSO communication based on a light band are well known technologies.
  • Communication based on optical radio technology can provide very high data rates, low latency and secure communication.
  • LiDAR can also be used for ultra-high-resolution 3D mapping in 6G communication based on wide bands.
  • FSO The transmitter and receiver characteristics of an FSO system are similar to those of a fiber optic network.
  • data transmission in an FSO system is similar to that of a fiber optic system. Therefore, FSO can be a good technology to provide backhaul connectivity in 6G systems along with fiber optic networks.
  • FSO supports high-capacity backhaul connections for remote and non-remote areas such as sea, space, underwater, and isolated islands.
  • FSO also supports cellular BS connectivity.
  • MIMO technology improves, so does the spectral efficiency. Therefore, large-scale MIMO technology will be important in 6G systems. Since the MIMO technology uses multiple paths, a multiplexing technique and a beam generation and operation technique suitable for the THz band should also be considered important so that a data signal can be transmitted through one or more paths.
  • Blockchain will become an important technology for managing large amounts of data in future communication systems.
  • Blockchain is a form of distributed ledger technology, which is a database distributed across numerous nodes or computing devices. Each node replicates and stores an identical copy of the ledger.
  • the blockchain is managed as a peer-to-peer network. It can exist without being managed by a centralized authority or server. Data on the blockchain is collected together and organized into blocks. Blocks are linked together and protected using encryption.
  • Blockchain in nature perfectly complements IoT at scale with improved interoperability, security, privacy, reliability and scalability. Therefore, blockchain technology provides several features such as interoperability between devices, traceability of large amounts of data, autonomous interaction of different IoT systems, and large-scale connection stability of 6G communication systems.
  • the 6G system integrates terrestrial and public networks to support vertical expansion of user communications.
  • 3D BS will be provided via low orbit satellites and UAVs. Adding a new dimension in terms of elevation and associated degrees of freedom makes 3D connections significantly different from traditional 2D networks.
  • UAVs Unmanned Aerial Vehicles
  • a BS entity is installed in the UAV to provide cellular connectivity.
  • UAVs have certain features not found in fixed BS infrastructure, such as easy deployment, strong line-of-sight links, and degrees of freedom with controlled mobility.
  • eMBB enhanced Mobile Broadband
  • URLLC Universal Mobile Broadband
  • mMTC massive Machine Type Communication
  • Tight integration of multiple frequencies and heterogeneous communication technologies is very important in 6G systems. As a result, users can seamlessly move from one network to another without having to make any manual configuration on the device. The best network is automatically selected from the available communication technologies. This will break the limitations of the cell concept in wireless communication. Currently, user movement from one cell to another causes too many handovers in high-density networks, causing handover failures, handover delays, data loss and ping-pong effects. 6G cell-free communication will overcome all of this and provide better QoS. Cell-free communication will be achieved through multi-connectivity and multi-tier hybrid technologies and different heterogeneous radios of devices.
  • WIET uses the same fields and waves as wireless communication systems.
  • the sensor and smartphone will be charged using wireless power transfer during communication.
  • WIET is a promising technology for extending the life of battery-charging wireless systems. Therefore, devices without batteries will be supported in 6G communication.
  • An autonomous wireless network is a function that can continuously detect dynamically changing environmental conditions and exchange information between different nodes.
  • sensing will be tightly integrated with communications to support autonomous systems.
  • the density of access networks in 6G will be enormous.
  • Each access network is connected by backhaul connections such as fiber optic and FSO networks.
  • backhaul connections such as fiber optic and FSO networks.
  • Beamforming is a signal processing procedure that adjusts an antenna array to transmit a radio signal in a specific direction.
  • Beamforming technology has several advantages such as high call-to-noise ratio, interference prevention and rejection, and high network efficiency.
  • Hologram beamforming (HBF) is a new beamforming method that is significantly different from MIMO systems because it uses a software-defined antenna. HBF will be a very effective approach for efficient and flexible transmission and reception of signals in multi-antenna communication devices in 6G.
  • Big data analytics is a complex process for analyzing various large data sets or big data. This process ensures complete data management by finding information such as hidden data, unknown correlations and customer propensity. Big data is gathered from a variety of sources such as videos, social networks, images and sensors. This technology is widely used to process massive amounts of data in 6G systems.
  • the linearity is strong, so there may be many shaded areas due to obstructions.
  • the LIS technology that expands the communication area, strengthens communication stability and enables additional additional services becomes important.
  • the LIS is an artificial surface made of electromagnetic materials, and can change the propagation of incoming and outgoing radio waves.
  • LIS can be seen as an extension of massive MIMO, but the array structure and operation mechanism are different from those of massive MIMO.
  • LIS has low power consumption in that it operates as a reconfigurable reflector with passive elements, that is, only passively reflects the signal without using an active RF chain.
  • each of the passive reflectors of the LIS must independently adjust the phase shift of the incoming signal, it can be advantageous for a wireless communication channel.
  • the reflected signal can be gathered at the target receiver to boost the received signal power.
  • the above salpin 6G communication technology may be applied in combination with the methods proposed in the present specification to be described later, or may be supplemented to specify or clarify the technical characteristics of the methods proposed in the present specification.
  • the communication service proposed in the present specification may be applied in combination with a communication service by 3G, 4G and/or 5G communication technology as well as the 6G communication technology described above.
  • FIG. 12 is a diagram illustrating a communication system for federated learning to which an embodiment is applied.
  • the communication system for federated learning includes a server (MS) and devices (DE).
  • the server MS of the base station BS has a global model.
  • the server MS may communicate with the plurality of devices DE in various ways.
  • the global model is for performing AI learning, and the server (MS) provides the global model to the devices (DE).
  • Each of the devices DE acquires local data and learns the local data using the provided global model.
  • the devices DE update the weight through learning, calculate the change amount of the updated weight, and transmit it to the server MS.
  • the weight change amount means a difference between the weights before the update and the weights after the update.
  • the devices DE may transmit the updated weight, not the weight change amount, to the server MS.
  • the server MS updates the weight of the global model based on the weight change amount provided from the devices DE. And, after the weight update is finished, the server MS may evaluate the loss function of the global model.
  • FIG. 13 is a diagram illustrating a federated learning protocol according to an embodiment.
  • federated learning consists of a plurality of rounds, and each round includes a selection section, a configuration section, and a reporting section.
  • the selection procedure is a process in which the devices DE register with the server MS to participate in federated learning, and the server MS selects a device to participate in the corresponding round from among a plurality of devices.
  • the configuration section is a process in which the server (MS) transmits the global model and necessary parameters to selected devices, and the devices learn the received global model using local data.
  • the reporting section is a section in which each device calculates the weight change amount from the learned model, and transmits the calculated weight change amount to the server (MS).
  • the server MS updates the weights of the global model based on the weight changes received within a predetermined time.
  • Each procedure in the round is performed in a state in which synchronization between the server (MS) and the devices (DE) is made.
  • the server MS updates the global model using the weight changes received from the selected devices within a given time.
  • the server (MS) proceeds with the next round of learning using the updated global model.
  • FIG. 14 is a diagram illustrating a communication method for joint learning according to an embodiment.
  • the server (MS) transmits a global model and a configuration parameter to the device (DE).
  • the globularization parameter includes a scaling factor of quantization.
  • the visualization parameter may include statistical information such as a maximum value, a minimum value, and a variance of the weight change amount distribution.
  • the scaling factor is determined based on the distribution of the weight change amount of each device received in the previous round. Each device is provided with the same scaling factor.
  • the device DE learns using the global model, and then performs quantization based on the scaling factor provided from the server MS.
  • the device DE may additionally perform variable bit-based quantization according to a distribution of a weight change amount of each layer.
  • the device DE may include quantization bit information in the weight change amount information and transmit it to the server MS.
  • 15 is a diagram illustrating a method of determining a scaling factor according to an embodiment of the present invention.
  • the server MS analyzes the weight changes received in the previous round and the loss function evaluation.
  • the server MS generates a cumulative distribution function (CDF) accumulating absolute values of the weight changes based on the weight change amounts provided from the respective devices DE.
  • CDF cumulative distribution function
  • the server MS acquires a threshold value gmax corresponding to a threshold value gthreshold based on the weight change cumulative distribution function.
  • a cumulative weight change having a size greater than or equal to the threshold gmax is not used for learning. This is because weight changes having a very small frequency and a large size may make learning inefficient compared to other weight changes.
  • the threshold value (gthreshold) may be preset, and the size of the threshold value (gthreshold) may vary for each round.
  • the server MS performs error evaluation of the loss function based on the global model updated in the previous round.
  • 17 is a diagram illustrating an example of error evaluation of a loss function obtained using a mean square error.
  • the error of the loss function tends to converge to a specific value as learning progresses. As such, as the error of the loss function converges, the gradient change also gradually decreases, so that a distribution centered on “0” may be exhibited. However, since the error of the loss function is not converged at the beginning of learning, even if the weight changes are distributed in a narrow range, it has the possibility of being widely distributed in the subsequent rounds.
  • the server MS determines a scaling factor based on the loss function error and the cumulative distribution function of the weight change amount.
  • the maximum quantization bit for the remaining weight change value after considering the scaling coefficient is determined to be 8 bits.
  • the weight change amount will have a value in the range of -127 to 127 excluding the scaling factor.
  • the maximum value of the weight change in the cumulative distribution function of the weight change is determined as the boundary value gmax, the distribution of the next round is in the range from "-gmax" to "gmax". Accordingly, the server MS determines the scaling factor of the next round as a value of “gmax/127”.
  • the server MS determines whether to apply the scaling factor based on the evaluation result of the loss function. For example, in FIG. 17 , since a lot of learning has not been performed before about 20 rounds, the distribution of the weight change amount for each round may represent a large difference. Therefore, the scaling factor can be adjusted according to the error of the loss function as follows.
  • the amount of change in the loss function may be calculated as the difference between the loss of the previous round and the loss of the current round.
  • the scaling factor of the next round may be calculated as in Equation 2 below.
  • the amount of change in the loss function ( Loss) is "0 ⁇
  • the scaling factor of the next round may be determined as in [Equation 3].
  • the scaling factor of the next round may be determined as in [Equation 4].
  • the loss obtained through [Equation 1] to [Equation 4] means the evaluation result of the loss function of the global model updated in the corresponding round as shown in FIG. 17 .
  • one or more scaling factors may be determined. For example, only one scaling factor may be determined within a range from “-gmax” to “gmax”.
  • a range within the absolute value of the boundary value gmax may be divided into two or more, and a scaling factor may be determined from each boundary value.
  • a first scaling factor may be determined in a range of “-gmax” to “-gmax/2”
  • a second scaling factor may be determined in a range of “-gmax/2” to “gmax”.
  • the first scaling factor may match the first quantization information
  • the second scaling factor may match the second quantization information.
  • the first quantization information and the second quantization information may be different from each other.
  • the range of the weight change can be reduced more quickly, and when the performance is deteriorated compared to the previous round, the scaling factor can be increased to wide the range of the weight change.
  • a third step (S1503) the server (MS) transmits the determined scaling factor to the devices.
  • the devices DE perform a variable bit-based quantization operation using a scaling factor from the server MS. Referring to FIG. 18, it is as follows.
  • 18 is a diagram for explaining the operation of devices.
  • the devices DE are provided with visualization information for joint learning from the server MS.
  • the configuration information includes a global model and a scaling factor.
  • the devices DE acquire local data and learn a global model based on the local data.
  • the devices DE calculate a weight change amount from the updated global model after learning is completed.
  • the devices (DE) quantize the weight change amount calculated in the second step (S1802) by using the scaling factor provided from the server (MS).
  • the quantization assumes an 8-bit quantization process.
  • the weight change amount ( Wt) can be expressed as the following [Equation 5].
  • the weight change amount excluding the scaling factor term is in the range of "-6 to +6". It is possible to perform 4-bit quantization for a weight change amount in the range of "-6 to +6". Accordingly, the device DE transmits the quantization bit (2), which is information about 4-bit quantization, and the quantized weight change amount together. "Quantization bit(2)" is information indicating how many bits quantization was performed.
  • the devices DE quantize values within the range of the weight change amount by 8-bit and transmit them as shown in [Equation 6].
  • the weight change amount quantized with a scaling factor of 32 bits (0.2/255) and 8 bits is transmitted. Comparing the amount of weight change transmitted for the above two methods, 64 bits are required for the existing method and 18 bits are required for the proposed method. Therefore, compared to the conventional method in which 8-bit quantization is performed and transmitted, the size of the transmitted weight change amount can be reduced by using the difference in the weight change amount for each device in the proposed method.
  • the devices DE transmit the weight change amount information on which quantization is completed to the server. And, when variable bit quantization is performed, the devices DE also transmit variable quantization bit information to the server MS.
  • the above-described embodiment of the present invention can reduce the amount of data in the process of transmitting the weight change amount.
  • a method of transmitting a general weight change amount may result in unnecessary waste of resources. This is because the weight variation distribution characteristics of the devices may be different. Looking at this:
  • weight changes may be widely distributed or weight changes may be distributed within a narrow range depending on the layer.
  • the weight change amount distribution characteristic for each device may be different.
  • the distribution of weight change may be distributed in a very small range.
  • a wide range of weight change distribution can be shown after learning on the global model.
  • the respective devices may have very different distribution characteristics of weight variations based on the same layer.
  • the weight change amount of the first device is in the range of "-1 to +1”
  • the weight change amount of the second device is in the range "-0.05 to +0.1”
  • the weight change amount of the third device is "-0.8 to +0.6” “It can be in scope.
  • all of the first to third devices must use a matrix of weight change amounts configured on an 8-bit basis.
  • the server sums up the weight changes at the same location and calculates an average.
  • the influence of the server on the weight updating process is negligible. In this way, in the process of updating the weights by the server, even a device that has little practical effect uses the same quantization bit, so communication bandwidth tends to be wasted.
  • the amount of weight change amount data provided to a device having a small weight change amount can be reduced.
  • the communication system 1 applied to the present invention includes a wireless device, a base station, and a network.
  • the wireless device refers to a device that performs communication using a radio access technology (eg, 5G NR (New RAT), LTE (Long Term Evolution)), and may be referred to as a communication/wireless/5G device.
  • a radio access technology eg, 5G NR (New RAT), LTE (Long Term Evolution)
  • the wireless device may include a robot 100a, a vehicle 100b-1, 100b-2, an eXtended Reality (XR) device 100c, a hand-held device 100d, and a home appliance 100e. ), an Internet of Thing (IoT) device 100f, and an AI device/server 400 .
  • the vehicle may include a vehicle equipped with a wireless communication function, an autonomous driving vehicle, a vehicle capable of performing inter-vehicle communication, and the like.
  • the vehicle may include an Unmanned Aerial Vehicle (UAV) (eg, a drone).
  • UAV Unmanned Aerial Vehicle
  • XR devices include AR (Augmented Reality)/VR (Virtual Reality)/MR (Mixed Reality) devices, and include a Head-Mounted Device (HMD), a Head-Up Display (HUD) provided in a vehicle, a television, a smartphone, It may be implemented in the form of a computer, a wearable device, a home appliance, a digital signage, a vehicle, a robot, and the like.
  • the portable device may include a smart phone, a smart pad, a wearable device (eg, a smart watch, smart glasses), a computer (eg, a laptop computer), and the like.
  • Home appliances may include a TV, a refrigerator, a washing machine, and the like.
  • the IoT device may include a sensor, a smart meter, and the like.
  • the base station and the network may be implemented as a wireless device, and the specific wireless device 200a may operate as a base station/network node to other wireless devices.
  • the wireless devices 100a to 100f may be connected to the network 300 through the base station 200 .
  • AI Artificial Intelligence
  • the network 300 may be configured using a 3G network, a 4G (eg, LTE) network, or a 5G (eg, NR) network.
  • the wireless devices 100a to 100f may communicate with each other through the base station 200/network 300, but may also communicate directly (e.g. sidelink communication) without passing through the base station/network.
  • the vehicles 100b-1 and 100b-2 may perform direct communication (e.g. Vehicle to Vehicle (V2V)/Vehicle to everything (V2X) communication).
  • the IoT device eg, sensor
  • the IoT device may communicate directly with other IoT devices (eg, sensor) or other wireless devices 100a to 100f.
  • Wireless communication/connection 150a and 150b may be performed between the wireless devices 100a to 100f/base station 200 - the base station 200/wireless devices 100a to 100f.
  • the wireless communication/connection may be performed through various wireless access technologies (eg, 5G NR) for uplink/downlink communication 150a and sidelink communication 150b (or D2D communication).
  • 5G NR wireless access technologies
  • the wireless device and the base station/wireless device may transmit/receive wireless signals to each other.
  • the wireless communication/connection 150a and 150b may transmit/receive signals through various physical channels based on all/part of the process of FIG. A1 .
  • various configuration information setting processes for wireless signal transmission/reception various signal processing processes (eg, channel encoding/decoding, modulation/demodulation, resource mapping/demapping, etc.) , at least a part of a resource allocation process may be performed.
  • various signal processing processes eg, channel encoding/decoding, modulation/demodulation, resource mapping/demapping, etc.
  • the first wireless device 100 and the second wireless device 200 may transmit/receive wireless signals through various wireless access technologies (eg, LTE, NR).
  • ⁇ first wireless device 100, second wireless device 200 ⁇ is ⁇ wireless device 100x, base station 200 ⁇ of FIG. 19 and/or ⁇ wireless device 100x, wireless device 100x) ⁇ can be matched.
  • the first wireless device 100 includes one or more processors 102 and one or more memories 104 , and may further include one or more transceivers 106 and/or one or more antennas 108 .
  • the processor 102 controls the memory 104 and/or the transceiver 106 and may be configured to implement the functions, procedures and/or methods described/suggested above. For example, the processor 102 may process information in the memory 104 to generate first information/signal, and then transmit a wireless signal including the first information/signal through the transceiver 106 . In addition, the processor 102 may receive the radio signal including the second information/signal through the transceiver 106 , and then store information obtained from signal processing of the second information/signal in the memory 104 .
  • the memory 104 may be connected to the processor 102 and may store various information related to the operation of the processor 102 .
  • the memory 104 may store software code including instructions for performing some or all of the processes controlled by the processor 102 , or for performing the procedures and/or methods described/suggested above.
  • the processor 102 and the memory 104 may be part of a communication modem/circuit/chip designed to implement a wireless communication technology (eg, LTE, NR).
  • the transceiver 106 may be coupled to the processor 102 , and may transmit and/or receive wireless signals via one or more antennas 108 .
  • the transceiver 106 may include a transmitter and/or a receiver.
  • the transceiver 106 may be used interchangeably with a radio frequency (RF) unit.
  • a wireless device may refer to a communication modem/circuit/chip.
  • the second wireless device 200 includes one or more processors 202 , one or more memories 204 , and may further include one or more transceivers 206 and/or one or more antennas 208 .
  • the processor 202 controls the memory 204 and/or the transceiver 206 and may be configured to implement the functions, procedures, and/or methods described/suggested above. For example, the processor 202 may process the information in the memory 204 to generate third information/signal, and then transmit a wireless signal including the third information/signal through the transceiver 206 . In addition, the processor 202 may receive the radio signal including the fourth information/signal through the transceiver 206 , and then store information obtained from signal processing of the fourth information/signal in the memory 204 .
  • the memory 204 may be connected to the processor 202 and may store various information related to the operation of the processor 202 .
  • the memory 204 may store software code including instructions for performing some or all of the processes controlled by the processor 202 , or for performing the procedures and/or methods described/suggested above.
  • the processor 202 and the memory 204 may be part of a communication modem/circuit/chip designed to implement a wireless communication technology (eg, LTE, NR).
  • the transceiver 206 may be coupled to the processor 202 and may transmit and/or receive wireless signals via one or more antennas 208 .
  • Transceiver 206 may include a transmitter and/or receiver. Transceiver 206 may be used interchangeably with an RF unit.
  • a wireless device may refer to a communication modem/circuit/chip.
  • one or more protocol layers may be implemented by one or more processors 102 , 202 .
  • one or more processors 102 , 202 may implement one or more layers (eg, functional layers such as PHY, MAC, RLC, PDCP, RRC, SDAP).
  • the one or more processors 102 and 202 may generate one or more Protocol Data Units (PDUs) and/or one or more Service Data Units (SDUs) according to the functions, procedures, proposals and/or methods disclosed herein.
  • PDUs Protocol Data Units
  • SDUs Service Data Units
  • One or more processors 102 , 202 may generate messages, control information, data, or information according to the functions, procedures, proposals and/or methods disclosed herein.
  • the one or more processors 102 and 202 generate a signal (eg, a baseband signal) including PDUs, SDUs, messages, control information, data or information according to the functions, procedures, proposals and/or methods disclosed herein. , to one or more transceivers 106 and 206 .
  • One or more processors 102 , 202 may receive signals (eg, baseband signals) from one or more transceivers 106 , 206 , PDUs, SDUs, and/or SDUs according to the functions, procedures, proposals and/or methods disclosed herein. , a message, control information, data or information can be obtained.
  • One or more processors 102, 202 may be referred to as a controller, microcontroller, microprocessor, or microcomputer.
  • One or more processors 102 , 202 may be implemented by hardware, firmware, software, or a combination thereof.
  • ASICs Application Specific Integrated Circuits
  • DSPs Digital Signal Processors
  • DSPDs Digital Signal Processing Devices
  • PLDs Programmable Logic Devices
  • FPGAs Field Programmable Gate Arrays
  • the functions, procedures, proposals and/or methods disclosed in this document may be implemented using firmware or software, and the firmware or software may be implemented to include modules, procedures, functions, and the like.
  • Firmware or software configured to perform the functions, procedures, proposals, and/or methods disclosed herein is included in one or more processors 102, 202, or stored in one or more memories 104, 204, to one or more processors 102, 202) can be driven.
  • the functions, procedures, proposals and/or methods disclosed in this document may be implemented using firmware or software in the form of code, instructions, and/or a set of instructions.
  • One or more memories 104 , 204 may be coupled with one or more processors 102 , 202 , and may store various forms of data, signals, messages, information, programs, code, instructions, and/or instructions.
  • the one or more memories 104 and 204 may be comprised of ROM, RAM, EPROM, flash memory, hard drives, registers, cache memory, computer readable storage media, and/or combinations thereof.
  • One or more memories 104 , 204 may be located inside and/or external to one or more processors 102 , 202 . Additionally, one or more memories 104 , 204 may be coupled to one or more processors 102 , 202 through various technologies, such as wired or wireless connections.
  • One or more transceivers 106 , 206 may transmit user data, control information, radio signals/channels, etc. referred to in the methods and/or operational flowcharts of this document to one or more other devices.
  • the one or more transceivers 106, 206 may receive user data, control information, radio signals/channels, etc., referred to in the functions, procedures, proposals, methods, and/or flowcharts of operations disclosed herein, or the like, from one or more other devices.
  • one or more transceivers 106 , 206 may be coupled to one or more processors 102 , 202 and may transmit and receive wireless signals.
  • one or more processors 102 , 202 may control one or more transceivers 106 , 206 to transmit user data, control information, or wireless signals to one or more other devices.
  • one or more processors 102 , 202 may control one or more transceivers 106 , 206 to receive user data, control information, or wireless signals from one or more other devices.
  • one or more transceivers 106, 206 may be coupled to one or more antennas 108, 208, and the one or more transceivers 106, 206 may be coupled to one or more of the transceivers 106, 206 via the one or more antennas 108, 208 for the functions, procedures, and procedures disclosed herein.
  • one or more antennas may be a plurality of physical antennas or a plurality of logical antennas (eg, antenna ports).
  • the one or more transceivers 106, 206 convert the received radio signal/channel, etc. from the RF band signal to process the received user data, control information, radio signal/channel, etc. using the one or more processors 102, 202. It can be converted into a baseband signal.
  • One or more transceivers 106 and 206 may convert user data, control information, radio signals/channels, etc. processed using one or more processors 102 and 202 from baseband signals to RF band signals.
  • one or more transceivers 106 , 206 may include (analog) oscillators and/or filters.
  • 21 illustrates a signal processing circuit for a transmission signal.
  • the signal processing circuit 1000 may include a scrambler 1010 , a modulator 1020 , a layer mapper 1030 , a precoder 1040 , a resource mapper 1050 , and a signal generator 1060 .
  • the operations/functions of FIG. 21 may be performed by the processors 102 and 202 and/or the transceivers 106 and 206 of FIG. 20 .
  • the hardware elements of FIG. 21 may be implemented in the processors 102 , 202 and/or transceivers 106 , 206 of FIG. 20 .
  • blocks 1010 to 1060 may be implemented in the processors 102 and 202 of FIG. 20 .
  • blocks 1010 to 1050 may be implemented in the processors 102 and 202 of FIG. 20
  • block 1060 may be implemented in the transceivers 106 and 206 of FIG. 20 .
  • the codeword may be converted into a wireless signal through the signal processing circuit 1000 of FIG. 21 .
  • the codeword is a coded bit sequence of an information block.
  • the information block may include a transport block (eg, a UL-SCH transport block, a DL-SCH transport block).
  • the radio signal may be transmitted through various physical channels (eg, PUSCH, PDSCH) of FIG. A1 .
  • the codeword may be converted into a scrambled bit sequence by the scrambler 1010 .
  • a scramble sequence used for scrambling is generated based on an initialization value, and the initialization value may include ID information of a wireless device, and the like.
  • the scrambled bit sequence may be modulated by a modulator 1020 into a modulation symbol sequence.
  • the modulation method may include pi/2-Binary Phase Shift Keying (pi/2-BPSK), m-Phase Shift Keying (m-PSK), m-Quadrature Amplitude Modulation (m-QAM), and the like.
  • the complex modulation symbol sequence may be mapped to one or more transport layers by the layer mapper 1030 .
  • Modulation symbols of each transport layer may be mapped to corresponding antenna port(s) by the precoder 1040 (precoding).
  • the output z of the precoder 1040 may be obtained by multiplying the output y of the layer mapper 1030 by the precoding matrix W of N*M.
  • N is the number of antenna ports
  • M is the number of transport layers.
  • the precoder 1040 may perform precoding after performing transform precoding (eg, DFT transform) on the complex modulation symbols. Also, the precoder 1040 may perform precoding without performing transform precoding.
  • the resource mapper 1050 may map modulation symbols of each antenna port to a time-frequency resource.
  • the time-frequency resource may include a plurality of symbols (eg, a CP-OFDMA symbol, a DFT-s-OFDMA symbol) in the time domain and a plurality of subcarriers in the frequency domain.
  • CP Cyclic Prefix
  • DAC Digital-to-Analog Converter
  • the signal processing process for the received signal in the wireless device may be configured in reverse of the signal processing process 1010 to 1060 of FIG. 21 .
  • the wireless device eg, 100 and 200 in FIG. 20
  • the received radio signal may be converted into a baseband signal through a signal restorer.
  • the signal restorer may include a frequency downlink converter, an analog-to-digital converter (ADC), a CP remover, and a Fast Fourier Transform (FFT) module.
  • ADC analog-to-digital converter
  • FFT Fast Fourier Transform
  • the baseband signal may be restored to a codeword through a resource de-mapper process, a postcoding process, a demodulation process, and a descrambling process.
  • the codeword may be restored to the original information block through decoding.
  • the signal processing circuit (not shown) for the received signal may include a signal restorer, a resource de-mapper, a post coder, a demodulator, a descrambler, and a decoder.
  • the wireless device 22 shows another example of a wireless device to which the present invention is applied.
  • the wireless device may be implemented in various forms according to use-examples/services (refer to FIGS. 19 and 23 to 26 ).
  • wireless devices 100 and 200 correspond to wireless devices 100 and 200 of FIG. 20 , and include various elements, components, units/units, and/or modules. ) may consist of
  • the wireless devices 100 and 200 may include a communication unit 110 , a control unit 120 , a memory unit 130 , and an additional element 140 .
  • the communication unit may include communication circuitry 112 and transceiver(s) 114 .
  • communication circuitry 112 may include one or more processors 102,202 and/or one or more memories 104,204 of FIG. 20 .
  • the transceiver(s) 114 may include one or more transceivers 106 , 206 and/or one or more antennas 108 , 208 of FIG.
  • the control unit 120 is electrically connected to the communication unit 110 , the memory unit 130 , and the additional element 140 , and controls general operations of the wireless device. For example, the controller 120 may control the electrical/mechanical operation of the wireless device based on the program/code/command/information stored in the memory unit 130 . In addition, the control unit 120 transmits information stored in the memory unit 130 to the outside (eg, other communication device) through the communication unit 110 through a wireless/wired interface, or externally (eg, through the communication unit 110 ) Information received through a wireless/wired interface from another communication device) may be stored in the memory unit 130 .
  • the outside eg, other communication device
  • Information received through a wireless/wired interface from another communication device may be stored in the memory unit 130 .
  • the additional element 140 may be configured in various ways according to the type of the wireless device.
  • the additional element 140 may include at least one of a power unit/battery, an input/output unit (I/O unit), a driving unit, and a computing unit.
  • the wireless device includes a robot ( FIGS. 19 and 100a ), a vehicle ( FIGS. 19 , 100b-1 , 100b-2 ), an XR device ( FIGS. 19 and 100c ), a mobile device ( FIGS. 19 and 100d ), and a home appliance. (FIG. 19, 100e), IoT device (FIG.
  • digital broadcasting terminal digital broadcasting terminal
  • hologram device public safety device
  • MTC device medical device
  • fintech device or financial device
  • security device climate/environment device
  • It may be implemented in the form of an AI server/device ( FIGS. 19 and 400 ), a base station ( FIGS. 19 and 200 ), and a network node.
  • the wireless device may be mobile or used in a fixed location depending on the use-example/service.
  • various elements, components, units/units, and/or modules in the wireless devices 100 and 200 may be entirely interconnected through a wired interface, or at least some of them may be wirelessly connected through the communication unit 110 .
  • the control unit 120 and the communication unit 110 are connected by wire, and the control unit 120 and the first unit (eg, 130 , 140 ) are connected to the communication unit 110 through the communication unit 110 . It can be connected wirelessly.
  • each element, component, unit/unit, and/or module within the wireless device 100 , 200 may further include one or more elements.
  • the controller 120 may be configured with one or more processor sets.
  • control unit 120 may be configured as a set of a communication control processor, an application processor, an electronic control unit (ECU), a graphic processing processor, a memory control processor, and the like.
  • memory unit 130 may include random access memory (RAM), dynamic RAM (DRAM), read only memory (ROM), flash memory, volatile memory, and non-volatile memory. volatile memory) and/or a combination thereof.
  • FIG. 22 will be described in more detail with reference to the drawings.
  • the portable device may include a smart phone, a smart pad, a wearable device (eg, a smart watch, smart glasses), and a portable computer (eg, a laptop computer).
  • a mobile device may be referred to as a mobile station (MS), a user terminal (UT), a mobile subscriber station (MSS), a subscriber station (SS), an advanced mobile station (AMS), or a wireless terminal (WT).
  • MS mobile station
  • UT user terminal
  • MSS mobile subscriber station
  • SS subscriber station
  • AMS advanced mobile station
  • WT wireless terminal
  • the portable device 100 includes an antenna unit 108 , a communication unit 110 , a control unit 120 , a memory unit 130 , a power supply unit 140a , an interface unit 140b , and an input/output unit 140c . ) may be included.
  • the antenna unit 108 may be configured as a part of the communication unit 110 .
  • Blocks 110 to 130/140a to 140c respectively correspond to blocks 110 to 130/140 of FIG. 22 .
  • the communication unit 110 may transmit and receive signals (eg, data, control signals, etc.) with other wireless devices and base stations.
  • the controller 120 may control components of the portable device 100 to perform various operations.
  • the controller 120 may include an application processor (AP).
  • the memory unit 130 may store data/parameters/programs/codes/commands necessary for driving the portable device 100 . Also, the memory unit 130 may store input/output data/information.
  • the power supply unit 140a supplies power to the portable device 100 and may include a wired/wireless charging circuit, a battery, and the like.
  • the interface unit 140b may support a connection between the portable device 100 and other external devices.
  • the interface unit 140b may include various ports (eg, an audio input/output port and a video input/output port) for connection with an external device.
  • the input/output unit 140c may receive or output image information/signal, audio information/signal, data, and/or information input from a user.
  • the input/output unit 140c may include a camera, a microphone, a user input unit, a display unit 140d, a speaker, and/or a haptic module.
  • the input/output unit 140c obtains information/signals (eg, touch, text, voice, image, video) input from the user, and the obtained information/signals are stored in the memory unit 130 . can be saved.
  • the communication unit 110 may convert the information/signal stored in the memory into a wireless signal, and transmit the converted wireless signal directly to another wireless device or to a base station. Also, after receiving a radio signal from another radio device or base station, the communication unit 110 may restore the received radio signal to original information/signal. After the restored information/signal is stored in the memory unit 130 , it may be output in various forms (eg, text, voice, image, video, haptic) through the input/output unit 140c.
  • various forms eg, text, voice, image, video, haptic
  • the vehicle or autonomous driving vehicle may be implemented as a mobile robot, a vehicle, a train, an aerial vehicle (AV), a ship, and the like.
  • AV aerial vehicle
  • the vehicle or autonomous driving vehicle 100 includes an antenna unit 108 , a communication unit 110 , a control unit 120 , a driving unit 140a , a power supply unit 140b , a sensor unit 140c and autonomous driving. It may include a part 140d.
  • the antenna unit 108 may be configured as a part of the communication unit 110 .
  • Blocks 110/130/140a-140d correspond to blocks 110/130/140 of FIG. 22, respectively.
  • the communication unit 110 may transmit/receive signals (eg, data, control signals, etc.) to and from external devices such as other vehicles, base stations (e.g., base stations, roadside units, etc.), servers, and the like.
  • the controller 120 may control elements of the vehicle or the autonomous driving vehicle 100 to perform various operations.
  • the controller 120 may include an Electronic Control Unit (ECU).
  • the driving unit 140a may cause the vehicle or the autonomous driving vehicle 100 to run on the ground.
  • the driving unit 140a may include an engine, a motor, a power train, a wheel, a brake, a steering device, and the like.
  • the power supply unit 140b supplies power to the vehicle or the autonomous driving vehicle 100 , and may include a wired/wireless charging circuit, a battery, and the like.
  • the sensor unit 140c may obtain vehicle status, surrounding environment information, user information, and the like.
  • the sensor unit 140c includes an inertial measurement unit (IMU) sensor, a collision sensor, a wheel sensor, a speed sensor, an inclination sensor, a weight sensor, a heading sensor, a position module, and a vehicle forward movement.
  • IMU inertial measurement unit
  • a collision sensor a wheel sensor
  • a speed sensor a speed sensor
  • an inclination sensor a weight sensor
  • a heading sensor a position module
  • a vehicle forward movement / may include a reverse sensor, a battery sensor, a fuel sensor, a tire sensor, a steering sensor, a temperature sensor, a humidity sensor, an ultrasonic sensor, an illuminance sensor, a pedal position sensor, and the like.
  • the autonomous driving unit 140d includes a technology for maintaining a driving lane, a technology for automatically adjusting speed such as adaptive cruise control, a technology for automatically driving along a predetermined route, and a technology for automatically setting a route when a destination is set. technology can be implemented.
  • the communication unit 110 may receive map data, traffic information data, and the like from an external server.
  • the autonomous driving unit 140d may generate an autonomous driving route and a driving plan based on the acquired data.
  • the controller 120 may control the driving unit 140a to move the vehicle or the autonomous driving vehicle 100 along the autonomous driving path (eg, speed/direction adjustment) according to the driving plan.
  • the communication unit 110 may obtain the latest traffic information data from an external server non/periodically, and may acquire surrounding traffic information data from surrounding vehicles.
  • the sensor unit 140c may acquire vehicle state and surrounding environment information.
  • the autonomous driving unit 140d may update the autonomous driving route and driving plan based on the newly acquired data/information.
  • the communication unit 110 may transmit information about a vehicle location, an autonomous driving route, a driving plan, and the like to an external server.
  • the external server may predict traffic information data in advance using AI technology or the like based on information collected from the vehicle or autonomous driving vehicles, and may provide the predicted traffic information data to the vehicle or autonomous driving vehicles.
  • the vehicle 25 illustrates a vehicle to which the present invention is applied.
  • the vehicle may also be implemented as a means of transportation, a train, an air vehicle, a ship, and the like.
  • the vehicle 100 may include a communication unit 110 , a control unit 120 , a memory unit 130 , an input/output unit 140a , and a position measurement unit 140b .
  • blocks 110 to 130/140a to 140b correspond to blocks 110 to 130/140 of FIG. 22 , respectively.
  • the communication unit 110 may transmit and receive signals (eg, data, control signals, etc.) with other vehicles or external devices such as a base station.
  • the controller 120 may control components of the vehicle 100 to perform various operations.
  • the memory unit 130 may store data/parameters/programs/codes/commands supporting various functions of the vehicle 100 .
  • the input/output unit 140a may output an AR/VR object based on information in the memory unit 130 .
  • the input/output unit 140a may include a HUD.
  • the position measuring unit 140b may acquire position information of the vehicle 100 .
  • the location information may include absolute location information of the vehicle 100 , location information within a driving line, acceleration information, location information with a surrounding vehicle, and the like.
  • the position measuring unit 140b may include a GPS and various sensors.
  • the communication unit 110 of the vehicle 100 may receive map information, traffic information, and the like from an external server and store it in the memory unit 130 .
  • the position measuring unit 140b may acquire vehicle position information through GPS and various sensors and store it in the memory unit 130 .
  • the controller 120 may generate a virtual object based on map information, traffic information, vehicle location information, and the like, and the input/output unit 140a may display the created virtual object on a window inside the vehicle ( 1410 and 1420 ).
  • the controller 120 may determine whether the vehicle 100 is normally operating within the driving line based on the vehicle location information. When the vehicle 100 deviates from the driving line abnormally, the controller 120 may display a warning on the windshield of the vehicle through the input/output unit 140a.
  • control unit 120 may broadcast a warning message regarding driving abnormality to surrounding vehicles through the communication unit 110 .
  • control unit 120 may transmit the location information of the vehicle and information on driving/vehicle abnormality to a related organization through the communication unit 110 .
  • the XR device may be implemented as an HMD, a head-up display (HUD) provided in a vehicle, a television, a smart phone, a computer, a wearable device, a home appliance, a digital signage, a vehicle, a robot, and the like.
  • HMD head-up display
  • a television a smart phone
  • a computer a wearable device
  • a home appliance a digital signage
  • a vehicle a robot, and the like.
  • the XR device 100a may include a communication unit 110 , a control unit 120 , a memory unit 130 , an input/output unit 140a , a sensor unit 140b , and a power supply unit 140c .
  • blocks 110 to 130/140a to 140c correspond to blocks 110 to 130/140 of FIG. 22 , respectively.
  • the communication unit 110 may transmit/receive signals (eg, media data, control signals, etc.) to/from external devices such as other wireless devices, portable devices, or media servers.
  • Media data may include images, images, and sounds.
  • the controller 120 may perform various operations by controlling the components of the XR device 100a.
  • the controller 120 may be configured to control and/or perform procedures such as video/image acquisition, (video/image) encoding, and metadata generation and processing.
  • the memory unit 130 may store data/parameters/programs/codes/commands necessary for driving the XR device 100a/creating an XR object.
  • the input/output unit 140a may obtain control information, data, and the like from the outside, and may output the generated XR object.
  • the input/output unit 140a may include a camera, a microphone, a user input unit, a display unit, a speaker, and/or a haptic module.
  • the sensor unit 140b may obtain an XR device state, surrounding environment information, user information, and the like.
  • the sensor unit 140b may include a proximity sensor, an illuminance sensor, an acceleration sensor, a magnetic sensor, a gyro sensor, an inertial sensor, an RGB sensor, an IR sensor, a fingerprint recognition sensor, an ultrasonic sensor, an optical sensor, a microphone, and/or a radar. there is.
  • the power supply unit 140c supplies power to the XR device 100a, and may include a wired/wireless charging circuit, a battery, and the like.
  • the memory unit 130 of the XR device 100a may include information (eg, data, etc.) necessary for generating an XR object (eg, AR/VR/MR object).
  • the input/output unit 140a may obtain a command to operate the XR device 100a from the user, and the controller 120 may drive the XR device 100a according to the user's driving command. For example, when the user intends to watch a movie or news through the XR device 100a, the controller 120 transmits the content request information to another device (eg, the mobile device 100b) through the communication unit 130 or can be sent to the media server.
  • another device eg, the mobile device 100b
  • the communication unit 130 may download/stream contents such as movies and news from another device (eg, the portable device 100b) or a media server to the memory unit 130 .
  • the controller 120 controls and/or performs procedures such as video/image acquisition, (video/image) encoding, and metadata generation/processing for the content, and is acquired through the input/output unit 140a/sensor unit 140b It is possible to generate/output an XR object based on information about one surrounding space or a real object.
  • the XR device 100a is wirelessly connected to the portable device 100b through the communication unit 110 , and the operation of the XR device 100a may be controlled by the portable device 100b.
  • the portable device 100b may operate as a controller for the XR device 100a.
  • the XR device 100a may obtain 3D location information of the portable device 100b, and then generate and output an XR object corresponding to the portable device 100b.
  • Robots can be classified into industrial, medical, home, military, etc. depending on the purpose or field of use.
  • the robot 100 may include a communication unit 110 , a control unit 120 , a memory unit 130 , an input/output unit 140a , a sensor unit 140b , and a driving unit 140c .
  • blocks 110 to 130/140a to 140c correspond to blocks 110 to 130/140 of FIG. 22 , respectively.
  • the communication unit 110 may transmit/receive signals (eg, driving information, control signals, etc.) with external devices such as other wireless devices, other robots, or control servers.
  • the controller 120 may perform various operations by controlling the components of the robot 100 .
  • the memory unit 130 may store data/parameters/programs/codes/commands supporting various functions of the robot 100 .
  • the input/output unit 140a may obtain information from the outside of the robot 100 and may output information to the outside of the robot 100 .
  • the input/output unit 140a may include a camera, a microphone, a user input unit, a display unit, a speaker, and/or a haptic module.
  • the sensor unit 140b may obtain internal information, surrounding environment information, user information, and the like of the robot 100 .
  • the sensor unit 140b may include a proximity sensor, an illumination sensor, an acceleration sensor, a magnetic sensor, a gyro sensor, an inertial sensor, an IR sensor, a fingerprint recognition sensor, an ultrasonic sensor, an optical sensor, a microphone, a radar, and the like.
  • the driving unit 140c may perform various physical operations such as moving a robot joint. In addition, the driving unit 140c may make the robot 100 travel on the ground or fly in the air.
  • the driving unit 140c may include an actuator, a motor, a wheel, a brake, a propeller, and the like.
  • AI devices include TVs, projectors, smartphones, PCs, laptops, digital broadcasting terminals, tablet PCs, wearable devices, set-top boxes (STBs), radios, washing machines, refrigerators, digital signage, robots, vehicles, etc. It may be implemented in any possible device or the like.
  • the AI device 100 includes a communication unit 110 , a control unit 120 , a memory unit 130 , input/output units 140a/140b , a learning processor unit 140c , and a sensor unit 140d). may include.
  • Blocks 110 to 130/140a to 140d correspond to blocks 110 to 130/140 of FIG. 22, respectively.
  • the communication unit 110 uses wired/wireless communication technology to communicate with other AI devices (eg, FIGS. 19, 100x, 200, 400) or external devices such as the AI server 200 and wired/wireless signals (eg, sensor information, user input, learning). models, control signals, etc.). To this end, the communication unit 110 may transmit information in the memory unit 130 to an external device or transmit a signal received from the external device to the memory unit 130 .
  • AI devices eg, FIGS. 19, 100x, 200, 400
  • wired/wireless signals eg, sensor information, user input, learning). models, control signals, etc.
  • the communication unit 110 may transmit information in the memory unit 130 to an external device or transmit a signal received from the external device to the memory unit 130 .
  • the controller 120 may determine at least one executable operation of the AI device 100 based on information determined or generated using a data analysis algorithm or a machine learning algorithm. In addition, the controller 120 may control the components of the AI device 100 to perform the determined operation. For example, the control unit 120 may request, search, receive, or utilize the data of the learning processor unit 140c or the memory unit 130 , and may be predicted or preferred among at least one executable operation. Components of the AI device 100 may be controlled to execute the operation. In addition, the control unit 120 collects history information including user feedback on the operation contents or operation of the AI device 100 and stores it in the memory unit 130 or the learning processor unit 140c, or the AI server ( 19 and 400) may be transmitted to an external device. The collected historical information may be used to update the learning model.
  • the memory unit 130 may store data supporting various functions of the AI device 100 .
  • the memory unit 130 may store data obtained from the input unit 140a , data obtained from the communication unit 110 , output data of the learning processor unit 140c , and data obtained from the sensing unit 140 .
  • the memory unit 130 may store control information and/or software codes necessary for the operation/execution of the control unit 120 .
  • the input unit 140a may acquire various types of data from the outside of the AI device 100 .
  • the input unit 120 may obtain training data for model learning, input data to which the learning model is applied, and the like.
  • the input unit 140a may include a camera, a microphone, and/or a user input unit.
  • the output unit 140b may generate an output related to sight, hearing, or touch.
  • the output unit 140b may include a display unit, a speaker, and/or a haptic module.
  • the sensing unit 140 may obtain at least one of internal information of the AI device 100 , surrounding environment information of the AI device 100 , and user information by using various sensors.
  • the sensing unit 140 may include a proximity sensor, an illuminance sensor, an acceleration sensor, a magnetic sensor, a gyro sensor, an inertial sensor, an RGB sensor, an IR sensor, a fingerprint recognition sensor, an ultrasonic sensor, an optical sensor, a microphone, and/or a radar. there is.
  • the learning processor unit 140c may train a model composed of an artificial neural network by using the training data.
  • the learning processor unit 140c may perform AI processing together with the learning processor unit of the AI server ( FIGS. 19 and 400 ).
  • the learning processor unit 140c may process information received from an external device through the communication unit 110 and/or information stored in the memory unit 130 .
  • the output value of the learning processor unit 140c may be transmitted to an external device through the communication unit 110 and/or stored in the memory unit 130 .
  • the wireless communication technology implemented in the wireless devices 100 and 200 of the present specification may include a narrowband Internet of Things for low-power communication as well as LTE, NR, and 6G.
  • NB-IoT technology may be an example of LPWAN (Low Power Wide Area Network) technology, and may be implemented in standards such as LTE Cat NB1 and/or LTE Cat NB2, and is limited to the above-mentioned names. not.
  • the wireless communication technology implemented in the wireless devices 100 and 200 of the present specification may perform communication based on the LTE-M technology.
  • the LTE-M technology may be an example of an LPWAN technology, and may be called various names such as enhanced machine type communication (eMTC).
  • eMTC enhanced machine type communication
  • LTE-M technology is 1) LTE CAT 0, 2) LTE Cat M1, 3) LTE Cat M2, 4) LTE non-BL (non-Bandwidth Limited), 5) LTE-MTC, 6) LTE Machine Type Communication, and/or 7) may be implemented in at least one of various standards such as LTE M, and is not limited to the above-described name.
  • the wireless communication technology implemented in the wireless devices 100 and 200 of the present specification is at least one of ZigBee, Bluetooth, and Low Power Wide Area Network (LPWAN) in consideration of low-power communication. It may include any one, and is not limited to the above-mentioned names.
  • the ZigBee technology can create PAN (personal area networks) related to small/low-power digital communication based on various standards such as IEEE 802.15.4, and can be called by various names.
  • the above-described specification can be implemented as computer-readable code on a medium in which a program is recorded.
  • the computer-readable medium includes all types of recording devices in which data readable by a computer system is stored. Examples of computer-readable media include Hard Disk Drive (HDD), Solid State Disk (SSD), Silicon Disk Drive (SDD), ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc.
  • HDD Hard Disk Drive
  • SSD Solid State Disk
  • SDD Silicon Disk Drive
  • ROM Read Only Memory
  • RAM Compact Disk Drive
  • CD-ROM Compact Disk Read Only Memory
  • magnetic tape floppy disk
  • optical data storage device etc.
  • carrier wave eg, transmission over the Internet

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Computational Linguistics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Health & Medical Sciences (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

A communication method for federated learning in which a server derives a final learning result on the basis of a result of learning by a plurality of devices according to an embodiment of the present specification comprises the steps in which: devices receive a scaling factor based on a global model and a weight variance from a server; the devices calculate the weight variance on the basis of the global model; the devices perform quantization on the basis of the weight variance; and the devices transmit the quantized weight variance to the server. A terminal of the present specification may be linked to an artificial intelligence module, a drone (unmanned aerial vehicle (UAV)), a robot, an augmented reality (AR) device, a virtual reality (VR) device, a device related to a 6G service, and the like.

Description

연합학습을 위한 통신방법 및 이를 수행하는 디바이스Communication method for federated learning and device for performing the same
본 명세서는 연합학습을 위한 통신방법 및 이를 수행하는 디바이스에 관한 것으로, 특히 디바이스들로부터 서버로 전송되는 가중치 정보의 사이즈를 줄일 수 있는 통신방법 및 디바이스를 개시하고 있다.The present specification relates to a communication method for federated learning and a device for performing the same, and in particular, discloses a communication method and device capable of reducing the size of weight information transmitted from devices to a server.
인공지능 학습 방법에서, 디바이스들 각각의 프라이버시(Privacy)를 유지하면서 클라우드(Cloud)의 글로벌 모델(Global Model)을 학습시킬 수 있는 연합학습(Federated Learning) 방식이 이용되고 있다. In the AI learning method, a federated learning method capable of learning a global model of the cloud while maintaining the privacy of each device is used.
연합학습(Federated Learning)은 디바이스들이 획득한 데이터를 서버로 전송하지 않기 때문에 데이터 전송량을 줄일 수 있지만, 디바이스들은 가중치 정보를 서버로 전송하여야 한다. 가중치 정보는 학습이 수행되는 라운드마다 전송되어야 하기 때문에, 이로 인한 데이터 소모량도 부담될 수 있다.Federated learning can reduce the amount of data transmission because the data acquired by devices is not transmitted to the server, but devices must transmit weight information to the server. Since the weight information has to be transmitted every round in which learning is performed, data consumption due to this may also be burdened.
본 명세서는 전술한 필요성 및/또는 문제점을 해결하는 것을 목적으로 한다.SUMMARY OF THE INVENTION The present specification aims to solve the above-mentioned needs and/or problems.
또한, 본 명세서는, 데이터 소모량을 줄일 수 있는 연합학습을 위한 통신 방법 및 이를 수행하는 디바이스를 제공하기 위한 것이다.In addition, the present specification is to provide a communication method for federated learning capable of reducing data consumption and a device for performing the same.
본 명세서의 실시 예에 따른 복수의 디바이스들의 학습 결과에 기초하여, 서버가 최종 학습 결과물을 도출하는 연합학습을 위한 통신방법은 디바이스들이 서버로부터 글로벌 모델 및 가중치 변화량에 기초한 스케일링 계수를 제공받는 단계, 다바이스들이 글로벌 모델에 기초하여 가중치 변화량을 산출하는 단계, 디바이스들이 가중치 변화량에 기초하여 양자화를 수행하는 단계 및 디바이스들이 양자화 된 가중치 변화량을 서버로 전송하는 단계를 포함한다.Based on the learning results of a plurality of devices according to an embodiment of the present specification, a communication method for federated learning in which a server derives a final learning result includes the steps of devices receiving a scaling factor based on a global model and a weight change amount from a server; It includes the steps of the devices calculating the weight change amount based on the global model, the devices performing quantization based on the weight change amount, and the devices transmitting the quantized weight change amount to the server.
상기 양자화를 수행하는 단계는 상기 디바이스들 각각이 동일한 값을 갖는 상기 스케일링 계수를 이용하는 것일 수 있다.The performing of the quantization may include using the scaling coefficients having the same value in each of the devices.
상기 디바이스들이 상기 스케일링 계수를 제공받는 단계는, (i-1)(i는 2 이상의 자연수) 번째 라운드에서 상기 가중치 변화량 분포에 기초하여 생성된 상기 스케일링 계수를, i 번째 라운드에서 제공받는 것일 수 있다.The step in which the devices are provided with the scaling factor may include receiving the scaling factor generated based on the distribution of the weight change in the (i-1)-th round (i is a natural number greater than or equal to 2) in the i-th round. .
상기 (i-1)번째 라운드에서 상기 가중치 변화량 분포를 생성하는 단계는, 상기 (i-1)번째 라운드 까지의 라운드 중에서 적어도 하나 이상의 라운드의 상기 가중치 변화량의 절대값을 누적한 누적 분포 함수를 생성하는 단계; 및 상기 누적 분포 함수에서 미리 설정된 경계값 이상의 상기 가중치 변화량의 절대값의 누적 분포를 클립핑하는 단계를 포함할 수 있다.The generating of the weight change distribution in the (i-1)-th round includes generating a cumulative distribution function accumulating the absolute values of the weight change amount of at least one round among rounds up to the (i-1)-th round to do; and clipping the cumulative distribution of the absolute value of the weight change amount greater than or equal to a preset threshold in the cumulative distribution function.
상기 스케일링 계수를 결정하는 단계는 상기 (i-1) 번째 라운드의 손실과 상기 (i-2) 번째 라운드의 손실의 차이값을 계산한 것에 기초하여 손실 변화량을 산출하는 단계; 및 상기 손실 변화량의 크기에 기초하여 스케일링 계수의 크기를 결정하는 단계를 포함하는 것일 수 있다.The determining of the scaling factor may include: calculating a change in loss based on a difference between the loss of the (i-1)-th round and the loss of the (i-2)-th round; and determining the magnitude of the scaling factor based on the magnitude of the change in loss.
상기 스케일링 계수를 결정하는 단계는 미리 설정된 최대 양자화 비트에 대비한 상기 경계값의 크기 보다 작은 범위 내에서 결정하는 것일 수 있다.The determining of the scaling factor may include determining within a range smaller than a size of the boundary value with respect to a preset maximum quantization bit.
상기 스케일링 계수를 결정하는 단계는 상기 경계값의 크기 보다 작은 범위를 둘 이상의 구간으로 구분하고, 구분된 구간에서 각각 서로 다른 스케일링 계수를 생성하는 것일 수 있다.The determining of the scaling factor may include dividing a range smaller than the size of the boundary value into two or more sections, and generating different scaling factors in the divided sections.
상기 양자화를 수행하는 단계는 상기 가중치 변화량을 상기 스케일링 계수 및 가변 가중치 변화량의 곱으로 산출하는 단계; 및 상기 가변 가중치 변화량의 양자화 범위를 산출하는 단계를 포함하는 것일 수 있다.The performing of the quantization may include calculating the weight change amount as a product of the scaling factor and the variable weight change amount; and calculating a quantization range of the variable weight change amount.
상기 양자화 된 가중치 변화량을 서버로 전송하는 단계는 상기 스케일링 계수에 기초하여 가변된 가변 양자화 정보를 전송하는 단계를 더 포함할 수 있다. Transmitting the quantized weight change amount to the server may further include transmitting variable quantization information changed based on the scaling factor.
본 명세서의 실시 예에 따른 서버로부터 제공받은 글로벌 모델에 기초하여 연합학습을 수행하는 디바이스는 서버와의 통신을 위한 트랜시버 및 글로벌 모델에 기초하여 연합학습을 수행하는 프로세서를 포함한다. 프로세서는 서버로부터 글로벌 모델 및 가중치 변화량에 기초한 스케일링 계수를 제공받고, 글로벌 모델에 기초하여 가중치 변화량을 산출하며, 가중치 변화량에 기초하여 양자화를 수행하고, 양자화 된 가중치 변화량을 서버로 전송한다.A device for performing federated learning based on a global model provided from a server according to an embodiment of the present specification includes a transceiver for communication with the server and a processor for performing federated learning based on the global model. The processor receives the global model and a scaling factor based on the weight change amount from the server, calculates the weight change amount based on the global model, performs quantization based on the weight change amount, and transmits the quantized weight change amount to the server.
상기 프로세서는 (i-1)(i는 2 이상의 자연수) 번째 라운드에서 상기 가중치 변화량 분포에 기초하여 생성된 상기 스케일링 계수를, i번째 라운드에서 제공받는 것일 수 있다.The processor may receive, in the i-th round, the scaling factor generated based on the distribution of the weight change amount in the (i-1)-th round (i is a natural number equal to or greater than 2).
상기 (i-1)번째 라운드에서 생성된 상기 가중치 변화량 분포는 상기 (i-1)번째 라운드 이전까지의 라운드 중에서 적어도 하나 이상의 라운드의 상기 가중치 변화량의 절대값을 누적한 누적 분포 함수에서, 미리 설정된 경계값 이상의 상기 가중치 변화량 절대값의 누적 분포를 클립핑하여 생성되는 것일 수 있다.The weight change distribution generated in the (i-1)-th round is a cumulative distribution function that accumulates the absolute values of the weight change amount of at least one round among rounds before the (i-1)-th round, It may be generated by clipping the cumulative distribution of the absolute value of the weight change amount equal to or greater than a threshold value.
상기 스케일링 계수는 상기 (i-1) 번째 라운드의 손실과 상기 (i-2) 번째 라운드의 손실의 차이값을 계산한 것에 기초하여 크기를 결정하는 것일 수 있다.The scaling factor may determine a size based on a difference between the loss of the (i-1)-th round and the loss of the (i-2)-th round.
본 명세서에 관한 이해를 돕기 위해 상세한 설명의 일부로 포함되는, 첨부 도면은 본 명세서에 대한 실시예를 제공하고, 상세한 설명과 함께 본 명세서의 기술적 특징을 설명한다.BRIEF DESCRIPTION OF THE DRAWINGS The accompanying drawings, which are included as a part of the detailed description to help the understanding of the present specification, provide embodiments of the present specification, and together with the detailed description, explain the technical features of the present specification.
도 1은 3GPP 시스템에 이용되는 물리 채널들 및 일반적인 신호 전송을 예시한다.1 illustrates physical channels and general signal transmission used in a 3GPP system.
도 2는 6G 시스템에서 제공 가능한 통신 구조의 일례를 나타낸 도이다.2 is a diagram illustrating an example of a communication structure that can be provided in a 6G system.
도 3은 퍼셉트론 구조를 예시한다.3 illustrates a perceptron structure.
도 4는 다층 퍼셉트론 구조를 예시한다.4 illustrates a multilayer perceptron structure.
도 5는 심층 신경망 구조를 예시한다.5 illustrates a deep neural network structure.
도 6은 컨볼루션 신경망 구조를 예시한다.6 illustrates a convolutional neural network structure.
도 7은 컨볼루션 신경망에서의 필터 연산을 예시한다.7 illustrates a filter operation in a convolutional neural network.
도 8은 순환 루프가 존재하는 신경망 구조를 예시한다.8 illustrates a neural network structure in which a cyclic loop exists.
도 9는 순환 신경망의 동작 구조를 예시한다.9 illustrates the operational structure of a recurrent neural network.
도 10은 전자기 스펙트럼의 일례를 나타낸다.10 shows an example of an electromagnetic spectrum.
도 11은 THz 통신 응용의 일례를 나타낸다.11 shows an example of a THz communication application.
도 12는 실시 예가 적용되는 연합학습을 위한 통신 시스템을 나타내는 도면이다.12 is a diagram illustrating a communication system for federated learning to which an embodiment is applied.
도 13은 실시 예에 따른 연합학습 프로토콜을 나타내는 도면이다. 13 is a diagram illustrating a federated learning protocol according to an embodiment.
도 14는 실시 예에 의한 연합학습을 위한 통신 방법을 나타내는 도면이다.14 is a diagram illustrating a communication method for joint learning according to an embodiment.
도 15는 본 발명의 실시 예에 따른 스케일링 계수를 결정하는 방법을 나타내는 도면이다.15 is a diagram illustrating a method of determining a scaling factor according to an embodiment of the present invention.
도 16은 서버가 생성한 가중치 변화량 누적분포함수의 일례를 나타내는 도면이다.16 is a diagram illustrating an example of a weight change amount cumulative distribution function generated by a server.
도 17은 평균 제곱 오차를 이용하여 구한 손실 함수의 오차 평가의 일례를 나타내는 도면이다.17 is a diagram showing an example of error evaluation of a loss function obtained using a mean square error.
도 18은 디바이스들의 동작을 설명하는 도면이다.18 is a diagram for explaining the operation of devices.
도 19은 본 발명에 적용되는 통신 시스템을 예시한다.19 illustrates a communication system applied to the present invention.
도 20은 본 발명에 적용될 수 있는 무선 기기를 예시한다.20 illustrates a wireless device applicable to the present invention.
도 21는 전송 신호를 위한 신호 처리 회로를 예시한다.21 illustrates a signal processing circuit for a transmission signal.
도 22은 본 발명에 적용되는 무선 기기의 다른 예를 나타낸다.22 shows another example of a wireless device to which the present invention is applied.
도 23는 본 발명에 적용되는 휴대 기기를 예시한다.23 illustrates a portable device to which the present invention is applied.
도 24는 본 발명에 적용되는 차량 또는 자율 주행 차량을 예시한다.24 illustrates a vehicle or an autonomous driving vehicle to which the present invention is applied.
도 25은 본 발명에 적용되는 차량을 예시한다.25 illustrates a vehicle to which the present invention is applied.
도 26는 본 발명에 적용되는 XR 기기를 예시한다.26 illustrates an XR device applied to the present invention.
도 27는 본 발명에 적용되는 로봇을 예시한다.27 illustrates a robot applied to the present invention.
도 28은 본 발명에 적용되는 AI 기기를 예시한다.28 illustrates an AI device applied to the present invention.
이하, 첨부된 도면을 참조하여 본 명세서에 개시된 실시예를 상세히 설명하되, 도면 부호에 관계없이 동일하거나 유사한 구성요소는 동일한 참조 번호를 부여하고 이에 대한 중복되는 설명은 생략하기로 한다. 이하의 설명에서 사용되는 구성요소에 대한 접미사 "모듈" 및 "부"는 명세서 작성의 용이함만이 고려되어 부여되거나 혼용되는 것으로서, 그 자체로 서로 구별되는 의미 또는 역할을 갖는 것은 아니다. 또한, 본 명세서에 개시된 실시예를 설명함에 있어서 관련된 공지 기술에 대한 구체적인 설명이 본 명세서에 개시된 실시예의 요지를 흐릴 수 있다고 판단되는 경우 그 상세한 설명을 생략한다. 또한, 첨부된 도면은 본 명세서에 개시된 실시예를 쉽게 이해할 수 있도록 하기 위한 것일 뿐, 첨부된 도면에 의해 본 명세서에 개시된 기술적 사상이 제한되지 않으며, 본 명세서의 사상 및 기술 범위에 포함되는 모든 변경, 균등물 내지 대체물을 포함하는 것으로 이해되어야 한다. Hereinafter, the embodiments disclosed in the present specification will be described in detail with reference to the accompanying drawings, but the same or similar components are assigned the same reference numbers regardless of reference numerals, and redundant description thereof will be omitted. The suffixes "module" and "part" for components used in the following description are given or mixed in consideration of only the ease of writing the specification, and do not have distinct meanings or roles by themselves. In addition, in describing the embodiments disclosed in the present specification, if it is determined that detailed descriptions of related known technologies may obscure the gist of the embodiments disclosed in the present specification, the detailed description thereof will be omitted. In addition, the accompanying drawings are only for easy understanding of the embodiments disclosed in the present specification, and the technical spirit disclosed in this specification is not limited by the accompanying drawings, and all changes included in the spirit and scope of the present specification , should be understood to include equivalents or substitutes.
제1, 제2 등과 같이 서수를 포함하는 용어는 다양한 구성요소들을 설명하는데 사용될 수 있지만, 상기 구성요소들은 상기 용어들에 의해 한정되지는 않는다. 상기 용어들은 하나의 구성요소를 다른 구성요소로부터 구별하는 목적으로만 사용된다.Terms including an ordinal number such as 1st, 2nd, etc. may be used to describe various elements, but the elements are not limited by the terms. The above terms are used only for the purpose of distinguishing one component from another.
어떤 구성요소가 다른 구성요소에 "연결되어" 있다거나 "접속되어" 있다고 언급된 때에는, 그 다른 구성요소에 직접적으로 연결되어 있거나 또는 접속되어 있을 수도 있지만, 중간에 다른 구성요소가 존재할 수도 있다고 이해되어야 할 것이다. 반면에, 어떤 구성요소가 다른 구성요소에 "직접 연결되어" 있다거나 "직접 접속되어" 있다고 언급된 때에는, 중간에 다른 구성요소가 존재하지 않는 것으로 이해되어야 할 것이다.When an element is referred to as being “connected” or “connected” to another element, it is understood that it may be directly connected or connected to the other element, but other elements may exist in between. it should be On the other hand, when it is said that a certain element is "directly connected" or "directly connected" to another element, it should be understood that the other element does not exist in the middle.
단수의 표현은 문맥상 명백하게 다르게 뜻하지 않는 한, 복수의 표현을 포함한다.The singular expression includes the plural expression unless the context clearly dictates otherwise.
본 출원에서, "포함한다" 또는 "가지다" 등의 용어는 명세서상에 기재된 특징, 숫자, 단계, 동작, 구성요소, 부품 또는 이들을 조합한 것이 존재함을 지정하려는 것이지, 하나 또는 그 이상의 다른 특징들이나 숫자, 단계, 동작, 구성요소, 부품 또는 이들을 조합한 것들의 존재 또는 부가 가능성을 미리 배제하지 않는 것으로 이해되어야 한다.In the present application, terms such as “comprises” or “have” are intended to designate that a feature, number, step, operation, component, part, or combination thereof described in the specification exists, but one or more other features It should be understood that this does not preclude the existence or addition of numbers, steps, operations, components, parts, or combinations thereof.
이하의 기술은 CDMA, FDMA, TDMA, OFDMA, SC-FDMA 등과 같은 다양한 무선 접속 시스템에 사용될 수 있다. CDMA는 UTRA(Universal Terrestrial Radio Access)나 CDMA2000과 같은 무선 기술로 구현될 수 있다. TDMA는 GSM(Global System for Mobile communications)/GPRS(General Packet Radio Service)/EDGE(Enhanced Data Rates for GSM Evolution)와 같은 무선 기술로 구현될 수 있다. OFDMA는 IEEE 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE 802-20, E-UTRA(Evolved UTRA) 등과 같은 무선 기술로 구현될 수 있다. UTRA는 UMTS(Universal Mobile Telecommunications System)의 일부이다. 3GPP(3rd Generation Partnership Project) LTE(Long Term Evolution)은 E-UTRA를 사용하는 E-UMTS(Evolved UMTS)의 일부이고 LTE-A(Advanced)/LTE-A pro는 3GPP LTE의 진화된 버전이다. 3GPP NR(New Radio or New Radio Access Technology)는 3GPP LTE/LTE-A/LTE-A pro의 진화된 버전이다. 3GPP 6G는 3GPP NR의 진화된 버전일 수 있다.The following techniques can be used in various radio access systems such as CDMA, FDMA, TDMA, OFDMA, SC-FDMA, and the like. CDMA may be implemented with a radio technology such as Universal Terrestrial Radio Access (UTRA) or CDMA2000. TDMA may be implemented with a radio technology such as Global System for Mobile communications (GSM)/General Packet Radio Service (GPRS)/Enhanced Data Rates for GSM Evolution (EDGE). OFDMA may be implemented with a radio technology such as IEEE 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE 802-20, Evolved UTRA (E-UTRA), and the like. UTRA is part of the Universal Mobile Telecommunications System (UMTS). 3GPP (3rd Generation Partnership Project) Long Term Evolution (LTE) is a part of Evolved UMTS (E-UMTS) using E-UTRA and LTE-A (Advanced)/LTE-A pro is an evolved version of 3GPP LTE. 3GPP NR (New Radio or New Radio Access Technology) is an evolved version of 3GPP LTE/LTE-A/LTE-A pro. 3GPP 6G may be an evolved version of 3GPP NR.
설명을 명확하게 하기 위해, 3GPP 통신 시스템(예, LTE, NR 등)을 기반으로 설명하지만 본 발명의 기술적 사상이 이에 제한되는 것은 아니다. LTE는 3GPP TS 36.xxx Release 8 이후의 기술을 의미한다. 세부적으로, 3GPP TS 36.xxx Release 10 이후의 LTE 기술은 LTE-A로 지칭되고, 3GPP TS 36.xxx Release 13 이후의 LTE 기술은 LTE-A pro로 지칭된다. 3GPP NR은 TS 38.xxx Release 15 이후의 기술을 의미한다. 3GPP 6G는 TS Release 17 및/또는 Release 18 이후의 기술을 의미할 수 있다. "xxx"는 표준 문서 세부 번호를 의미한다. LTE/NR/6G는 3GPP 시스템으로 통칭될 수 있다. 본 발명의 설명에 사용된 배경기술, 용어, 약어 등에 관해서는 본 발명 이전에 공개된 표준 문서에 기재된 사항을 참조할 수 있다. 예를 들어, 다음 문서를 참조할 수 있다.For clarity of explanation, although description is based on a 3GPP communication system (eg, LTE, NR, etc.), the technical spirit of the present invention is not limited thereto. LTE refers to technology after 3GPP TS 36.xxx Release 8. In detail, LTE technology after 3GPP TS 36.xxx Release 10 is referred to as LTE-A, and LTE technology after 3GPP TS 36.xxx Release 13 is referred to as LTE-A pro. 3GPP NR refers to technology after TS 38.xxx Release 15. 3GPP 6G may refer to technology after TS Release 17 and/or Release 18. "xxx" stands for standard document detail number. LTE/NR/6G may be collectively referred to as a 3GPP system. For background art, terms, abbreviations, etc. used in the description of the present invention, reference may be made to matters described in standard documents published before the present invention. For example, you can refer to the following documents:
3GPP LTE3GPP LTE
- 36.211: Physical channels and modulation- 36.211: Physical channels and modulation
- 36.212: Multiplexing and channel coding- 36.212: Multiplexing and channel coding
- 36.213: Physical layer procedures- 36.213: Physical layer procedures
- 36.300: Overall description- 36.300: Overall description
- 36.331: Radio Resource Control (RRC)- 36.331: Radio Resource Control (RRC)
3GPP NR3GPP NR
- 38.211: Physical channels and modulation- 38.211: Physical channels and modulation
- 38.212: Multiplexing and channel coding- 38.212: Multiplexing and channel coding
- 38.213: Physical layer procedures for control- 38.213: Physical layer procedures for control
- 38.214: Physical layer procedures for data- 38.214: Physical layer procedures for data
- 38.300: NR and NG-RAN Overall Description- 38.300: NR and NG-RAN Overall Description
- 38.331: Radio Resource Control (RRC) protocol specification- 38.331: Radio Resource Control (RRC) protocol specification
물리 채널 및 프레임 구조Physical Channels and Frame Structure
물리 채널 및 일반적인 신호 전송Physical channels and general signal transmission
도 1은 3GPP 시스템에 이용되는 물리 채널들 및 일반적인 신호 전송을 예시한다. 무선 통신 시스템에서 단말은 기지국으로부터 하향링크(Downlink, DL)를 통해 정보를 수신하고, 단말은 기지국으로 상향링크(Uplink, UL)를 통해 정보를 전송한다. 기지국과 단말이 송수신하는 정보는 데이터 및 다양한 제어 정보를 포함하고, 이들이 송수신 하는 정보의 종류/용도에 따라 다양한 물리 채널이 존재한다.1 illustrates physical channels and general signal transmission used in a 3GPP system. In a wireless communication system, a terminal receives information through a downlink (DL) from a base station, and the terminal transmits information through an uplink (UL) to the base station. Information transmitted and received between the base station and the terminal includes data and various control information, and various physical channels exist according to the type/use of the information they transmit and receive.
단말은 전원이 켜지거나 새로이 셀에 진입한 경우 기지국과 동기를 맞추는 등의 초기 셀 탐색(Initial cell search) 작업을 수행한다(S11). 이를 위해, 단말은 기지국으로부터 주 동기 신호(Primary Synchronization Signal, PSS) 및 부 동기 신호(Secondary Synchronization Signal, SSS)을 수신하여 기지국과 동기를 맞추고, 셀 ID 등의 정보를 획득할 수 있다. 그 후, 단말은 기지국으로부터 물리 방송 채널(Physical Broadcast Channel, PBCH)를 수신하여 셀 내 방송 정보를 획득할 수 있다. 한편, 단말은 초기 셀 탐색 단계에서 하향링크 참조 신호(Downlink Reference Signal, DL RS)를 수신하여 하향링크 채널 상태를 확인할 수 있다.When the terminal is powered on or newly enters a cell, the terminal performs an initial cell search operation, such as synchronizing with the base station (S11). To this end, the terminal receives a primary synchronization signal (PSS) and a secondary synchronization signal (SSS) from the base station, synchronizes with the base station, and obtains information such as a cell ID. Thereafter, the terminal may receive a physical broadcast channel (PBCH) from the base station to obtain intra-cell broadcast information. On the other hand, the UE may receive a downlink reference signal (DL RS) in the initial cell search step to check the downlink channel state.
초기 셀 탐색을 마친 단말은 물리 하향링크 제어 채널(Physical Downlink Control Channel, PDCCH) 및 상기 PDCCH에 실린 정보에 따라 물리 하향링크 공유 채널(Physical Downlink Control Channel; PDSCH)을 수신함으로써 좀더 구체적인 시스템 정보를 획득할 수 있다(S12).After the initial cell search, the UE receives a Physical Downlink Control Channel (PDCCH) and a Physical Downlink Control Channel (PDSCH) according to information carried on the PDCCH to obtain more specific system information. It can be done (S12).
한편, 기지국에 최초로 접속하거나 신호 송신을 위한 무선 자원이 없는 경우, 단말은 기지국에 대해 임의 접속 과정(Random Access Procedure, RACH)을 수행할 수 있다(S13 내지 S16). 이를 위해, 단말은 물리 임의 접속 채널(Physical Random Access Channel, PRACH)을 통해 특정 시퀀스를 프리앰블로 송신하고(S13 및 S15), PDCCH 및 대응하는 PDSCH를 통해 프리앰블에 대한 응답 메시지((RAR(Random Access Response) message)를 수신할 수 있다. 경쟁 기반 RACH의 경우, 추가적으로 충돌 해결 절차(Contention Resolution Procedure)를 수행할 수 있다(S16).On the other hand, when first accessing the base station or there is no radio resource for signal transmission, the terminal may perform a random access procedure (RACH) for the base station (S13 to S16). To this end, the UE transmits a specific sequence as a preamble through a Physical Random Access Channel (PRACH) (S13 and S15), and a response message to the preamble through the PDCCH and the corresponding PDSCH ((Random Access (RAR)) Response) message) In the case of contention-based RACH, a contention resolution procedure may be additionally performed (S16).
상술한 바와 같은 절차를 수행한 단말은 이후 일반적인 상/하향링크 신호 송신 절차로서 PDCCH/PDSCH 수신(S17) 및 물리 상향링크 공유 채널(Physical Uplink Shared Channel, PUSCH)/물리 상향링크 제어 채널(Physical Uplink Control Channel; PUCCH) 송신(S18)을 수행할 수 있다. 특히 단말은 PDCCH를 통하여 하향링크 제어 정보(Downlink Control Information, DCI)를 수신할 수 있다. 여기서, DCI는 단말에 대한 자원 할당 정보와 같은 제어 정보를 포함하며, 사용 목적에 따라 포맷이 서로 다르게 적용될 수 있다. After performing the procedure as described above, the UE performs PDCCH/PDSCH reception (S17) and Physical Uplink Shared Channel (PUSCH)/Physical Uplink Control Channel (Physical Uplink) as a general uplink/downlink signal transmission procedure. Control Channel (PUCCH) transmission (S18) may be performed. In particular, the UE may receive downlink control information (DCI) through the PDCCH. Here, the DCI includes control information such as resource allocation information for the terminal, and different formats may be applied according to the purpose of use.
한편, 단말이 상향링크를 통해 기지국에 송신하는 또는 단말이 기지국으로부터 수신하는 제어 정보는 하향링크/상향링크 ACK/NACK 신호, CQI(Channel Quality Indicator), PMI(Precoding Matrix 인덱스), RI(Rank Indicator) 등을 포함할 수 있다. 단말은 상술한 CQI/PMI/RI 등의 제어 정보를 PUSCH 및/또는 PUCCH를 통해 송신할 수 있다.On the other hand, the control information that the terminal transmits to the base station through the uplink or the terminal receives from the base station includes a downlink/uplink ACK/NACK signal, a channel quality indicator (CQI), a precoding matrix index (PMI), and a rank indicator (RI). ) and the like. The UE may transmit the above-described control information such as CQI/PMI/RI through PUSCH and/or PUCCH.
상향링크 및 하향링크 채널의 구조Structures of uplink and downlink channels
하향링크 채널 구조Downlink Channel Structure
기지국은 후술하는 하향링크 채널을 통해 관련 신호를 단말에게 전송하고, 단말은 후술하는 하향링크 채널을 통해 관련 신호를 기지국으로부터 수신한다.The base station transmits a related signal to the terminal through a downlink channel to be described later, and the terminal receives the related signal from the base station through a downlink channel to be described later.
(1) 물리 하향링크 공유 채널(PDSCH)(1) Physical Downlink Shared Channel (PDSCH)
PDSCH는 하향링크 데이터(예, DL-shared channel transport block, DL-SCH TB)를 운반하고, QPSK(Quadrature Phase Shift Keying), 16 QAM(Quadrature Amplitude Modulation), 64 QAM, 256 QAM 등의 변조 방법이 적용된다. TB를 인코딩하여 코드워드(codeword)가 생성된다. PDSCH는 다수의 코드워드들을 나를 수 있다. 코드워드(codeword) 별로 스크램블링(scrambling) 및 변조 매핑(modulation mapping)이 수행되고, 각 코드워드로부터 생성된 변조 심볼들은 하나 이상의 레이어로 매핑된다(Layer mapping). 각 레이어는 DMRS(Demodulation Reference Signal)과 함께 자원에 매핑되어 OFDM 심볼 신호로 생성되고, 해당 안테나 포트를 통해 전송된다.PDSCH carries downlink data (eg, DL-shared channel transport block, DL-SCH TB), and modulation methods such as Quadrature Phase Shift Keying (QPSK), 16 Quadrature Amplitude Modulation (QAM), 64 QAM, and 256 QAM are available. applies. A codeword is generated by encoding the TB. A PDSCH can carry multiple codewords. Scrambling and modulation mapping are performed for each codeword, and modulation symbols generated from each codeword are mapped to one or more layers (Layer mapping). Each layer is mapped to a resource together with a demodulation reference signal (DMRS), is generated as an OFDM symbol signal, and is transmitted through a corresponding antenna port.
(2) 물리 하향링크 제어 채널(PDCCH)(2) Physical Downlink Control Channel (PDCCH)
PDCCH는 하향링크 제어 정보(DCI)를 운반하고 QPSK 변조 방법 등이 적용된다. 하나의 PDCCH는 AL(Aggregation Level)에 따라 1, 2, 4, 8, 16 개 등의 CCE(Control Channel Element)로 구성된다. 하나의 CCE는 6개의 REG(Resource Element Group)로 구성된다. 하나의 REG는 하나의 OFDM 심볼과 하나의 (P)RB로 정의된다. The PDCCH carries downlink control information (DCI) and a QPSK modulation method is applied. One PDCCH is composed of 1, 2, 4, 8, 16 CCEs (Control Channel Elements) according to an Aggregation Level (AL). One CCE consists of six REGs (Resource Element Groups). One REG is defined as one OFDM symbol and one (P)RB.
단말은 PDCCH 후보들의 세트에 대한 디코딩(일명, 블라인드 디코딩)을 수행하여 PDCCH를 통해 전송되는 DCI를 획득한다. 단말이 디코딩하는 PDCCH 후보들의 세트는 PDCCH 검색 공간(Search Space) 세트라 정의한다. 검색 공간 세트는 공통 검색 공간 (common search space) 또는 단말-특정 검색 공간 (UE-specific search space)일 수 있다. 단말은 MIB 또는 상위 계층 시그널링에 의해 설정된 하나 이상의 검색 공간 세트 내 PDCCH 후보를 모니터링하여 DCI를 획득할 수 있다. The UE obtains DCI transmitted through the PDCCH by performing decoding (aka, blind decoding) on the set of PDCCH candidates. A set of PDCCH candidates decoded by the UE is defined as a PDCCH search space set. The search space set may be a common search space or a UE-specific search space. The UE may acquire DCI by monitoring PDCCH candidates in one or more search space sets configured by MIB or higher layer signaling.
상향링크 채널 구조Uplink Channel Structure
단말은 후술하는 상향링크 채널을 통해 관련 신호를 기지국으로 전송하고, 기지국은 후술하는 상향링크 채널을 통해 관련 신호를 단말로부터 수신한다.The terminal transmits a related signal to the base station through an uplink channel to be described later, and the base station receives the related signal from the terminal through an uplink channel to be described later.
(1) 물리 상향링크 공유 채널(PUSCH)(1) Physical Uplink Shared Channel (PUSCH)
PUSCH는 상향링크 데이터(예, UL-shared channel transport block, UL-SCH TB) 및/또는 상향링크 제어 정보(UCI)를 운반하고, CP-OFDM (Cyclic Prefix - Orthogonal Frequency Division Multiplexing) 파형(waveform), DFT-s-OFDM (Discrete Fourier Transform - spread - Orthogonal Frequency Division Multiplexing) 파형 등에 기초하여 전송된다. PUSCH가 DFT-s-OFDM 파형에 기초하여 전송되는 경우, 단말은 변환 프리코딩(transform precoding)을 적용하여 PUSCH를 전송한다. 일 예로, 변환 프리코딩이 불가능한 경우(예, transform precoding is disabled) 단말은 CP-OFDM 파형에 기초하여 PUSCH를 전송하고, 변환 프리코딩이 가능한 경우(예, transform precoding is enabled) 단말은 CP-OFDM 파형 또는 DFT-s-OFDM 파형에 기초하여 PUSCH를 전송할 수 있다. PUSCH 전송은 DCI 내 UL 그랜트에 의해 동적으로 스케줄링 되거나, 상위 계층(예, RRC) 시그널링 (및/또는 Layer 1(L1) 시그널링(예, PDCCH))에 기초하여 반-정적(semi-static)으로 스케줄링 될 수 있다(configured grant). PUSCH 전송은 코드북 기반 또는 비-코드북 기반으로 수행될 수 있다.PUSCH carries uplink data (eg, UL-shared channel transport block, UL-SCH TB) and/or uplink control information (UCI), and CP-OFDM (Cyclic Prefix - Orthogonal Frequency Division Multiplexing) waveform (waveform) , DFT-s-OFDM (Discrete Fourier Transform - spread - Orthogonal Frequency Division Multiplexing) is transmitted based on the waveform. When the PUSCH is transmitted based on the DFT-s-OFDM waveform, the UE transmits the PUSCH by applying transform precoding. For example, when transform precoding is not possible (eg, transform precoding is disabled), the UE transmits a PUSCH based on the CP-OFDM waveform, and when transform precoding is possible (eg, transform precoding is enabled), the UE transmits the CP-OFDM PUSCH may be transmitted based on a waveform or a DFT-s-OFDM waveform. PUSCH transmission is dynamically scheduled by a UL grant in DCI, or based on higher layer (eg, RRC) signaling (and/or Layer 1 (L1) signaling (eg, PDCCH)) semi-statically. Can be scheduled (configured grant). PUSCH transmission may be performed on a codebook-based or non-codebook-based basis.
(2) 물리 상향링크 제어 채널(PUCCH)(2) Physical Uplink Control Channel (PUCCH)
PUCCH는 상향링크 제어 정보, HARQ-ACK 및/또는 스케줄링 요청(SR)을 운반하고, PUCCH 전송 길이에 따라 다수의 PUCCH들로 구분될 수 있다.The PUCCH carries uplink control information, HARQ-ACK and/or a scheduling request (SR), and may be divided into a plurality of PUCCHs according to the PUCCH transmission length.
6G 시스템 일반6G system general
6G (무선통신) 시스템은 (i) 디바이스 당 매우 높은 데이터 속도, (ii) 매우 많은 수의 연결된 디바이스들, (iii) 글로벌 연결성(global connectivity), (iv) 매우 낮은 지연, (v) 배터리-프리(battery-free) IoT 디바이스들의 에너지 소비를 낮추고, (vi) 초고신뢰성 연결, (vii) 머신 러닝 능력을 가지는 연결된 지능 등에 목적이 있다. 6G 시스템의 비젼은 intelligent connectivity, deep connectivity, holographic connectivity, ubiquitous connectivity와 같은 4가지 측면일 수 있으며, 6G 시스템은 아래 표 1과 같은 요구 사항을 만족시킬 수 있다. 즉, 표 1은 6G 시스템의 요구 사항의 일례를 나타낸 표이다.6G (wireless) systems have (i) very high data rates per device, (ii) very large number of connected devices, (iii) global connectivity, (iv) very low latency, (v) battery- It aims to reduce energy consumption of battery-free IoT devices, (vi) ultra-reliable connections, and (vii) connected intelligence with machine learning capabilities. The vision of the 6G system can be in four aspects: intelligent connectivity, deep connectivity, holographic connectivity, and ubiquitous connectivity, and the 6G system can satisfy the requirements shown in Table 1 below. That is, Table 1 is a table showing an example of the requirements of the 6G system.
Figure PCTKR2020011878-appb-img-000001
Figure PCTKR2020011878-appb-img-000001
6G 시스템은 Enhanced mobile broadband (eMBB), Ultra-reliable low latency communications (URLLC), massive machine-type communication (mMTC), AI integrated communication, Tactile internet, High throughput, High network capacity, High energy efficiency, Low backhaul and access network congestion, Enhanced data security와 같은 핵심 요소(key factor)들을 가질 수 있다.도 2는 6G 시스템에서 제공 가능한 통신 구조의 일례를 나타낸 도이다.6G systems include Enhanced mobile broadband (eMBB), Ultra-reliable low latency communications (URLLC), massive machine-type communication (mMTC), AI integrated communication, Tactile internet, High throughput, High network capacity, High energy efficiency, Low backhaul and It may have key factors such as access network congestion and enhanced data security. FIG. 2 is a diagram showing an example of a communication structure that can be provided in a 6G system.
6G 시스템은 5G 무선통신 시스템보다 50배 더 높은 동시 무선통신 연결성을 가질 것으로 예상된다. 5G의 key feature인 URLLC는 6G 통신에서 1ms보다 적은 단-대-단(end-to-end) 지연을 제공함으로써 보다 더 주요한 기술이 될 것이다. 6G 시스템은 자주 사용되는 영역 스펙트럼 효율과 달리 체적 스펙트럼 효율이 훨씬 우수할 것이다. 6G 시스템은 매우 긴 배터리 수명과 에너지 수확을 위한 고급 배터리 기술을 제공할 수 있어, 6G 시스템에서 모바일 디바이스들은 별도로 충전될 필요가 없을 것이다. 6G에서 새로운 네트워크 특성들은 다음과 같을 수 있다.6G systems are expected to have 50 times higher simultaneous wireless connectivity than 5G wireless communication systems. URLLC, a key feature of 5G, will become an even more important technology by providing an end-to-end delay of less than 1ms in 6G communication. 6G systems will have much better volumetric spectral efficiencies as opposed to frequently used areal spectral efficiencies. The 6G system can provide very long battery life and advanced battery technology for energy harvesting, so mobile devices will not need to be charged separately in the 6G system. New network characteristics in 6G may be as follows.
- 위성 통합 네트워크(Satellites integrated network): 글로벌 모바일 집단을 제공하기 위해 6G는 위성과 통합될 것으로 예상된다. 지상파, 위성 및 공중 네트워크를 하나의 무선통신 시스템으로 통합은 6G에 매우 중요하다.- Satellites integrated network: 6G is expected to be integrated with satellites to provide a global mobile population. The integration of terrestrial, satellite and public networks into one wireless communication system is very important for 6G.
- 연결된 인텔리전스(Connected intelligence): 이전 세대의 무선 통신 시스템과 달리 6G는 혁신적이며, "연결된 사물"에서 "연결된 지능"으로 무선 진화가 업데이트될 것이다. AI는 통신 절차의 각 단계(또는 후술할 신호 처리의 각 절차)에서 적용될 수 있다.- Connected intelligence: Unlike previous generations of wireless communication systems, 6G is revolutionary and will update the evolution of wireless from “connected things” to “connected intelligence”. AI may be applied in each step of a communication procedure (or each procedure of signal processing to be described later).
- 무선 정보 및 에너지 전달의 완벽한 통합(Seamless integration wireless information and energy transfer): 6G 무선 네트워크는 스마트폰들과 센서들과 같이 디바이스들의 배터리를 충전하기 위해 전력을 전달할 것이다. 그러므로, 무선 정보 및 에너지 전송 (WIET)은 통합될 것이다.- Seamless integration wireless information and energy transfer: The 6G wireless network will deliver power to charge the batteries of devices such as smartphones and sensors. Therefore, wireless information and energy transfer (WIET) will be integrated.
- 유비쿼터스 슈퍼 3D 연결(Ubiquitous super 3D connectivity): 드론 및 매우 낮은 지구 궤도 위성의 네트워크 및 핵심 네트워크 기능에 접속은 6G 유비쿼터스에서 슈퍼 3D 연결을 만들 것이다.- Ubiquitous super 3D connectivity: access to networks and core network functions of drones and very low Earth orbiting satellites will create super 3D connectivity in 6G ubiquitous.
위와 같은 6G의 새로운 네트워크 특성들에서 몇 가지 일반적인 요구 사항은 다음과 같을 수 있다.In the above new network characteristics of 6G, some general requirements may be as follows.
- 스몰 셀 네트워크(small cell networks): 스몰 셀 네트워크의 아이디어는 셀룰러 시스템에서 처리량, 에너지 효율 및 스펙트럼 효율 향상의 결과로 수신 신호 품질을 향상시키기 위해 도입되었다. 결과적으로, 스몰 셀 네트워크는 5G 및 비욘드 5G (5GB) 이상의 통신 시스템에 필수적인 특성이다. 따라서, 6G 통신 시스템 역시 스몰 셀 네트워크의 특성을 채택한다.- Small cell networks: The idea of small cell networks was introduced to improve the received signal quality as a result of improved throughput, energy efficiency and spectral efficiency in cellular systems. As a result, small cell networks are essential characteristics for communication systems beyond 5G and Beyond 5G (5GB). Accordingly, the 6G communication system also adopts the characteristics of the small cell network.
- 초 고밀도 이기종 네트워크(Ultra-dense heterogeneous network): 초 고밀도 이기종 네트워크들은 6G 통신 시스템의 또 다른 중요한 특성이 될 것이다. 이기종 네트워크로 구성된 멀티-티어 네트워크는 전체 QoS를 개선하고 비용을 줄인다.- Ultra-dense heterogeneous network: Ultra-dense heterogeneous networks will be another important characteristic of 6G communication systems. A multi-tier network composed of heterogeneous networks improves overall QoS and reduces costs.
- 대용량 백홀(High-capacity backhaul): 백홀 연결은 대용량 트래픽을 지원하기 위해 대용량 백홀 네트워크로 특징 지어진다. 고속 광섬유 및 자유 공간 광학 (FSO) 시스템이 이 문제에 대한 가능한 솔루션일 수 있다.- High-capacity backhaul: A backhaul connection is characterized as a high-capacity backhaul network to support high-capacity traffic. High-speed fiber optics and free-space optics (FSO) systems may be possible solutions to this problem.
- 모바일 기술과 통합된 레이더 기술: 통신을 통한 고정밀 지역화(또는 위치 기반 서비스)는 6G 무선통신 시스템의 기능 중 하나이다. 따라서, 레이더 시스템은 6G 네트워크와 통합될 것이다.- Radar technology integrated with mobile technology: High-precision localization (or location-based service) through communication is one of the functions of the 6G wireless communication system. Therefore, the radar system will be integrated with the 6G network.
- 소프트화 및 가상화(Softwarization and virtualization): 소프트화 및 가상화는 유연성, 재구성성 및 프로그래밍 가능성을 보장하기 위해 5GB 네트워크에서 설계 프로세스의 기초가 되는 두 가지 중요한 기능이다. 또한, 공유 물리적 인프라에서 수십억 개의 장치가 공유될 수 있다.- Softwarization and virtualization: Softening and virtualization are two important features that underlie the design process in 5GB networks to ensure flexibility, reconfigurability and programmability. In addition, billions of devices can be shared in a shared physical infrastructure.
6G 시스템의 핵심 구현 기술Core implementation technology of 6G system
인공 지능(Artificial Intelligence)Artificial Intelligence
6G 시스템에 가장 중요하며, 새로 도입될 기술은 AI이다. 4G 시스템에는 AI가 관여하지 않았다. 5G 시스템은 부분 또는 매우 제한된 AI를 지원할 것이다. 그러나, 6G 시스템은 완전히 자동화를 위해 AI가 지원될 것이다. 머신 러닝의 발전은 6G에서 실시간 통신을 위해 보다 지능적인 네트워크를 만들 것이다. 통신에 AI를 도입하면 실시간 데이터 전송이 간소화되고 향상될 수 있다. AI는 수많은 분석을 사용하여 복잡한 대상 작업이 수행되는 방식을 결정할 수 있다. 즉, AI는 효율성을 높이고 처리 지연을 줄일 수 있다.The most important and newly introduced technology for 6G systems is AI. AI was not involved in the 4G system. 5G systems will support partial or very limited AI. However, the 6G system will be AI-enabled for full automation. Advances in machine learning will create more intelligent networks for real-time communication in 6G. Incorporating AI into communications can simplify and enhance real-time data transmission. AI can use numerous analytics to determine how complex target tasks are performed. In other words, AI can increase efficiency and reduce processing delays.
핸드 오버, 네트워크 선택, 자원 스케쥴링과 같은 시간 소모적인 작업은 AI를 사용함으로써 즉시 수행될 수 있다. AI는 M2M, 기계-대-인간 및 인간-대-기계 통신에서도 중요한 역할을 할 수 있다. 또한, AI는 BCI(Brain Computer Interface)에서 신속한 통신이 될 수 있다. AI 기반 통신 시스템은 메타 물질, 지능형 구조, 지능형 네트워크, 지능형 장치, 지능형 인지 라디오(radio), 자체 유지 무선 네트워크 및 머신 러닝에 의해 지원될 수 있다.Time-consuming tasks such as handovers, network selection, and resource scheduling can be performed instantly by using AI. AI can also play an important role in M2M, machine-to-human and human-to-machine communication. In addition, AI can be a rapid communication in BCI (Brain Computer Interface). AI-based communication systems can be supported by metamaterials, intelligent structures, intelligent networks, intelligent devices, intelligent cognitive radios, self-sustaining wireless networks, and machine learning.
최근에는 AI를 무선 통신 시스템과 통합하려고 하는 시도들이 나타나고 있으나, 이는 application layer, network layer 특히, 딥러닝을 wireless resource management and allocation 분야에 집중되어 왔다. 그러나, 이러한 연구는 점점 MAC layer 와 Physical layer로 발전하고 있으며, 특히 물리계층에서 딥러닝을 무선 전송(wireless transmission)과 결합하고자 하는 시도들이 나타나고 있다. AI 기반의 물리계층 전송은, 근본적인 신호 처리 및 통신 메커니즘에 있어서, 전통적인 통신 프레임워크가 아니라 AI 드라이버에 기초한 신호 처리 및 통신 메커니즘을 적용하는 것을 의미한다. 예를 들어, 딥러닝 기반의 채널 코딩 및 디코딩(channel coding and decoding), 딥러닝 기반의 신호 추정(estimation) 및 검출(detection), 딥러닝 기반의 MIMO mechanism, AI 기반의 자원 스케줄링(scheduling) 및 할당(allocation) 등을 포함할 수 있다.Recently, attempts have been made to integrate AI with wireless communication systems, but these have been focused on the application layer and network layer, especially deep learning, in the field of wireless resource management and allocation. However, these studies are gradually developing into the MAC layer and the physical layer, and in particular, attempts to combine deep learning with wireless transmission in the physical layer are appearing. AI-based physical layer transmission means applying a signal processing and communication mechanism based on an AI driver rather than a traditional communication framework in a fundamental signal processing and communication mechanism. For example, deep learning-based channel coding and decoding, deep learning-based signal estimation and detection, deep learning-based MIMO mechanism, AI-based resource scheduling and It may include an allocation (allocation) and the like.
머신 러닝은 채널 추정 및 채널 트래킹을 위해 사용될 수 있으며, DL(downlink)의 물리 계층(physical layer)에서 전력 할당(power allocation), 간섭 제거 (interference cancellation) 등에 사용될 수 있다. 또한, 머신 러닝은 MIMO 시스템에서 안테나 선택, 전력 제어(power control), 심볼 검출(symbol detection) 등에도 사용될 수 있다.Machine learning may be used for channel estimation and channel tracking, and may be used for power allocation, interference cancellation, and the like in a physical layer of a downlink (DL). In addition, machine learning may be used for antenna selection, power control, symbol detection, and the like in a MIMO system.
그러나 물리계층에서의 전송을 위한 DNN의 적용은 아래와 같은 문제점이 있을 수 있다.However, the application of DNN for transmission in the physical layer may have the following problems.
딥러닝 기반의 AI 알고리즘은 훈련 파라미터를 최적화하기 위해 수많은 훈련 데이터가 필요하다. 그러나 특정 채널 환경에서의 데이터를 훈련 데이터로 획득하는데 있어서의 한계로 인해, 오프라인 상에서 많은 훈련 데이터를 사용한다. 이는 특정 채널 환경에서 훈련 데이터에 대한 정적 훈련(static training)은, 무선 채널의 동적 특성 및 다이버시티(diversity) 사이에 모순(contradiction)이 생길 수 있다.Deep learning-based AI algorithms require large amounts of training data to optimize training parameters. However, due to a limitation in acquiring data in a specific channel environment as training data, a lot of training data is used offline. This is because static training on training data in a specific channel environment may cause a contradiction between dynamic characteristics and diversity of a wireless channel.
또한, 현재 딥러닝은 주로 실제 신호(real signal)을 대상으로 한다. 그러나, 무선 통신의 물리 계층의 신호들은 복소 신호(complex signal)이다. 무선 통신 신호의 특성을 매칭시키기 위해 복소 도메인 신호의 검출하는 신경망(neural network)에 대한 연구가 더 필요하다.In addition, current deep learning mainly targets real signals. However, signals of the physical layer of wireless communication are complex signals. In order to match the characteristics of a wireless communication signal, further research on a neural network for detecting a complex domain signal is needed.
이하, 머신 러닝에 대해 보다 구체적으로 살펴본다.Hereinafter, machine learning will be described in more detail.
머신 러닝은 사람이 할 수 있거나 혹은 하기 어려운 작업을 대신해낼 수 있는 기계를 만들어내기 위해 기계를 학습시키는 일련의 동작을 의미한다. 머신 러닝을 위해서는 데이터와 러닝 모델이 필요하다. 머신 러닝에서 데이터의 학습 방법은 크게 3가지 즉, 지도 학습(supervised learning), 비지도 학습(unsupervised learning) 그리고 강화 학습(reinforcement learning)으로 구분될 수 있다.Machine learning refers to a set of actions that trains a machine to create a machine that can perform tasks that humans can or cannot do. Machine learning requires data and a learning model. In machine learning, data learning methods can be roughly divided into three types: supervised learning, unsupervised learning, and reinforcement learning.
신경망 학습은 출력의 오류를 최소화하기 위한 것이다. 신경망 학습은 반복적으로 학습 데이터를 신경망에 입력시키고 학습 데이터에 대한 신경망의 출력과 타겟의 에러를 계산하고, 에러를 줄이기 위한 방향으로 신경망의 에러를 신경망의 출력 레이어에서부터 입력 레이어 방향으로 역전파(backpropagation) 하여 신경망의 각 노드의 가중치를 업데이트하는 과정이다.Neural network learning is to minimize output errors. Neural network learning repeatedly inputs learning data into the neural network, calculates the output and target errors of the neural network for the training data, and backpropagates the neural network error from the output layer of the neural network to the input layer in the direction to reduce the error. ) to update the weight of each node in the neural network.
지도 학습은 학습 데이터에 정답이 라벨링된 학습 데이터를 사용하며 비지도 학습은 학습 데이터에 정답이 라벨링되어 있지 않을 수 있다. 즉, 예를 들어 데이터 분류에 관한 지도 학습의 경우의 학습 데이터는 학습 데이터 각각에 카테고리가 라벨링된 데이터 일 수 있다. 라벨링된 학습 데이터가 신경망에 입력되고 신경망의 출력(카테고리)과 학습 데이터의 라벨을 비교하여 오차(error)가 계산될 수 있다. 계산된 오차는 신경망에서 역방향(즉, 출력 레이어에서 입력 레이어 방향)으로 역전파 되며, 역전파에 따라 신경망의 각 레이어의 각 노드들의 연결 가중치가 업데이트 될 수 있다. 업데이트 되는 각 노드의 연결 가중치는 학습률(learing rate)에 따라 변화량이 결정될 수 있다. 입력 데이터에 대한 신경망의 계산과 에러의 역전파는 학습 사이클(epoch)을 구성할 수 있다. 학습률은 신경망의 학습 사이클의 반복 횟수에 따라 상이하게 적용될 수 있다. 예를 들어, 신경망의 학습 초기에는 높은 학습률을 사용하여 신경망이 빠르게 일정 수준의 성능을 확보하도록 하여 효율성을 높이고, 학습 후기에는 낮은 학습률을 사용하여 정확도를 높일 수 있다Supervised learning uses training data in which the correct answer is labeled in the training data, and in unsupervised learning, the correct answer may not be labeled in the training data. That is, for example, learning data in the case of supervised learning related to data classification may be data in which categories are labeled for each of the training data. The labeled training data is input to the neural network, and an error can be calculated by comparing the output (category) of the neural network with the label of the training data. The calculated error is back propagated in the reverse direction (ie, from the output layer to the input layer) in the neural network, and the connection weight of each node of each layer of the neural network may be updated according to the back propagation. A change amount of the connection weight of each node to be updated may be determined according to a learning rate. The computation of the neural network on the input data and the backpropagation of errors can constitute a learning cycle (epoch). The learning rate may be applied differently depending on the number of repetitions of the learning cycle of the neural network. For example, in the early stage of learning a neural network, a high learning rate can be used to increase the efficiency by allowing the neural network to quickly obtain a certain level of performance, and in the late learning period, a low learning rate can be used to increase the accuracy.
데이터의 특징에 따라 학습 방법은 달라질 수 있다. 예를 들어, 통신 시스템 상에서 송신단에서 전송한 데이터를 수신단에서 정확하게 예측하는 것을 목적으로 하는 경우, 비지도 학습 또는 강화 학습 보다는 지도 학습을 이용하여 학습을 수행하는 것이 바람직하다.The learning method may vary depending on the characteristics of the data. For example, when the purpose of accurately predicting data transmitted from a transmitter in a communication system is at a receiver, it is preferable to perform learning using supervised learning rather than unsupervised learning or reinforcement learning.
러닝 모델은 인간의 뇌에 해당하는 것으로서, 가장 기본적인 선형 모델을 생각할 수 있으나, 인공 신경망(artificial neural networks)와 같은 복잡성이 높은 신경망 구조를 러닝 모델로 사용하는 머신 러닝의 패러다임을 딥러닝(deep learning)이라 한다.The learning model corresponds to the human brain, and the most basic linear model can be considered. ) is called
학습(learning) 방식으로 사용하는 신경망 코어(neural network cord)는 크게 심층 신경망(DNN, deep neural networks), 합성곱 신경망(CNN, convolutional deep neural networks), 순환 신경망(RNN, Recurrent Boltzmann Machine) 방식이 있다.The neural network cord used as a learning method is largely divided into deep neural networks (DNN), convolutional deep neural networks (CNN), and Recurrent Boltzmann Machine (RNN) methods. there is.
인공 신경망(artificial neural network)은 여러 개의 퍼셉트론을 연결한 예시이다.An artificial neural network is an example of connecting several perceptrons.
도 3을 참조하면, 입력 벡터 x=(x1,x2,...,xd) 가 입력되면 각 성분에 가중치(W1,W2,...,Wd)를 곱하고, 그 결과를 모두 합산한 후, 활성함수 σ() 를 적용하는 전체 과정을 퍼셉트론(perceptron)이라 한다. 거대한 인공 신경망 구조는 도 3에 도시한 단순화된 퍼셉트론 구조를 확장하여 입력벡터를 서로 다른 다 차원의 퍼셉트론에 적용할 수도 있다. 설명의 편의를 위해 입력값 또는 출력값을 노드(node)라 칭한다.Referring to FIG. 3, when an input vector x=(x1,x2,...,xd) is input, each component is multiplied by a weight (W1,W2,...,Wd), and after summing all the results, The whole process of applying the activation function σ() is called a perceptron. The huge artificial neural network structure may extend the simplified perceptron structure shown in FIG. 3 to apply input vectors to different multidimensional perceptrons. For convenience of description, an input value or an output value is referred to as a node.
한편, 도 3에 도시된 퍼셉트론 구조는 입력값, 출력값을 기준으로 총 3개의 층(layer)로 구성되는 것으로 설명할 수 있다. 1st layer와 2nd layer 사이에는 (d+1) 차원의 퍼셉트론 H개, 2nd layer와 3rd layer 사이에는 (H+1)차원 퍼셉트론이 K 개 존재하는 인공신경망을 도 4와 같이 표현할 수 있다.Meanwhile, the perceptron structure shown in FIG. 3 can be described as being composed of a total of three layers based on an input value and an output value. An artificial neural network in which H (d+1)-dimensional perceptrons exist between the 1st layer and the 2nd layer and K (H+1)-dimensional perceptrons exist between the 2nd layer and the 3rd layer can be expressed as shown in FIG. 4 .
입력벡터가 위치하는 층을 입력층(input layer), 최종 출력값이 위치하는 층을 출력층(output layer), 입력층과 출력층 사이에 위치하는 모든 층을 은닉층(hidden layer)라 한다. 도 4의 예시는 3개의 층이 개시되나, 실제 인공신경망 층의 개수를 카운트할 때는 입력층을 제외하고 카운트하므로 총 2개의 층으로 볼 수 있다. 인공신경망은 기본 블록의 퍼셉트론을 2차원적으로 연결되어 구성된다.The layer where the input vector is located is called the input layer, the layer where the final output value is located is called the output layer, and all the layers located between the input layer and the output layer are called hidden layers. In the example of FIG. 4 , three layers are disclosed, but when counting the actual number of artificial neural network layers, the input layer is counted except for the input layer, so it can be viewed as a total of two layers. The artificial neural network is constructed by connecting the perceptrons of the basic blocks in two dimensions.
전술한 입력층, 은닉층, 출력층은 다층 퍼셉트론 뿐 아니라 후술할 CNN, RNN 등 다양한 인공신경망 구조에서 공동적으로 적용될 수 있다. 은닉층의 개수가 많아질수록 인공신경망이 깊어진 것이며, 충분히 깊어진 인공신경망을 러닝모델로 사용하는 머신러닝 패러다임을 딥러닝(Deep Learning)이라 한다. 또한 딥러닝을 위해 사용하는 인공신경망을 심층 신경망(DNN: Deep neural network)라 한다.The aforementioned input layer, hidden layer, and output layer can be jointly applied in various artificial neural network structures such as CNN and RNN to be described later as well as multi-layer perceptron. As the number of hidden layers increases, the artificial neural network becomes deeper, and a machine learning paradigm that uses a sufficiently deep artificial neural network as a learning model is called deep learning. Also, an artificial neural network used for deep learning is called a deep neural network (DNN).
도 5에 도시된 심층 신경망은 은닉층+출력층이 8개로 구성된 다층 퍼셉트론이다. 상기 다층 퍼셉트론 구조를 완전 연결 신경망(fully-connected neural network)이라 표현한다. 완전 연결 신경망은 서로 같은 층에 위치하는 노드 간에는 연결 관계가 존재하지 않으며, 인접한 층에 위치한 노드들 간에만 연결 관계가 존재한다. DNN은 완전 연결 신경망 구조를 가지고 다수의 은닉층과 활성함수들의 조합으로 구성되어 입력과 출력 사이의 상관관계 특성을 파악하는데 유용하게 적용될 수 있다. 여기서 상관관계 특성은 입출력의 결합확률(joint probability)을 의미할 수 있다.The deep neural network shown in FIG. 5 is a multilayer perceptron composed of eight hidden layers + output layers. The multi-layered perceptron structure is referred to as a fully-connected neural network. In a fully connected neural network, a connection relationship does not exist between nodes located in the same layer, and a connection relationship exists only between nodes located in adjacent layers. DNN has a fully connected neural network structure and is composed of a combination of multiple hidden layers and activation functions, so it can be usefully applied to identify the correlation characteristics between input and output. Here, the correlation characteristic may mean a joint probability of input/output.
‘한편, 복수의 퍼셉트론을 서로 어떻게 연결하느냐에 따라 전술한 DNN과 다른 다양한 인공 신경망 구조를 형성할 수 있다. ‘On the other hand, depending on how a plurality of perceptrons are connected to each other, various artificial neural network structures different from the aforementioned DNN can be formed.
DNN은 하나의 층 내부에 위치한 노드들이 1차원적의 세로 방향으로 배치되어 있다. 그러나, 도 6는 노드들이 2차원적으로 가로 w개, 세로 h개의 노드가 배치할 경우를 가정할 수 있다(도 6의 컨볼루션 신경망 구조). 이 경우, 하나의 입력노드에서 은닉층으로 이어지는 연결과정에서 연결 하나당 가중치가 부가되므로 총 hХw 개의 가중치를 고려해야한다. 입력층에 hХw 개의 노드가 존재하므로 인접한 두 층 사이에는 총 h 2w 2 개의 가중치가 필요하다.In DNN, nodes located inside one layer are arranged in a one-dimensional vertical direction. However, in FIG. 6 , it may be assumed that the nodes are two-dimensionally arranged with w horizontally and h vertical nodes (convolutional neural network structure of FIG. 6 ). In this case, since a weight is added per connection in the connection process from one input node to the hidden layer, a total of hХw weights must be considered. Since there are hХw nodes in the input layer, a total of h 2 w 2 weights are needed between two adjacent layers.
도 6의 컨볼루션 신경망은 연결개수에 따라 가중치의 개수가 기하급수적으로 증가하는 문제가 있어 인접한 층 간의 모든 모드의 연결을 고려하는 대신, 크기가 작은 필터(filter)가 존재하는 것으로 가정하여 도 7에서와 같이 필터가 겹치는 부분에 대해서는 가중합 및 활성함수 연산을 수행하도록 한다.The convolutional neural network of FIG. 6 has a problem in that the number of weights increases exponentially according to the number of connections, so instead of considering the connection of all modes between adjacent layers, it is assumed that a filter with a small size exists in FIG. 7 As in Fig., the weighted sum and activation function calculations are performed on the overlapping filters.
하나의 필터는 그 크기만큼의 개수에 해당하는 가중치를 가지며, 이미지 상의 어느 특정한 특징을 요인으로 추출하여 출력할 수 있도록 가중치의 학습이 이루어질 수 있다. 도 7에서는 3Х3 크기의 필터가 입력층의 가장 좌측 상단 3Х3 영역에 적용되고, 해당 노드에 대한 가중합 및 활성함수 연산을 수행한 결과 출력값을 z22에 저장한다.One filter has a weight corresponding to the number corresponding to its size, and weight learning can be performed so that a specific feature on an image can be extracted and output as a factor. In FIG. 7 , a filter with a size of 3Х3 is applied to the upper left 3Х3 region of the input layer, and an output value obtained by performing weighted sum and activation function operations on the corresponding node is stored in z22.
상기 필터는 입력층을 스캔하면서 가로, 세로 일정 간격 만큼 이동하면서 가중합 및 활성함수 연산을 수행하고 그 출력값을 현재 필터의 위치에 위치시킨다. 이러한 연산 방식은 컴퓨터 비전(computer vision) 분야에서 이미지에 대한 컨볼루션(convolution) 연산과 유사하여 이러한 구조의 심층 신경망을 컨볼루션 신경망(CNN: convolutional neural network)라 하고, 컨볼루션 연산 결과 생성되는 은닉층을 컨볼루션 층(convolutional layer)라 한다. 또한, 복수의 컨볼루션 층이 존재하는 신경망을 심층 컨볼루션 신경망(DCNN: Deep convolutional)이라 한다.The filter performs weight sum and activation function calculations while moving horizontally and vertically at regular intervals while scanning the input layer, and places the output value at the current filter position. This calculation method is similar to a convolution operation on an image in the field of computer vision, so a deep neural network with such a structure is called a convolutional neural network (CNN), and a hidden layer generated as a result of the convolution operation is called a convolutional layer. Also, a neural network having a plurality of convolutional layers is called a deep convolutional neural network (DCNN).
컨볼루션 층에서는 현재 필터가 위치한 노드에서, 상기 필터가 커버하는 영역에 위치한 노드만을 포괄하여 가중합을 계산함으로써, 가중치의 개수를 줄여줄 수 있다. 이로 인해, 하나의 필터가 로컬(local) 영역에 대한 특징에 집중하도록 이용될 수 있다. 이에 따라 CNN은 2차원 영역 상의 물리적 거리가 중요한 판단 기준이 되는 이미지 데이터 처리에 효과적으로 적용될 수 있다. 한편, CNN은 컨볼루션 층의 직전에 복수의 필터가 적용될 수 있으며, 각 필터의 컨볼루션 연산을 통해 복수의 출력 결과를 생성할 수도 있다.In the convolution layer, the number of weights can be reduced by calculating the weighted sum by including only nodes located in the region covered by the filter in the node where the filter is currently located. Due to this, one filter can be used to focus on features for a local area. Accordingly, CNN can be effectively applied to image data processing in which physical distance in a two-dimensional domain is an important criterion. Meanwhile, in CNN, a plurality of filters may be applied immediately before the convolution layer, and a plurality of output results may be generated through the convolution operation of each filter.
한편, 데이터 속성에 따라 시퀀스(sequence) 특성이 중요한 데이터들이 있을 수 있다. 이러한 시퀀스 데이터들의 길이 가변성, 선후 관계를 고려하여 데이터 시퀀스 상의 원소를 매 시점(timestep) 마다 하나씩 입력하고, 특정 시점에 출력된 은닉층의 출력 벡터(은닉 벡터)를, 시퀀스 상의 바로 다음 원소와 함께 입력하는 방식을 인공 신경망에 적용한 구조를 순환 신경망 구조라 한다.Meanwhile, there may be data whose sequence characteristics are important according to data properties. Considering the length variability and precedence relationship of the sequence data, one element in the data sequence is input at each timestep, and the output vector (hidden vector) of the hidden layer output at a specific time is input together with the next element in the sequence. A structure in which this method is applied to an artificial neural network is called a recurrent neural network structure.
도 8를 참조하면, 순환 신경망(RNN: recurrent neural netwok)은 데이터 시퀀스 상의 어느 시선 t의 원소 (x1(t), x2(t), ,..., xd(t))를 완전 연결 신경망에 입력하는 과정에서, 바로 이전 시점 t-1은 은닉 벡터 (z1(t-1), z2(t*?*1),..., zH(t*?*1))을 함께 입력하여 가중합 및 활성함수를 적용하는 구조이다. 이와 같이 은닉 벡터를 다음 시점으로 전달하는 이유는 앞선 시점들에서의 입력 벡터속 정보들이 현재 시점의 은닉 벡터에 누적된 것으로 간주하기 때문이다.Referring to FIG. 8 , a recurrent neural network (RNN) connects elements (x1(t), x2(t), ,..., xd(t)) of a certain gaze t on a data sequence to a fully connected neural network. In the input process, the previous time point t-1 is weighted by inputting the hidden vectors (z1(t-1), z2(t*?*1),..., zH(t*?*1)) together. and a structure to which an activation function is applied. The reason why the hidden vector is transferred to the next time point in this way is that information in the input vector at previous time points is considered to be accumulated in the hidden vector of the current time point.
도 8을 참조하면, 순환 신경망은 입력되는 데이터 시퀀스에 대하여 소정의 시점 순서대로 동작한다.Referring to FIG. 8 , the recurrent neural network operates in a predetermined time sequence with respect to an input data sequence.
시점 1에서의 입력 벡터 (x1(t), x2(t), ..., xd(t))가 순환 신경망에 입력되었을 때의 은닉 벡터 (z1(1), z2(1), ...,zH(1))가 시점 2의 입력 벡터 (x1(2),x2(2),...,xd(2))와 함께 입력되어 가중합 및 활성 함수를 통해 은닉층의 벡터 (z1(2),z2(2) ,...,zH(2))를 결정한다. 이러한 과정은 시점 2, 시점 3, ,,, 시점 T 까지 반복적으로 수행된다.The hidden vector (z1(1), z2(1), ... ,zH(1)) is input together with the input vector (x1(2),x2(2),...,xd(2)) of time point 2, and then the vector of the hidden layer (z1(2)) through weighted sum and activation functions ),z2(2) ,...,zH(2)). This process is repeatedly performed until time point 2, time point 3, ,, and time point T.
한편, 순환 신경망 내에서 복수의 은닉층이 배치될 경우, 이를 심층 순환 신경망(DRNN: Deep recurrent neural network)라 한다. 순환 신경망은 시퀀스 데이터(예를 들어, 자연어 처리(natural language processing)에 유용하게 적용되도록 설계되어 있다.On the other hand, when a plurality of hidden layers are arranged in a recurrent neural network, this is called a deep recurrent neural network (DRNN). The recurrent neural network is designed to be usefully applied to sequence data (eg, natural language processing).
학습(learning) 방식으로 사용하는 신경망 코어로서 DNN, CNN, RNN 외에 제한 볼츠만 머신(RBM, Restricted Boltzmann Machine), 심층 신뢰 신경망(DBN, deep belief networks), 심층 Q-네트워크(Deep Q-Network)와 같은 다양한 딥 러닝 기법들을 포함하며, 컴퓨터비젼, 음성인식, 자연어처리, 음성/신호처리 등의 분야에 적용될 수 있다.As a neural network core used as a learning method, in addition to DNN, CNN, and RNN, Restricted Boltzmann Machine (RBM), deep belief networks (DBN), Deep Q-Network and It includes various deep learning techniques such as, and can be applied to fields such as computer vision, voice recognition, natural language processing, and voice/signal processing.
최근에는 AI를 무선 통신 시스템과 통합하려고 하는 시도들이 나타나고 있으나, 이는 application layer, network layer 특히, 딥러닝을 wireless resource management and allocation 분야에 집중되어 왔다. 그러나, 이러한 연구는 점점 MAC layer 와 Physical layer로 발전하고 있으며, 특히 물리계층에서 딥러닝을 무선 전송(wireless transmission)과 결합하고자 하는 시도들이 나타나고 있다. AI 기반의 물리계층 전송은, 근본적인 신호 처리 및 통신 메커니즘에 있어서, 전통적인 통신 프레임워크가 아니라 AI 드라이버에 기초한 신호 처리 및 통신 메커니즘을 적용하는 것을 의미한다. 예를 들어, 딥러닝 기반의 채널 코딩 및 디코딩(channel coding and decoding), 딥러닝 기반의 신호 추정(estimation) 및 검출(detection), 딥러닝 기반의 MIMO mechanism, AI 기반의 자원 스케줄링(scheduling) 및 할당(allocation) 등을 포함할 수 있다.Recently, attempts have been made to integrate AI with wireless communication systems, but these have been focused on the application layer and network layer, especially deep learning, in the field of wireless resource management and allocation. However, these studies are gradually developing into the MAC layer and the physical layer, and in particular, attempts to combine deep learning with wireless transmission in the physical layer are appearing. AI-based physical layer transmission means applying a signal processing and communication mechanism based on an AI driver rather than a traditional communication framework in a fundamental signal processing and communication mechanism. For example, deep learning-based channel coding and decoding, deep learning-based signal estimation and detection, deep learning-based MIMO mechanism, AI-based resource scheduling and It may include an allocation (allocation) and the like.
THz(Terahertz) 통신THz (Terahertz) communication
데이터 전송률은 대역폭을 늘려 높일 수 있다. 이것은 넓은 대역폭으로 sub-THz 통신을 사용하고, 진보된 대규모 MIMO 기술을 적용하여 수행될 수 있다. 밀리미터 이하의 방사선으로도 알려진 THz파는 일반적으로 0.03mm-3mm 범위의 해당 파장을 가진 0.1THz와 10THz 사이의 주파수 대역을 나타낸다. 100GHz-300GHz 대역 범위(Sub THz 대역)는 셀룰러 통신을 위한 THz 대역의 주요 부분으로 간주된다. Sub-THz 대역 mmWave 대역 에 추가하면 6G 셀룰러 통신 용량은 늘어난다.. 정의된 THz 대역 중 300GHz-3THz는 원적외선 (IR) 주파수 대역에 있다. 300GHz-3THz 대역은 광 대역의 일부이지만 광 대역의 경계에 있으며, RF 대역 바로 뒤에 있다. 따라서, 이 300 GHz-3 THz 대역은 RF와 유사성을 나타낸다. 도 10는 전자기 스펙트럼의 일례를 나타낸다.The data rate can be increased by increasing the bandwidth. This can be accomplished by using sub-THz communication with a wide bandwidth and applying advanced large-scale MIMO technology. THz waves, also known as sub-millimeter radiation, typically exhibit a frequency band between 0.1 THz and 10 THz with corresponding wavelengths in the range of 0.03 mm-3 mm. The 100GHz-300GHz band range (Sub THz band) is considered a major part of the THz band for cellular communication. Sub-THz band Addition to mmWave band increases 6G cellular communication capacity. Among the defined THz bands, 300GHz-3THz is in the far-infrared (IR) frequency band. The 300GHz-3THz band is part of the broadband, but at the edge of the wideband, just behind the RF band. Thus, this 300 GHz-3 THz band shows similarities to RF. 10 shows an example of an electromagnetic spectrum.
THz파는 RF(Radio Frequency)/밀리미터(mm)와 적외선 대역 사이에 위치하며, (i) 가시광/적외선에 비해 비금속/비분극성 물질을 잘 투과하며 RF/밀리미터파에 비해 파장이 짧아 높은 직진성을 가지며 빔 집속이 가능할 수 있다. 또한, THz파의 광자 에너지는 수 meV에 불과하기 때문에 인체에 무해한 특성이 있다. THz 무선통신에 이용될 것으로 기대되는 주파수 대역은 공기 중 분자 흡수에 의한 전파 손실이 작은 D-밴드(110GHz~170GHz) 혹은 H-밴드(220GHz~325GHz) 대역일 수 있다. THz 무선통신에 대한 표준화 논의는 3GPP 이외에도 IEEE 802.15 THz working group을 중심으로 논의되고 있으며, IEEE 802.15의 Task Group (TG3d, TG3e)에서 발행되는 표준문서는 본 명세서에서 설명되는 내용을 구체화하거나 보충할 수 있다. THz 무선통신은 무선 인식(wireless cognition), 센싱(sensing), 이미징(imaging), 무선 통신(wireless), THz 네비게이션(navigation) 등에 응용될 수 있다. THz 통신의 주요 특성은 (i) 매우 높은 데이터 전송률을 지원하기 위해 광범위하게 사용 가능한 대역폭, (ii) 고주파에서 발생하는 높은 경로 손실 (고 지향성 안테나는 필수 불가결)을 포함한다. 높은 지향성 안테나에서 생성된 좁은 빔 폭은 간섭을 줄인다. THz 신호의 작은 파장은 훨씬 더 많은 수의 안테나 소자가 이 대역에서 동작하는 장치 및 BS에 통합될 수 있게 한다. 이를 통해 범위 제한을 극복할 수 있는 고급 적응형 배열 기술을 사용할 수 있다.THz wave is located between RF (Radio Frequency)/millimeter (mm) and infrared band, (i) It transmits non-metal/non-polar material better than visible light/infrared light, and has a shorter wavelength than RF/millimeter wave, so it has high straightness. Beam focusing may be possible. In addition, since the photon energy of the THz wave is only a few meV, it is harmless to the human body. The frequency band expected to be used for THz wireless communication may be a D-band (110 GHz to 170 GHz) or H-band (220 GHz to 325 GHz) band with low propagation loss due to absorption of molecules in the air. The standardization discussion on THz wireless communication is being discussed centered on the IEEE 802.15 THz working group in addition to 3GPP, and the standard documents issued by the IEEE 802.15 Task Group (TG3d, TG3e) may specify or supplement the content described in this specification. there is. THz wireless communication may be applied to wireless recognition, sensing, imaging, wireless communication, THz navigation, and the like. The main characteristics of THz communication include (i) widely available bandwidth to support very high data rates, and (ii) high path loss occurring at high frequencies (high directional antennas are indispensable). The narrow beamwidth produced by the highly directional antenna reduces interference. The small wavelength of the THz signal allows a much larger number of antenna elements to be integrated into devices and BSs operating in this band. This allows the use of advanced adaptive nesting techniques that can overcome range limitations.
도 11은 THz 통신 응용의 일례를 나타낸다.11 shows an example of a THz communication application.
도 11에 도시된 바와 같이, THz 무선통신 시나리오는 매크로 네트워크(macro network), 마이크로 네트워크(micro network), 나노스케일 네트워크(nanoscale network)로 분류될 수 있다. 매크로 네트워크에서 THz 무선통신은 vehicle-to-vehicle 연결 및 backhaul/fronthaul 연결에 응용될 수 있다. 마이크로 네트워크에서 THz 무선통신은 인도어 스몰 셀(small cell), 데이터 센터에서 무선 연결과 같은 고정된 point-to-point 또는 multi-point 연결, 키오스크 다운로딩과 같은 근거리 통신(near-field communication)에 응용될 수 있다.As shown in FIG. 11 , a THz wireless communication scenario may be classified into a macro network, a micro network, and a nanoscale network. In the macro network, THz wireless communication can be applied to vehicle-to-vehicle connection and backhaul/fronthaul connection. THz wireless communication in micro networks is applied to indoor small cells, fixed point-to-point or multi-point connections such as wireless connections in data centers, and near-field communication such as kiosk downloading. can be
아래 표 2는 THz 파에서 이용될 수 있는 기술의 일례를 나타낸 표이다.Table 2 below is a table showing an example of a technique that can be used in the THz wave.
Figure PCTKR2020011878-appb-img-000002
Figure PCTKR2020011878-appb-img-000002
광 무선 기술 (Optical wireless technology)Optical wireless technology
OWC 기술은 가능한 모든 장치-대-액세스 네트워크를 위한 RF 기반 통신 외에도 6G 통신을 위해 계획되었다. 이러한 네트워크는 네트워크-대-백홀/프론트홀 네트워크 연결에 접속한다. OWC 기술은 4G 통신 시스템 이후 이미 사용되고 있으나 6G 통신 시스템의 요구를 충족시키기 위해 더 널리 사용될 것이다. 광 충실도(light fidelity), 가시광 통신, 광 카메라 통신 및 광 대역에 기초한 FSO 통신과 같은 OWC 기술은 이미 잘 알려진 기술이다. 광 무선 기술 기반의 통신은 매우 높은 데이터 속도, 낮은 지연 시간 및 안전한 통신을 제공할 수 있다. LiDAR 또한 광 대역을 기반으로 6G 통신에서 초 고해상도 3D 매핑을 위해 이용될 수 있다.OWC technology is envisioned for 6G communications in addition to RF-based communications for all possible device-to-access networks. These networks connect to network-to-backhaul/fronthaul network connections. OWC technology has already been used since the 4G communication system, but will be used more widely to meet the needs of the 6G communication system. OWC technologies such as light fidelity, visible light communication, optical camera communication, and FSO communication based on a light band are well known technologies. Communication based on optical radio technology can provide very high data rates, low latency and secure communication. LiDAR can also be used for ultra-high-resolution 3D mapping in 6G communication based on wide bands.
FSO 백홀 네트워크FSO backhaul network
FSO 시스템의 송신기 및 수신기 특성은 광섬유 네트워크의 특성과 유사하다. 따라서, FSO 시스템의 데이터 전송은 광섬유 시스템과 비슷하다. 따라서, FSO는 광섬유 네트워크와 함께 6G 시스템에서 백홀 연결을 제공하는 좋은 기술이 될 수 있다. FSO를 사용하면, 10,000km 이상의 거리에서도 매우 장거리 통신이 가능하다. FSO는 바다, 우주, 수중, 고립된 섬과 같은 원격 및 비원격 지역을 위한 대용량 백홀 연결을 지원한다. FSO는 셀룰러 BS 연결도 지원한다.The transmitter and receiver characteristics of an FSO system are similar to those of a fiber optic network. Thus, data transmission in an FSO system is similar to that of a fiber optic system. Therefore, FSO can be a good technology to provide backhaul connectivity in 6G systems along with fiber optic networks. Using FSO, very long-distance communication is possible even at distances of 10,000 km or more. FSO supports high-capacity backhaul connections for remote and non-remote areas such as sea, space, underwater, and isolated islands. FSO also supports cellular BS connectivity.
대규모 MIMO 기술Massive MIMO technology
스펙트럼 효율을 향상시키는 핵심 기술 중 하나는 MIMO 기술을 적용하는 것이다. MIMO 기술이 향상되면 스펙트럼 효율도 향상된다. 따라서, 6G 시스템에서 대규모 MIMO 기술이 중요할 것이다. MIMO 기술은 다중 경로를 이용하기 때문에 데이터 신호가 하나 이상의 경로로 전송될 수 있도록 다중화 기술 및 THz 대역에 적합한 빔 생성 및 운영 기술도 중요하게 고려되어야 한다.One of the key technologies to improve spectral efficiency is to apply MIMO technology. As MIMO technology improves, so does the spectral efficiency. Therefore, large-scale MIMO technology will be important in 6G systems. Since the MIMO technology uses multiple paths, a multiplexing technique and a beam generation and operation technique suitable for the THz band should also be considered important so that a data signal can be transmitted through one or more paths.
블록 체인blockchain
블록 체인은 미래의 통신 시스템에서 대량의 데이터를 관리하는 중요한 기술이 될 것이다. 블록 체인은 분산 원장 기술의 한 형태로서, 분산 원장은 수많은 노드 또는 컴퓨팅 장치에 분산되어 있는 데이터베이스이다. 각 노드는 동일한 원장 사본을 복제하고 저장한다. 블록 체인은 P2P 네트워크로 관리된다. 중앙 집중식 기관이나 서버에서 관리하지 않고 존재할 수 있다. 블록 체인의 데이터는 함께 수집되어 블록으로 구성된다. 블록은 서로 연결되고 암호화를 사용하여 보호된다. 블록 체인은 본질적으로 향상된 상호 운용성(interoperability), 보안, 개인 정보 보호, 안정성 및 확장성을 통해 대규모 IoT를 완벽하게 보완한다. 따라서, 블록 체인 기술은 장치 간 상호 운용성, 대용량 데이터 추적성, 다른 IoT 시스템의 자율적 상호 작용 및 6G 통신 시스템의 대규모 연결 안정성과 같은 여러 기능을 제공한다.Blockchain will become an important technology for managing large amounts of data in future communication systems. Blockchain is a form of distributed ledger technology, which is a database distributed across numerous nodes or computing devices. Each node replicates and stores an identical copy of the ledger. The blockchain is managed as a peer-to-peer network. It can exist without being managed by a centralized authority or server. Data on the blockchain is collected together and organized into blocks. Blocks are linked together and protected using encryption. Blockchain in nature perfectly complements IoT at scale with improved interoperability, security, privacy, reliability and scalability. Therefore, blockchain technology provides several features such as interoperability between devices, traceability of large amounts of data, autonomous interaction of different IoT systems, and large-scale connection stability of 6G communication systems.
3D 네트워킹3D Networking
6G 시스템은 지상 및 공중 네트워크를 통합하여 수직 확장의 사용자 통신을 지원한다. 3D BS는 저궤도 위성 및 UAV를 통해 제공될 것이다. 고도 및 관련 자유도 측면에서 새로운 차원을 추가하면 3D 연결이 기존 2D 네트워크와 상당히 다르다.The 6G system integrates terrestrial and public networks to support vertical expansion of user communications. 3D BS will be provided via low orbit satellites and UAVs. Adding a new dimension in terms of elevation and associated degrees of freedom makes 3D connections significantly different from traditional 2D networks.
양자 커뮤니케이션quantum communication
6G 네트워크의 맥락에서 네트워크의 감독되지 않은 강화 학습이 유망하다. 지도 학습 방식은 6G에서 생성된 방대한 양의 데이터에 레이블을 지정할 수 없다. 비지도 학습에는 라벨링이 필요하지 않다. 따라서, 이 기술은 복잡한 네트워크의 표현을 자율적으로 구축하는 데 사용할 수 있다. 강화 학습과 비지도 학습을 결합하면 진정한 자율적인 방식으로 네트워크를 운영할 수 있다.In the context of 6G networks, unsupervised reinforcement learning of networks is promising. Supervised learning methods cannot label the massive amounts of data generated by 6G. Unsupervised learning does not require labeling. Thus, this technique can be used to autonomously build representations of complex networks. Combining reinforcement learning and unsupervised learning allows networks to operate in a truly autonomous way.
무인 항공기drone
UAV(Unmanned Aerial Vehicle) 또는 드론은 6G 무선 통신에서 중요한 요소가 될 것이다. 대부분의 경우, UAV 기술을 사용하여 고속 데이터 무선 연결이 제공된다. BS 엔티티는 셀룰러 연결을 제공하기 위해 UAV에 설치된다. UAV는 쉬운 배치, 강력한 가시선 링크 및 이동성이 제어되는 자유도와 같은 고정 BS 인프라에서 볼 수 없는 특정 기능을 가지고 있다. 천재 지변 등의 긴급 상황 동안, 지상 통신 인프라의 배치는 경제적으로 실현 가능하지 않으며, 때로는 휘발성 환경에서 서비스를 제공할 수 없다. UAV는 이러한 상황을 쉽게 처리할 수 있다. UAV는 무선 통신 분야의 새로운 패러다임이 될 것이다. 이 기술은 eMBB, URLLC 및 mMTC 인 무선 네트워크의 세 가지 기본 요구 사항을 용이하게 한다. UAV는 또한, 네트워크 연결성 향상, 화재 감지, 재난 응급 서비스, 보안 및 감시, 오염 모니터링, 주차 모니터링, 사고 모니터링 등과 같은 여러 가지 목적을 지원할 수 있다. 따라서, UAV 기술은 6G 통신에 가장 중요한 기술 중 하나로 인식되고 있다.Unmanned Aerial Vehicles (UAVs) or drones will become an important element in 6G wireless communication. In most cases, high-speed data wireless connections are provided using UAV technology. A BS entity is installed in the UAV to provide cellular connectivity. UAVs have certain features not found in fixed BS infrastructure, such as easy deployment, strong line-of-sight links, and degrees of freedom with controlled mobility. During emergencies such as natural disasters, the deployment of terrestrial communications infrastructure is not economically feasible and sometimes cannot provide services in volatile environments. A UAV can easily handle this situation. UAV will become a new paradigm in the field of wireless communication. This technology facilitates the three basic requirements of wireless networks: eMBB, URLLC and mMTC. UAVs can also serve several purposes, such as improving network connectivity, fire detection, disaster emergency services, security and surveillance, pollution monitoring, parking monitoring, incident monitoring, and more. Therefore, UAV technology is recognized as one of the most important technologies for 6G communication.
셀-프리 통신(Cell-free Communication)Cell-free Communication
여러 주파수와 이기종 통신 기술의 긴밀한 통합은 6G 시스템에서 매우 중요하다. 결과적으로, 사용자는 디바이스에서 어떤 수동 구성을 만들 필요 없이 네트워크에서 다른 네트워크로 원활하게 이동할 수 있다. 사용 가능한 통신 기술에서 최상의 네트워크가 자동으로 선택된다. 이것은 무선 통신에서 셀 개념의 한계를 깨뜨릴 것이다. 현재, 하나의 셀에서 다른 셀로의 사용자 이동은 고밀도 네트워크에서 너무 많은 핸드 오버를 야기하고, 핸드 오버 실패, 핸드 오버 지연, 데이터 손실 및 핑퐁 효과를 야기한다. 6G 셀-프리 통신은 이 모든 것을 극복하고 더 나은 QoS를 제공할 것이다. 셀-프리 통신은 멀티 커넥티비티 및 멀티-티어 하이브리드 기술과 장치의 서로 다른 이기종 라디오를 통해 달성될 것이다.Tight integration of multiple frequencies and heterogeneous communication technologies is very important in 6G systems. As a result, users can seamlessly move from one network to another without having to make any manual configuration on the device. The best network is automatically selected from the available communication technologies. This will break the limitations of the cell concept in wireless communication. Currently, user movement from one cell to another causes too many handovers in high-density networks, causing handover failures, handover delays, data loss and ping-pong effects. 6G cell-free communication will overcome all of this and provide better QoS. Cell-free communication will be achieved through multi-connectivity and multi-tier hybrid technologies and different heterogeneous radios of devices.
무선 정보 및 에너지 전송 통합Integration of wireless information and energy transmission
WIET은 무선 통신 시스템과 같이 동일한 필드와 웨이브(wave)를 사용한다. 특히, 센서와 스마트폰은 통신 중 무선 전력 전송을 사용하여 충전될 것이다. WIET은 배터리 충전 무선 시스템의 수명을 연장하기 위한 유망한 기술이다. 따라서, 배터리가 없는 장치는 6G 통신에서 지원될 것이다.WIET uses the same fields and waves as wireless communication systems. In particular, the sensor and smartphone will be charged using wireless power transfer during communication. WIET is a promising technology for extending the life of battery-charging wireless systems. Therefore, devices without batteries will be supported in 6G communication.
센싱과 커뮤니케이션의 통합Integration of sensing and communication
자율 무선 네트워크는 동적으로 변화하는 환경 상태를 지속적으로 감지하고 서로 다른 노드간에 정보를 교환할 수 있는 기능이다. 6G에서, 감지는 자율 시스템을 지원하기 위해 통신과 긴밀하게 통합될 것이다.An autonomous wireless network is a function that can continuously detect dynamically changing environmental conditions and exchange information between different nodes. In 6G, sensing will be tightly integrated with communications to support autonomous systems.
액세스 백홀 네트워크의 통합Consolidation of access backhaul networks
6G에서 액세스 네트워크의 밀도는 엄청날 것이다. 각 액세스 네트워크는 광섬유와 FSO 네트워크와 같은 백홀 연결로 연결된다. 매우 많은 수의 액세스 네트워크들에 대처하기 위해, 액세스 및 백홀 네트워크 사이에 긴밀한 통합이 있을 것이다.The density of access networks in 6G will be enormous. Each access network is connected by backhaul connections such as fiber optic and FSO networks. To cope with a very large number of access networks, there will be tight integration between the access and backhaul networks.
홀로그램 빔 포밍Holographic Beamforming
빔 포밍은 특정 방향으로 무선 신호를 전송하기 위해 안테나 배열을 조정하는 신호 처리 절차이다. 스마트 안테나 또는 진보된 안테나 시스템의 하위 집합이다. 빔 포밍 기술은 높은 호 대잡음비, 간섭 방지 및 거부, 높은 네트워크 효율과 같은 몇 가지 장점이 있다. 홀로그램 빔 포밍 (HBF)은 소프트웨어-정의된 안테나를 사용하기 때문에 MIMO 시스템과 상당히 다른 새로운 빔 포밍 방법이다. HBF는 6G에서 다중 안테나 통신 장치에서 신호의 효율적이고 유연한 전송 및 수신을 위해 매우 효과적인 접근 방식이 될 것이다.Beamforming is a signal processing procedure that adjusts an antenna array to transmit a radio signal in a specific direction. A smart antenna or a subset of an advanced antenna system. Beamforming technology has several advantages such as high call-to-noise ratio, interference prevention and rejection, and high network efficiency. Hologram beamforming (HBF) is a new beamforming method that is significantly different from MIMO systems because it uses a software-defined antenna. HBF will be a very effective approach for efficient and flexible transmission and reception of signals in multi-antenna communication devices in 6G.
빅 데이터 분석Big Data Analytics
빅 데이터 분석은 다양한 대규모 데이터 세트 또는 빅 데이터를 분석하기 위한 복잡한 프로세스이다. 이 프로세스는 숨겨진 데이터, 알 수 없는 상관 관계 및 고객 성향과 같은 정보를 찾아 완벽한 데이터 관리를 보장한다. 빅 데이터는 비디오, 소셜 네트워크, 이미지 및 센서와 같은 다양한 소스에서 수집된다. 이 기술은 6G 시스템에서 방대한 데이터를 처리하는 데 널리 사용된다.Big data analytics is a complex process for analyzing various large data sets or big data. This process ensures complete data management by finding information such as hidden data, unknown correlations and customer propensity. Big data is gathered from a variety of sources such as videos, social networks, images and sensors. This technology is widely used to process massive amounts of data in 6G systems.
Large Intelligent Surface(LIS)Large Intelligent Surface (LIS)
THz 대역 신호의 경우 직진성이 강하여 방해물로 인한 음영 지역이 많이 생길 수 있는데, 이러한 음영 지역 근처에 LIS 설치함으로써 통신 권역을 확대하고 통신 안정성 강화 및 추가적인 부가 서비스가 가능한 LIS 기술이 중요하게 된다. LIS는 전자기 물질(electromagnetic materials)로 만들어진 인공 표면(artificial surface)이고, 들어오는 무선파와 나가는 무선파의 전파(propagation)을 변경시킬 수 있다. LIS는 massive MIMO의 확장으로 보여질 수 있으나, massive MIMO와 서로 다른 array 구조 및 동작 메커니즘이 다르다. 또한, LIS는 수동 엘리먼트(passive elements)를 가진 재구성 가능한 리플렉터(reflector)로서 동작하는 점 즉, 활성(active) RF chain을 사용하지 않고 신호를 수동적으로만 반사(reflect)하는 점에서 낮은 전력 소비를 가지는 장점이 있다. 또한, LIS의 수동적인 리플렉터 각각은 입사되는 신호의 위상 편이를 독립적으로 조절해야 하기 때문에, 무선 통신 채널에 유리할 수 있다. LIS 컨트롤러를 통해 위상 편이를 적절히 조절함으로써, 반사된 신호는 수신된 신호 전력을 부스트(boost)하기 위해 타겟 수신기에서 모여질 수 있다.In the case of the THz band signal, the linearity is strong, so there may be many shaded areas due to obstructions. By installing the LIS near these shaded areas, the LIS technology that expands the communication area, strengthens communication stability and enables additional additional services becomes important. The LIS is an artificial surface made of electromagnetic materials, and can change the propagation of incoming and outgoing radio waves. LIS can be seen as an extension of massive MIMO, but the array structure and operation mechanism are different from those of massive MIMO. In addition, LIS has low power consumption in that it operates as a reconfigurable reflector with passive elements, that is, only passively reflects the signal without using an active RF chain. There are advantages to having Also, since each of the passive reflectors of the LIS must independently adjust the phase shift of the incoming signal, it can be advantageous for a wireless communication channel. By properly adjusting the phase shift via the LIS controller, the reflected signal can be gathered at the target receiver to boost the received signal power.
앞서 살핀 6G 통신 기술은 후술할 본 명세서에서 제안되는 방법들과 결합되어 적용될 수 있으며, 또는 본 명세서에서 제안하는 방법들의 기술적 특징을 구체화하거나 명확하게 하는데 보충될 수 있다. 한편, 본 명세서에서 제안하는 통신 서비스는 앞서 설명한 6G 통신 기술뿐만 아니라, 3G, 4G 및/또는 5G 통신 기술에 의한 통신 서비스와 결합되어 적용될 수도 있다. The above salpin 6G communication technology may be applied in combination with the methods proposed in the present specification to be described later, or may be supplemented to specify or clarify the technical characteristics of the methods proposed in the present specification. On the other hand, the communication service proposed in the present specification may be applied in combination with a communication service by 3G, 4G and/or 5G communication technology as well as the 6G communication technology described above.
도 12는 실시 예가 적용되는 연합학습을 위한 통신 시스템을 나타내는 도면이다. 12 is a diagram illustrating a communication system for federated learning to which an embodiment is applied.
도 12를 참조하면, 연합학습을 위한 통신 시스템은 서버(MS) 및 디바이스(DE)들을 포함한다. 12, the communication system for federated learning includes a server (MS) and devices (DE).
베이스 스테이션(BS)의 서버(MS)는 글로벌 모델(Global Model)을 구비한다. 서버(MS)는 복수의 디바이스(DE)들과 다양한 방식으로 통신할 수 있다. 글로벌 모델은 인공지능 학습을 수행하기 위한 것으로, 서버(MS)는 글로벌 모델을 디바이스(DE)들로 제공한다.The server MS of the base station BS has a global model. The server MS may communicate with the plurality of devices DE in various ways. The global model is for performing AI learning, and the server (MS) provides the global model to the devices (DE).
디바이스(DE)들 각각은 로컬 데이터(Local Data)를 획득하고, 제공받은 글로벌 모델을 이용하여 로컬 데이터를 학습한다. 디바이스(DE)들은 학습을 통해서 가중치를 업데이트하고, 업데이트 된 가중치의 변화량을 계산하여 서버(MS)로 전송한다. 이때, 가중치 변화량은 업데이트되기 이전의 가중치와 업데이트 된 이후의 가중치들 간의 차이를 의미한다. 또는, 디바이스(DE)들은 가중치 변화량이 아닌 업데이트 된 가중치를 서버(MS)에 전송할 수도 있다.Each of the devices DE acquires local data and learns the local data using the provided global model. The devices DE update the weight through learning, calculate the change amount of the updated weight, and transmit it to the server MS. In this case, the weight change amount means a difference between the weights before the update and the weights after the update. Alternatively, the devices DE may transmit the updated weight, not the weight change amount, to the server MS.
서버(MS)는 디바이스(DE)들로부터 제공받은 가중치 변화량에 기초하여, 글로벌 모델의 가중치 업데이트를 진행한다. 그리고, 서버(MS)는 가중치 업데이트가 종료된 이후, 글로벌 모델의 손실 함수(Loss function)에 대한 평가를 수행할 수 있다. The server MS updates the weight of the global model based on the weight change amount provided from the devices DE. And, after the weight update is finished, the server MS may evaluate the loss function of the global model.
도 13은 실시 예에 따른 연합학습 프로토콜을 나타내는 도면이다. 13 is a diagram illustrating a federated learning protocol according to an embodiment.
도 13을 참조하면, 연합학습 다수의 라운드로 이루어져 있으며, 각각의 라운드는 선택(Selection) 구간, 구성화(Configuration) 구간, 리포팅(Reporting) 구간을 포함한다.Referring to FIG. 13, federated learning consists of a plurality of rounds, and each round includes a selection section, a configuration section, and a reporting section.
선택 절차는 디바이스(DE)들이 서버(MS)에 연합 학습에 참여하기 위해 등록을 수행하고, 서버(MS)는 다수의 디바이스들 중에서 해당 라운드에 참여할 디바이스를 선택하는 과정이다.The selection procedure is a process in which the devices DE register with the server MS to participate in federated learning, and the server MS selects a device to participate in the corresponding round from among a plurality of devices.
구성화 구간은 서버(MS)가 선택된 디바이스들에게 글로벌 모델 및 필요한 파라미터들을 전송하고, 디바이스들은 로컬 데이터를 이용하여 수신된 글로벌 모델을 학습하는 과정이다.The configuration section is a process in which the server (MS) transmits the global model and necessary parameters to selected devices, and the devices learn the received global model using local data.
리포팅 구간은 각각의 디바이스들이 학습된 모델로부터 가중치 변화량을 계산하고, 계산된 가중치 변화량을 서버(MS)로 전송하는 구간이다. 서버(MS)는 일정 시간 내에 수신된 가중치 변화량들에 기초하여, 글로벌 모델의 가중치들을 업데이트한다.The reporting section is a section in which each device calculates the weight change amount from the learned model, and transmits the calculated weight change amount to the server (MS). The server MS updates the weights of the global model based on the weight changes received within a predetermined time.
라운드 내의 각 절차들은 서버(MS)와 디바이스(DE)들 간의 동기화(Synchronous)가 이루어진 상태에서 수행된다. 서버(MS)는 주어진 시간 내에 선택된 디바이스들로부터 수신된 가중치 변화량들을 이용하여 글로벌 모델을 업데이트 한다. 서버(MS)는 업데이트 된 글로벌 모델을 이용하여 다음 라운드의 학습을 진행한다. Each procedure in the round is performed in a state in which synchronization between the server (MS) and the devices (DE) is made. The server MS updates the global model using the weight changes received from the selected devices within a given time. The server (MS) proceeds with the next round of learning using the updated global model.
도 14는 실시 예에 의한 연합학습을 위한 통신 방법을 나타내는 도면이다.14 is a diagram illustrating a communication method for joint learning according to an embodiment.
도 14를 참조하면, 연합학습을 위한 통신 방법은 제1 단계(S1401)에서, 서버(MS)는 디바이스(DE)로 글로벌 모델 및 구상화 파라미터(configuration parameter)를 전송한다. 구상화 파라미터는 양자화의 스케일링 계수(scaling factor)를 포함한다. 또한 구상화 파라미터는 가중치 변화량 분포에 대한 최대값, 최소값, 분산 등의 통계 정보를 포함할 수도 있다. Referring to FIG. 14 , in the communication method for federated learning, in a first step ( S1401 ), the server (MS) transmits a global model and a configuration parameter to the device (DE). The globularization parameter includes a scaling factor of quantization. In addition, the visualization parameter may include statistical information such as a maximum value, a minimum value, and a variance of the weight change amount distribution.
스케일링 계수는 이전 라운드에서 수신된 각 디바이스들의 가중치 변화량 분포에 기초하여 결정된다. 각각의 디바이스들은 동일한 스케일링 계수를 제공받는다. The scaling factor is determined based on the distribution of the weight change amount of each device received in the previous round. Each device is provided with the same scaling factor.
제2 단계(S1402)에서, 디바이스(DE)는 글로벌 모델을 이용하여 학습한 이후, 서버(MS)로부터 제공받은 스케일링 계수를 기반으로 양자화를 수행한다. 디바이스(DE)는 양자화를 수행하는 과정에서, 각 레이어(Layer)의 가중치 변화량 분포에 따라 가변 비트(Variable Bit) 기반의 양자화를 추가로 수행할 수 있다. 그리고, 디바이스(DE)는 가중치 변화량 정보에 양자화 비트(Quantization bit) 정보를 포함해서 서버(MS)에 전달할 수 있다.In the second step (S1402), the device DE learns using the global model, and then performs quantization based on the scaling factor provided from the server MS. In the process of performing quantization, the device DE may additionally perform variable bit-based quantization according to a distribution of a weight change amount of each layer. In addition, the device DE may include quantization bit information in the weight change amount information and transmit it to the server MS.
도 15는 본 발명의 실시 예에 따른 스케일링 계수를 결정하는 방법을 나타내는 도면이다. 15 is a diagram illustrating a method of determining a scaling factor according to an embodiment of the present invention.
도 15를 참조하면, 제1 단계(S1501)에서, 서버(MS)는 이전 라운드에서 수신된 가중치 변화량들과 손실 함수 평가(Loss Function Evaluation)를 분석한다. Referring to FIG. 15 , in the first step ( S1501 ), the server MS analyzes the weight changes received in the previous round and the loss function evaluation.
이를 위해서, 서버(MS)는 각 디바이스(DE)들로부터 제공받은 가중치 변화량들에 기초하여, 가중치 변화량들의 절대값을 누적한 누적분포함수(Cumulative distribution function; CDF)를 생성한다. 도 16은 서버(MS)가 생성한 가중치 변화량 누적분포함수의 일례를 나타내는 도면이다. To this end, the server MS generates a cumulative distribution function (CDF) accumulating absolute values of the weight changes based on the weight change amounts provided from the respective devices DE. 16 is a diagram showing an example of the weight change amount cumulative distribution function generated by the server (MS).
그리고, 서버(MS)는 가중치 변화량 누적분포함수에 기초하여, 임계값(gthreshold)에 대응하는 경계값(gmax)을 획득한다. 경계값(gmax) 이상의 크기를 갖는 누적 가중치 변화량은 학습에 이용되지 않는다. 이는 다른 가중치 변화량에 대비하여 빈도수가 매우 작으면서 크기가 큰 값을 갖는 가중치 변화량들은 학습을 비효율적으로 할 가능성이 있기 때문이다. 임계값(gthreshold)은 미리 설정될 수 있고, 라운드마다 임계값(gthreshold)의 크기는 달라질 수 있다. Then, the server MS acquires a threshold value gmax corresponding to a threshold value gthreshold based on the weight change cumulative distribution function. A cumulative weight change having a size greater than or equal to the threshold gmax is not used for learning. This is because weight changes having a very small frequency and a large size may make learning inefficient compared to other weight changes. The threshold value (gthreshold) may be preset, and the size of the threshold value (gthreshold) may vary for each round.
서버(MS)는 이전 라운드에서 업데이트 된 글로벌 모델에 기초하여 손실 함수의 오차 평가를 수행한다. The server MS performs error evaluation of the loss function based on the global model updated in the previous round.
도 17은 평균 제곱 오차(Mean Square Error)를 이용하여 구한 손실 함수의 오차 평가의 일례를 나타내는 도면이다.17 is a diagram illustrating an example of error evaluation of a loss function obtained using a mean square error.
도 17을 참조하면, 손실 함수의 오차는 학습이 진행될수록 특정값으로 수렴하는 경향을 보인다. 이와 같이 손실 함수의 오차가 수렴될수록 그래디언트(gradient) 변화도 점점 감소하여 "0"을 중심으로 몰리는 분포를 보일 수 있다. 그러나 학습 초기에는 손실 함수의 오차가 수렴되지 않은 상태이기 때문에, 가중치 변화량들이 비록 좁은 범위에 분포하고 있다고 할지라도, 후속 라운드에서는 넓게 분포될 가능성을 내포한다. Referring to FIG. 17 , the error of the loss function tends to converge to a specific value as learning progresses. As such, as the error of the loss function converges, the gradient change also gradually decreases, so that a distribution centered on “0” may be exhibited. However, since the error of the loss function is not converged at the beginning of learning, even if the weight changes are distributed in a narrow range, it has the possibility of being widely distributed in the subsequent rounds.
제2 단계(S1502)에서, 서버(MS)는 가중치 변화량의 누적 분포 함수와 손실 함수 오차에 기초하여, 스케일링 계수를 결정한다.In the second step S1502, the server MS determines a scaling factor based on the loss function error and the cumulative distribution function of the weight change amount.
스케일링 계수를 결정하는 과정에서, 스케일링 계수를 고려하고 남은 가중치 변화량 값에 대한 최대 양자화 비트를 8bit으로 결정한 것으로 가정하여 설명하기로 한다.In the process of determining the scaling coefficient, it is assumed that the maximum quantization bit for the remaining weight change value after considering the scaling coefficient is determined to be 8 bits.
이때, 가중치 변화량은 스케일링 계수를 제외하고 -127 ~ 127 범위의 값을 가지게 될 것이다. 가중치 변화량의 누적 분포 함수에서 가중치 변화량의 최대 값을 경계값(gmax)으로 결정되었을 경우, 다음 라운드의 분포는 "-gmax" 부터 "gmax" 내의 범위가 된다. 따라서, 서버(MS)는 다음 라운드의 스케일링 계수를 " gmax/127"값으로 결정한다.In this case, the weight change amount will have a value in the range of -127 to 127 excluding the scaling factor. When the maximum value of the weight change in the cumulative distribution function of the weight change is determined as the boundary value gmax, the distribution of the next round is in the range from "-gmax" to "gmax". Accordingly, the server MS determines the scaling factor of the next round as a value of “gmax/127”.
그리고, 서버(MS)는 손실 함수의 평가 결과에 기초하여, 스케일링 계수에 대한 적용 여부를 결정한다. 예를 들어, 도 17에서, 약 20 라운드 이전에는 아직 학습이 많이 진행되지 않은 상태이므로 라운드별로 가중치 변화량의 분포가 큰 차이를 나타낼 수 있다. 따라서 다음과 같이, 손실 함수의 오차에 따라 스케일링 계수를 조절할 수 있다. Then, the server MS determines whether to apply the scaling factor based on the evaluation result of the loss function. For example, in FIG. 17 , since a lot of learning has not been performed before about 20 rounds, the distribution of the weight change amount for each round may represent a large difference. Therefore, the scaling factor can be adjusted according to the error of the loss function as follows.
먼저, [수학식 1]을 이용하여 손실 함수의 변화량을 획득한다.First, the amount of change in the loss function is obtained using [Equation 1].
[수학식 1][Equation 1]
Figure PCTKR2020011878-appb-img-000003
Figure PCTKR2020011878-appb-img-000003
즉, 손실 함수의 변화량은 이전 라운드의 손실과 현재 라운드의 손실의 차이로 산출될 수 있다. That is, the amount of change in the loss function may be calculated as the difference between the loss of the previous round and the loss of the current round.
그리고, 손실 함수의 변화량(
Figure PCTKR2020011878-appb-img-000004
Loss)이 임계치(E threshold) 보다 크다면, 다음 라운드의 스케일링 계수는 다음의 [수학식 2]와 같이 산출될 수 있다.And, the amount of change of the loss function (
Figure PCTKR2020011878-appb-img-000004
Loss) is greater than the threshold E threshold , the scaling factor of the next round may be calculated as in Equation 2 below.
[수학식 2][Equation 2]
Figure PCTKR2020011878-appb-img-000005
Figure PCTKR2020011878-appb-img-000005
또는, 손실 함수의 변화량(
Figure PCTKR2020011878-appb-img-000006
Loss)이 "0<
Figure PCTKR2020011878-appb-img-000007
Loss<E threshold "인 조건을 만족하면, 다음 라운드의 스케일링 계수는 [수학식 3]과 같이 결정될 수 있다.Alternatively, the amount of change in the loss function (
Figure PCTKR2020011878-appb-img-000006
Loss) is "0<
Figure PCTKR2020011878-appb-img-000007
When the condition of "Loss<E threshold " is satisfied, the scaling factor of the next round may be determined as in [Equation 3].
[수학식 3][Equation 3]
Figure PCTKR2020011878-appb-img-000008
Figure PCTKR2020011878-appb-img-000008
그 이외의 경우, 다음 라운드의 스케일링 계수는 [수학식 4]와 같이 결정될 수 있다. In other cases, the scaling factor of the next round may be determined as in [Equation 4].
[수학식 4][Equation 4]
Figure PCTKR2020011878-appb-img-000009
Figure PCTKR2020011878-appb-img-000009
[수학식 1] 내지 [수학식 4]를 통해서 획득되는 손실은 도 17과 같이 해당 라운드에서 업데이트 된 글로벌 모델의 손실 함수에 대한 평가 결과를 의미한다.The loss obtained through [Equation 1] to [Equation 4] means the evaluation result of the loss function of the global model updated in the corresponding round as shown in FIG. 17 .
[수학식 1] 내지 [수학식 4]을 통해서 스케일링 계수를 산출할 때, 스케일링 계수는 하나 이상이 결정될 수 있다. 예를 들어, 스케일링 계수는 "-gmax" 부터 "gmax" 내의 범위 내에서 하나만 결정될 수 있다. When calculating the scaling factor through [Equation 1] to [Equation 4], one or more scaling factors may be determined. For example, only one scaling factor may be determined within a range from “-gmax” to “gmax”.
또는 경계값(gmax)의 절대값 이내의 범위는 둘 이상으로 구분되고, 각각의 경계값에서 스케일링 계수가 결정될 수 있다. 예를 들어, "-gmax" 부터 "-gmax/2" 범위에서 제1 스케일링 계수가 결정되고, "-gmax/2" 부터 "gmax" 범위에서 제2 스케일링 계수가 결정될 수 있다. 또한, 제1 스케일링 계수는 제1 양자화 정보와 매칭될 수 있고, 제2 스케일링 계수는 제2 양자화 정보와 매칭될 수 있다. 제1 양자화 정보 및 제2 양자화 정보는 서로 다를 수 있다.Alternatively, a range within the absolute value of the boundary value gmax may be divided into two or more, and a scaling factor may be determined from each boundary value. For example, a first scaling factor may be determined in a range of “-gmax” to “-gmax/2”, and a second scaling factor may be determined in a range of “-gmax/2” to “gmax”. Also, the first scaling factor may match the first quantization information, and the second scaling factor may match the second quantization information. The first quantization information and the second quantization information may be different from each other.
[수학식 1] 내지 [수학식 4]을 통해서 알 수 있는 바와 같이, 손실 함수의 평가 결과가 급격하게 떨어지는 경우에는 학습이 아직 안정되지 않았다고 판단함으로써, 이전 라운드에 대비하여 가중치 변화량의 범위가 너무 급작스럽게 줄어드는 것을 방지할 수 있다. As can be seen from [Equation 1] to [Equation 4], when the evaluation result of the loss function falls sharply, it is determined that the learning is not stable yet, so the range of the weight change is too large compared to the previous round. A sudden decrease can be prevented.
또한, 손실 함수의 평가 결과의 변화가 작아진 상태에서는 보다 빠르게 가중치 변화량의 범위를 줄일 수 있으며 이전 라운드에 대비하여 성능이 저하되는 경우에는 가중치 변화량의 범위를 넓히기 위하여 스케일링 계수를 크게 할 수 있다.In addition, when the change in the evaluation result of the loss function is small, the range of the weight change can be reduced more quickly, and when the performance is deteriorated compared to the previous round, the scaling factor can be increased to wide the range of the weight change.
제3 단계(S1503)에서, 서버(MS)는 결정된 스케일링 계수를 디바이스들에 전송한다. In a third step (S1503), the server (MS) transmits the determined scaling factor to the devices.
디바이스(DE)들은 서버(MS)로부터의 스케일링 계수를 이용하여, 가변 비트 기반의 양자화 동작을 수행한다. 이를 도 18을 참조하여 살펴보면 다음과 같다.The devices DE perform a variable bit-based quantization operation using a scaling factor from the server MS. Referring to FIG. 18, it is as follows.
도 18은 디바이스들의 동작을 설명하는 도면이다.18 is a diagram for explaining the operation of devices.
도 18을 참조하면, 제1 단계(S1801)에서, 디바이스(DE)들은 서버(MS)로부터 연합학습을 위한 구상화 정보들 제공받는다. 구성화 정보들은 글로벌 모델 및 스케일링 계수 등을 포함한다. 디바이스(DE)들은 로컬 데이터를 획득하고, 로컬 데이터에 기초하여 글로벌 모델을 학습한다. Referring to FIG. 18 , in a first step ( S1801 ), the devices DE are provided with visualization information for joint learning from the server MS. The configuration information includes a global model and a scaling factor. The devices DE acquire local data and learn a global model based on the local data.
제2 단계(S1802)에서, 디바이스(DE)들은 학습이 완료된 이후, 업데이트 된 글로벌 모델로부터 가중치 변화량을 계산한다.In the second step ( S1802 ), the devices DE calculate a weight change amount from the updated global model after learning is completed.
제3 단계(S1803)에서, 디바이스(DE)들은 서버(MS)로부터 제공받은 스케일링 계수를 이용하여, 제2 단계(S1802)에서 산출된 가중치 변화량을 양자화한다. 이때, 양자화는 8bit 양자화 과정을 가정한다. In the third step (S1803), the devices (DE) quantize the weight change amount calculated in the second step (S1802) by using the scaling factor provided from the server (MS). In this case, the quantization assumes an 8-bit quantization process.
만약, 스케일링 계수가 "1/64"일 경우, 가중치 변화량(
Figure PCTKR2020011878-appb-img-000010
Wt)은 다음의 [수학식 5]와 같이 표현될 수 있다.If the scaling factor is "1/64", the weight change amount (
Figure PCTKR2020011878-appb-img-000010
Wt) can be expressed as the following [Equation 5].
[수학식 5][Equation 5]
Figure PCTKR2020011878-appb-img-000011
Figure PCTKR2020011878-appb-img-000011
결과적으로, 스케일링 계수 텀을 제외한 가중치 변화량은 "-6 ~ +6" 범위가 된다. "-6 ~ +6" 범위의 가중치 변화량은 4bit 양자화를 수행하는 것이 가능하다. 따라서, 디바이스(DE)는 4bit 양자화에 대한 정보인 Quantization bit(2) 와 양자화 된 가중치 변화량을 함께 전송한다. "Quantization bit(2)"는 양자화를 몇 bit로 수행하였는지를 나타내는 정보이다.As a result, the weight change amount excluding the scaling factor term is in the range of "-6 to +6". It is possible to perform 4-bit quantization for a weight change amount in the range of "-6 to +6". Accordingly, the device DE transmits the quantization bit (2), which is information about 4-bit quantization, and the quantized weight change amount together. "Quantization bit(2)" is information indicating how many bits quantization was performed.
만약, 서버(MS)로부터 스케일링 계수를 제공받지 않을 경우를 가정해 보면, 디바이스(DE)들은 가중치 변화량 범위 내에 있는 값들을 8 bit 양자화하여 다음의 [수학식 6]과 같이 전송하게 된다.If it is assumed that the scaling factor is not provided from the server MS, the devices DE quantize values within the range of the weight change amount by 8-bit and transmit them as shown in [Equation 6].
[수학식 6][Equation 6]
Figure PCTKR2020011878-appb-img-000012
Figure PCTKR2020011878-appb-img-000012
즉, 스케일링 계수 32bit(0.2/255)와 8bit로 양자화된 가중치 변화량이 전송되는 것을 알 수 있다. 위 두 방법에 대하여 전송되는 가중치 변화량을 비교해 보면 기존 방식의 경우 64bit이 필요하며 제안된 방식의 경우 18bit이 필요하다. 따라서 전반적으로 8bit 양자화를 수행하여 전송해야 하는 기존 방식에 비하여 제안된 방식의 경우 디바이스들 별로 가중치 변화량의 차이를 이용하여 전송되는 가중치 변화량의 크기는 줄일 수 있다. That is, it can be seen that the weight change amount quantized with a scaling factor of 32 bits (0.2/255) and 8 bits is transmitted. Comparing the amount of weight change transmitted for the above two methods, 64 bits are required for the existing method and 18 bits are required for the proposed method. Therefore, compared to the conventional method in which 8-bit quantization is performed and transmitted, the size of the transmitted weight change amount can be reduced by using the difference in the weight change amount for each device in the proposed method.
제4 단계(S1804)에서, 디바이스(DE)들은 양자화가 완료된 가중치 변화량 정보를 서버로 전송한다. 그리고, 가변 비트 양자화가 수행되었을 경우, 디바이스(DE)들은 가변 양자화 비트 정보 또한 서버(MS)로 전송한다. In a fourth step ( S1804 ), the devices DE transmit the weight change amount information on which quantization is completed to the server. And, when variable bit quantization is performed, the devices DE also transmit variable quantization bit information to the server MS.
상술한 본 발명의 실시 예는 가중치 변화량을 전송하는 과정에서 데이터 양을 줄일 수 있다.The above-described embodiment of the present invention can reduce the amount of data in the process of transmitting the weight change amount.
일반적인 가중치 변화량을 전송하는 방법은 불필요한 리소스의 낭비를 초래할 수 있다. 이는 디바이스들의 가중치 변화량 분포 특성이 상이할 수 있기 때문이다. 이를 살펴보면 다음과 같다.A method of transmitting a general weight change amount may result in unnecessary waste of resources. This is because the weight variation distribution characteristics of the devices may be different. Looking at this:
데이터의 특성에 따라 학습 모델의 레이어(Layer) 별로 영향이 다르기 때문에, 레이어에 따라서 가중치 변화량들이 넓게 분포하거나 가중치 변화량들이 좁은 범위 내에서 분포할 수 있다. Since the effects are different for each layer of the learning model according to the characteristics of the data, weight changes may be widely distributed or weight changes may be distributed within a narrow range depending on the layer.
또한, 동일한 레이어에 대해서도 디바이스 별로 가중치 변화량 분포 특성이 달라질 수 있다. 예를 들어, 이미 글로벌 모델에서 학습된 데이터와 비슷한 특성으로 인해 글로벌 모델 학습에 많은 영향을 주지 않는 디바이스인 경우 가중치 변화량의 분포는 매우 작은 범위에서 분포되어 있을 수 있다. 반대로, 글로벌 모델에서 학습된 데이터와 특성이 많이 다른 데이터를 보유하고 있는 디바이스의 경우, 글로벌 모델에 대한 학습 이후에 넓은 범위의 가중치 변화량 분포를 나타낼 수 있다. Also, even for the same layer, the weight change amount distribution characteristic for each device may be different. For example, in the case of a device that does not significantly affect global model learning due to similar characteristics to data already learned from a global model, the distribution of weight change may be distributed in a very small range. Conversely, in the case of a device having data that has a lot different characteristics from data learned from the global model, a wide range of weight change distribution can be shown after learning on the global model.
연합 학습은 학습 초기에 상이한 특성을 가진 데이터들이 충분히 반영되지 않기 때문에 디바이스들로부터 수집된 가중치 변화량들의 분산이 큰 경우가 많이 존재하고, 라운드가 반복될수록 많은 데이터들의 특성을 학습함에 따라 점차 손실 함수의 오차가 줄어들면서 가중치 변화량들의 분산 및 평균값들이 수렴한다. 따라서 초기에 대비하여 라운드가 진행될수록 가중치 변화량들의 분포는 점차 줄어든다.In federated learning, since data with different characteristics are not sufficiently reflected at the beginning of learning, there are many cases in which the variance of weight changes collected from devices is large, and as the rounds are repeated, as the characteristics of more data are learned, the loss function gradually decreases. As the error decreases, the variance and average values of the weight changes converge. Therefore, as the round progresses in preparation for the initial stage, the distribution of weight changes gradually decreases.
이와 같이, 각각의 디바이스들은 동일한 레이어를 기준으로 가중치 변화량들의 분포 특성이 매우 상이할 수 있다.As such, the respective devices may have very different distribution characteristics of weight variations based on the same layer.
가중치 변화량이 상이할 경우에, 가중치 변화량 전송의 문제점은 다음과 같다.When the weight change amount is different, the problem of weight change amount transmission is as follows.
제1 디바이스의 가중치 변화량이 "-1 부터 +1" 범위에 속하고, 제2 디바이스의 가중치 변화량이 "-0.05 부터 +0.1" 범위에 속하며, 제3 디바이스의 가중치 변화량이 "-0.8 부터 +0.6" 범위에 속할 수 있다. 이런 경우, 제1 내지 제3 디바이스들이 동일하게 8비트 양자화가 수행되면, 제1 내지 제3 디바이스들 모두 8비트 기반으로 구성된 가중치 변화량의 매트릭스를 이용하여야 한다. 그리고, 서버는 가중치를 업데이트하기 위해서, 동일한 위치의 가중치 변화량들을 합산하고 평균을 산출한다. 이 경우, 제2 디바이스와 같이 가중치 변화량이 매우 작은 값들의 경우에는 서버가 가중치를 업데이트하는 과정에 끼치는 영향이 미비하다. 이처럼 서버가 가중치를 업데이트 하는 과정에서 실질적으로 영향이 거의 없는 디바이스조차 동일한 양자화 비트를 이용하기 때문에 통신대역폭을 낭비하는 경향이 있다. The weight change amount of the first device is in the range of "-1 to +1", the weight change amount of the second device is in the range "-0.05 to +0.1", and the weight change amount of the third device is "-0.8 to +0.6" “It can be in scope. In this case, if 8-bit quantization is equally performed by the first to third devices, all of the first to third devices must use a matrix of weight change amounts configured on an 8-bit basis. Then, in order to update the weight, the server sums up the weight changes at the same location and calculates an average. In this case, in the case of values with a very small weight change amount, such as in the second device, the influence of the server on the weight updating process is negligible. In this way, in the process of updating the weights by the server, even a device that has little practical effect uses the same quantization bit, so communication bandwidth tends to be wasted.
이에 반해서, 본 발명의 실시 예는 가중치 변화량에 기초하여 생성된 스케일링 계수를 바탕으로 양자화를 수행하기 때문에, 가중치 변화량의 크기가 작은 디바이스에 제공하는 가중치 변화량 데이터 양을 감소시킬 수 있다. On the other hand, in the embodiment of the present invention, since quantization is performed based on a scaling coefficient generated based on the weight change amount, the amount of weight change amount data provided to a device having a small weight change amount can be reduced.
무선 통신 시스템에 사용되는 장치Devices used in wireless communication systems
이로 제한되는 것은 아니지만, 상술한 본 발명의 다양한 제안들은 기기들간에 무선 통신/연결(예, 6G)을 필요로 하는 다양한 분야에 적용될 수 있다.Although not limited thereto, the various proposals of the present invention described above may be applied to various fields requiring wireless communication/connection (eg, 6G) between devices.
이하, 도면을 참조하여 보다 구체적으로 예시한다. 이하의 도면/설명에서 동일한 도면 부호는 다르게 기술하지 않는 한, 동일하거나 대응되는 하드웨어 블록, 소프트웨어 블록 또는 기능 블록을 예시할 수 있다. Hereinafter, it will be exemplified in more detail with reference to the drawings. In the following drawings/descriptions, the same reference numerals may represent the same or corresponding hardware blocks, software blocks, or functional blocks, unless otherwise indicated.
도 19은 본 발명에 적용되는 통신 시스템을 예시한다.19 illustrates a communication system applied to the present invention.
도 19을 참조하면, 본 발명에 적용되는 통신 시스템(1)은 무선 기기, 기지국 및 네트워크를 포함한다. 여기서, 무선 기기는 무선 접속 기술(예, 5G NR(New RAT), LTE(Long Term Evolution))을 이용하여 통신을 수행하는 기기를 의미하며, 통신/무선/5G 기기로 지칭될 수 있다. 이로 제한되는 것은 아니지만, 무선 기기는 로봇(100a), 차량(100b-1, 100b-2), XR(eXtended Reality) 기기(100c), 휴대 기기(Hand-held device)(100d), 가전(100e), IoT(Internet of Thing) 기기(100f), AI기기/서버(400)를 포함할 수 있다. 예를 들어, 차량은 무선 통신 기능이 구비된 차량, 자율 주행 차량, 차량간 통신을 수행할 수 있는 차량 등을 포함할 수 있다. 여기서, 차량은 UAV(Unmanned Aerial Vehicle)(예, 드론)를 포함할 수 있다. XR 기기는 AR(Augmented Reality)/VR(Virtual Reality)/MR(Mixed Reality) 기기를 포함하며, HMD(Head-Mounted Device), 차량에 구비된 HUD(Head-Up Display), 텔레비전, 스마트폰, 컴퓨터, 웨어러블 디바이스, 가전 기기, 디지털 사이니지(signage), 차량, 로봇 등의 형태로 구현될 수 있다. 휴대 기기는 스마트폰, 스마트패드, 웨어러블 기기(예, 스마트워치, 스마트글래스), 컴퓨터(예, 노트북 등) 등을 포함할 수 있다. 가전은 TV, 냉장고, 세탁기 등을 포함할 수 있다. IoT 기기는 센서, 스마트미터 등을 포함할 수 있다. 예를 들어, 기지국, 네트워크는 무선 기기로도 구현될 수 있으며, 특정 무선 기기(200a)는 다른 무선 기기에게 기지국/네트워크 노드로 동작할 수도 있다.Referring to FIG. 19 , the communication system 1 applied to the present invention includes a wireless device, a base station, and a network. Here, the wireless device refers to a device that performs communication using a radio access technology (eg, 5G NR (New RAT), LTE (Long Term Evolution)), and may be referred to as a communication/wireless/5G device. Although not limited thereto, the wireless device may include a robot 100a, a vehicle 100b-1, 100b-2, an eXtended Reality (XR) device 100c, a hand-held device 100d, and a home appliance 100e. ), an Internet of Thing (IoT) device 100f, and an AI device/server 400 . For example, the vehicle may include a vehicle equipped with a wireless communication function, an autonomous driving vehicle, a vehicle capable of performing inter-vehicle communication, and the like. Here, the vehicle may include an Unmanned Aerial Vehicle (UAV) (eg, a drone). XR devices include AR (Augmented Reality)/VR (Virtual Reality)/MR (Mixed Reality) devices, and include a Head-Mounted Device (HMD), a Head-Up Display (HUD) provided in a vehicle, a television, a smartphone, It may be implemented in the form of a computer, a wearable device, a home appliance, a digital signage, a vehicle, a robot, and the like. The portable device may include a smart phone, a smart pad, a wearable device (eg, a smart watch, smart glasses), a computer (eg, a laptop computer), and the like. Home appliances may include a TV, a refrigerator, a washing machine, and the like. The IoT device may include a sensor, a smart meter, and the like. For example, the base station and the network may be implemented as a wireless device, and the specific wireless device 200a may operate as a base station/network node to other wireless devices.
무선 기기(100a~100f)는 기지국(200)을 통해 네트워크(300)와 연결될 수 있다. 무선 기기(100a~100f)에는 AI(Artificial Intelligence) 기술이 적용될 수 있으며, 무선 기기(100a~100f)는 네트워크(300)를 통해 AI 서버(400)와 연결될 수 있다. 네트워크(300)는 3G 네트워크, 4G(예, LTE) 네트워크 또는 5G(예, NR) 네트워크 등을 이용하여 구성될 수 있다. 무선 기기(100a~100f)는 기지국(200)/네트워크(300)를 통해 서로 통신할 수도 있지만, 기지국/네트워크를 통하지 않고 직접 통신(e.g. 사이드링크 통신(sidelink communication))할 수도 있다. 예를 들어, 차량들(100b-1, 100b-2)은 직접 통신(e.g. V2V(Vehicle to Vehicle)/V2X(Vehicle to everything) communication)을 할 수 있다. 또한, IoT 기기(예, 센서)는 다른 IoT 기기(예, 센서) 또는 다른 무선 기기(100a~100f)와 직접 통신을 할 수 있다.The wireless devices 100a to 100f may be connected to the network 300 through the base station 200 . AI (Artificial Intelligence) technology may be applied to the wireless devices 100a to 100f , and the wireless devices 100a to 100f may be connected to the AI server 400 through the network 300 . The network 300 may be configured using a 3G network, a 4G (eg, LTE) network, or a 5G (eg, NR) network. The wireless devices 100a to 100f may communicate with each other through the base station 200/network 300, but may also communicate directly (e.g. sidelink communication) without passing through the base station/network. For example, the vehicles 100b-1 and 100b-2 may perform direct communication (e.g. Vehicle to Vehicle (V2V)/Vehicle to everything (V2X) communication). Also, the IoT device (eg, sensor) may communicate directly with other IoT devices (eg, sensor) or other wireless devices 100a to 100f.
무선 기기(100a~100f)/기지국(200)-기지국(200)/무선 기기(100a~100f) 간에는 무선 통신/연결(150a, 150b)이 이뤄질 수 있다. 여기서, 무선 통신/연결은 상향/하향링크 통신(150a)과 사이드링크 통신(150b)(또는, D2D 통신)은 다양한 무선 접속 기술(예, 5G NR)을 통해 이뤄질 수 있다. 무선 통신/연결(150a, 150b)을 통해 무선 기기와 기지국/무선 기기는 서로 무선 신호를 송신/수신할 수 있다. 예를 들어, 무선 통신/연결(150a, 150b)은 도 A1의 전체/일부 과정에 기반하여 다양한 물리 채널을 통해 신호를 송신/수신할 수 있다. 이를 위해, 본 발명의 다양한 제안들에 기반하여, 무선 신호의 송신/수신을 위한 다양한 구성정보 설정 과정, 다양한 신호 처리 과정(예, 채널 인코딩/디코딩, 변조/복조, 자원 매핑/디매핑 등), 자원 할당 과정 등 중 적어도 일부가 수행될 수 있다.Wireless communication/ connection 150a and 150b may be performed between the wireless devices 100a to 100f/base station 200 - the base station 200/wireless devices 100a to 100f. Here, the wireless communication/connection may be performed through various wireless access technologies (eg, 5G NR) for uplink/downlink communication 150a and sidelink communication 150b (or D2D communication). Through the wireless communication/ connection 150a and 150b, the wireless device and the base station/wireless device may transmit/receive wireless signals to each other. For example, the wireless communication/ connection 150a and 150b may transmit/receive signals through various physical channels based on all/part of the process of FIG. A1 . To this end, based on various proposals of the present invention, various configuration information setting processes for wireless signal transmission/reception, various signal processing processes (eg, channel encoding/decoding, modulation/demodulation, resource mapping/demapping, etc.) , at least a part of a resource allocation process may be performed.
도 20은 본 발명에 적용될 수 있는 무선 기기를 예시한다.20 illustrates a wireless device applicable to the present invention.
도 20을 참조하면, 제1 무선 기기(100)와 제2 무선 기기(200)는 다양한 무선 접속 기술(예, LTE, NR)을 통해 무선 신호를 송수신할 수 있다. 여기서, {제1 무선 기기(100), 제2 무선 기기(200)}은 도 19의 {무선 기기(100x), 기지국(200)} 및/또는 {무선 기기(100x), 무선 기기(100x)}에 대응할 수 있다.Referring to FIG. 20 , the first wireless device 100 and the second wireless device 200 may transmit/receive wireless signals through various wireless access technologies (eg, LTE, NR). Here, {first wireless device 100, second wireless device 200} is {wireless device 100x, base station 200} of FIG. 19 and/or {wireless device 100x, wireless device 100x) } can be matched.
제1 무선 기기(100)는 하나 이상의 프로세서(102) 및 하나 이상의 메모리(104)를 포함하며, 추가적으로 하나 이상의 송수신기(106) 및/또는 하나 이상의 안테나(108)을 더 포함할 수 있다. 프로세서(102)는 메모리(104) 및/또는 송수신기(106)를 제어하며, 앞에서 설명/제안한 기능, 절차 및/또는 방법들을 구현하도록 구성될 수 있다. 예를 들어, 프로세서(102)는 메모리(104) 내의 정보를 처리하여 제1 정보/신호를 생성한 뒤, 송수신기(106)을 통해 제1 정보/신호를 포함하는 무선 신호를 전송할 수 있다. 또한, 프로세서(102)는 송수신기(106)를 통해 제2 정보/신호를 포함하는 무선 신호를 수신한 뒤, 제2 정보/신호의 신호 처리로부터 얻은 정보를 메모리(104)에 저장할 수 있다. 메모리(104)는 프로세서(102)와 연결될 수 있고, 프로세서(102)의 동작과 관련한 다양한 정보를 저장할 수 있다. 예를 들어, 메모리(104)는 프로세서(102)에 의해 제어되는 프로세스들 중 일부 또는 전부를 수행하거나, 앞에서 설명/제안한 절차 및/또는 방법들을 수행하기 위한 명령들을 포함하는 소프트웨어 코드를 저장할 수 있다. 여기서, 프로세서(102)와 메모리(104)는 무선 통신 기술(예, LTE, NR)을 구현하도록 설계된 통신 모뎀/회로/칩의 일부일 수 있다. 송수신기(106)는 프로세서(102)와 연결될 수 있고, 하나 이상의 안테나(108)를 통해 무선 신호를 송신 및/또는 수신할 수 있다. 송수신기(106)는 송신기 및/또는 수신기를 포함할 수 있다. 송수신기(106)는 RF(Radio Frequency) 유닛과 혼용될 수 있다. 본 발명에서 무선 기기는 통신 모뎀/회로/칩을 의미할 수도 있다.The first wireless device 100 includes one or more processors 102 and one or more memories 104 , and may further include one or more transceivers 106 and/or one or more antennas 108 . The processor 102 controls the memory 104 and/or the transceiver 106 and may be configured to implement the functions, procedures and/or methods described/suggested above. For example, the processor 102 may process information in the memory 104 to generate first information/signal, and then transmit a wireless signal including the first information/signal through the transceiver 106 . In addition, the processor 102 may receive the radio signal including the second information/signal through the transceiver 106 , and then store information obtained from signal processing of the second information/signal in the memory 104 . The memory 104 may be connected to the processor 102 and may store various information related to the operation of the processor 102 . For example, the memory 104 may store software code including instructions for performing some or all of the processes controlled by the processor 102 , or for performing the procedures and/or methods described/suggested above. . Here, the processor 102 and the memory 104 may be part of a communication modem/circuit/chip designed to implement a wireless communication technology (eg, LTE, NR). The transceiver 106 may be coupled to the processor 102 , and may transmit and/or receive wireless signals via one or more antennas 108 . The transceiver 106 may include a transmitter and/or a receiver. The transceiver 106 may be used interchangeably with a radio frequency (RF) unit. In the present invention, a wireless device may refer to a communication modem/circuit/chip.
제2 무선 기기(200)는 하나 이상의 프로세서(202), 하나 이상의 메모리(204)를 포함하며, 추가적으로 하나 이상의 송수신기(206) 및/또는 하나 이상의 안테나(208)를 더 포함할 수 있다. 프로세서(202)는 메모리(204) 및/또는 송수신기(206)를 제어하며, 앞에서 설명/제안한 기능, 절차 및/또는 방법들을 구현하도록 구성될 수 있다. 예를 들어, 프로세서(202)는 메모리(204) 내의 정보를 처리하여 제3 정보/신호를 생성한 뒤, 송수신기(206)를 통해 제3 정보/신호를 포함하는 무선 신호를 전송할 수 있다. 또한, 프로세서(202)는 송수신기(206)를 통해 제4 정보/신호를 포함하는 무선 신호를 수신한 뒤, 제4 정보/신호의 신호 처리로부터 얻은 정보를 메모리(204)에 저장할 수 있다. 메모리(204)는 프로세서(202)와 연결될 수 있고, 프로세서(202)의 동작과 관련한 다양한 정보를 저장할 수 있다. 예를 들어, 메모리(204)는 프로세서(202)에 의해 제어되는 프로세스들 중 일부 또는 전부를 수행하거나, 앞에서 설명/제안한 절차 및/또는 방법들을 수행하기 위한 명령들을 포함하는 소프트웨어 코드를 저장할 수 있다. 여기서, 프로세서(202)와 메모리(204)는 무선 통신 기술(예, LTE, NR)을 구현하도록 설계된 통신 모뎀/회로/칩의 일부일 수 있다. 송수신기(206)는 프로세서(202)와 연결될 수 있고, 하나 이상의 안테나(208)를 통해 무선 신호를 송신 및/또는 수신할 수 있다. 송수신기(206)는 송신기 및/또는 수신기를 포함할 수 있다 송수신기(206)는 RF 유닛과 혼용될 수 있다. 본 발명에서 무선 기기는 통신 모뎀/회로/칩을 의미할 수도 있다.The second wireless device 200 includes one or more processors 202 , one or more memories 204 , and may further include one or more transceivers 206 and/or one or more antennas 208 . The processor 202 controls the memory 204 and/or the transceiver 206 and may be configured to implement the functions, procedures, and/or methods described/suggested above. For example, the processor 202 may process the information in the memory 204 to generate third information/signal, and then transmit a wireless signal including the third information/signal through the transceiver 206 . In addition, the processor 202 may receive the radio signal including the fourth information/signal through the transceiver 206 , and then store information obtained from signal processing of the fourth information/signal in the memory 204 . The memory 204 may be connected to the processor 202 and may store various information related to the operation of the processor 202 . For example, the memory 204 may store software code including instructions for performing some or all of the processes controlled by the processor 202 , or for performing the procedures and/or methods described/suggested above. . Here, the processor 202 and the memory 204 may be part of a communication modem/circuit/chip designed to implement a wireless communication technology (eg, LTE, NR). The transceiver 206 may be coupled to the processor 202 and may transmit and/or receive wireless signals via one or more antennas 208 . Transceiver 206 may include a transmitter and/or receiver. Transceiver 206 may be used interchangeably with an RF unit. In the present invention, a wireless device may refer to a communication modem/circuit/chip.
이하, 무선 기기(100, 200)의 하드웨어 요소에 대해 보다 구체적으로 설명한다. 이로 제한되는 것은 아니지만, 하나 이상의 프로토콜 계층이 하나 이상의 프로세서(102, 202)에 의해 구현될 수 있다. 예를 들어, 하나 이상의 프로세서(102, 202)는 하나 이상의 계층(예, PHY, MAC, RLC, PDCP, RRC, SDAP와 같은 기능적 계층)을 구현할 수 있다. 하나 이상의 프로세서(102, 202)는 본 문서에 개시된 기능, 절차, 제안 및/또는 방법에 따라 하나 이상의 PDU(Protocol Data Unit) 및/또는 하나 이상의 SDU(Service Data Unit)를 생성할 수 있다. 하나 이상의 프로세서(102, 202)는 본 문서에 개시된 기능, 절차, 제안 및/또는 방법에 따라 메시지, 제어정보, 데이터 또는 정보를 생성할 수 있다. 하나 이상의 프로세서(102, 202)는 본 문서에 개시된 기능, 절차, 제안 및/또는 방법에 따라 PDU, SDU, 메시지, 제어정보, 데이터 또는 정보를 포함하는 신호(예, 베이스밴드 신호)를 생성하여, 하나 이상의 송수신기(106, 206)에게 제공할 수 있다. 하나 이상의 프로세서(102, 202)는 하나 이상의 송수신기(106, 206)로부터 신호(예, 베이스밴드 신호)를 수신할 수 있고, 본 문서에 개시된 기능, 절차, 제안 및/또는 방법에 따라 PDU, SDU, 메시지, 제어정보, 데이터 또는 정보를 획득할 수 있다.Hereinafter, hardware elements of the wireless devices 100 and 200 will be described in more detail. Although not limited thereto, one or more protocol layers may be implemented by one or more processors 102 , 202 . For example, one or more processors 102 , 202 may implement one or more layers (eg, functional layers such as PHY, MAC, RLC, PDCP, RRC, SDAP). The one or more processors 102 and 202 may generate one or more Protocol Data Units (PDUs) and/or one or more Service Data Units (SDUs) according to the functions, procedures, proposals and/or methods disclosed herein. One or more processors 102 , 202 may generate messages, control information, data, or information according to the functions, procedures, proposals and/or methods disclosed herein. The one or more processors 102 and 202 generate a signal (eg, a baseband signal) including PDUs, SDUs, messages, control information, data or information according to the functions, procedures, proposals and/or methods disclosed herein. , to one or more transceivers 106 and 206 . One or more processors 102 , 202 may receive signals (eg, baseband signals) from one or more transceivers 106 , 206 , PDUs, SDUs, and/or SDUs according to the functions, procedures, proposals and/or methods disclosed herein. , a message, control information, data or information can be obtained.
하나 이상의 프로세서(102, 202)는 컨트롤러, 마이크로 컨트롤러, 마이크로 프로세서 또는 마이크로 컴퓨터로 지칭될 수 있다. 하나 이상의 프로세서(102, 202)는 하드웨어, 펌웨어, 소프트웨어, 또는 이들의 조합에 의해 구현될 수 있다. 일 예로, 하나 이상의 ASIC(Application Specific Integrated Circuit), 하나 이상의 DSP(Digital Signal Processor), 하나 이상의 DSPD(Digital Signal Processing Device), 하나 이상의 PLD(Programmable Logic Device) 또는 하나 이상의 FPGA(Field Programmable Gate Arrays)가 하나 이상의 프로세서(102, 202)에 포함될 수 있다. 본 문서에 개시된 기능, 절차, 제안 및/또는 방법들은 펌웨어 또는 소프트웨어를 사용하여 구현될 수 있고, 펌웨어 또는 소프트웨어는 모듈, 절차, 기능 등을 포함하도록 구현될 수 있다. 본 문서에 개시된 기능, 절차, 제안 및/또는 방법을 수행하도록 설정된 펌웨어 또는 소프트웨어는 하나 이상의 프로세서(102, 202)에 포함되거나, 하나 이상의 메모리(104, 204)에 저장되어 하나 이상의 프로세서(102, 202)에 의해 구동될 수 있다. 본 문서에 개시된 기능, 절차, 제안 및 또는 방법들은 코드, 명령어 및/또는 명령어의 집합 형태로 펌웨어 또는 소프트웨어를 사용하여 구현될 수 있다. One or more processors 102, 202 may be referred to as a controller, microcontroller, microprocessor, or microcomputer. One or more processors 102 , 202 may be implemented by hardware, firmware, software, or a combination thereof. For example, one or more Application Specific Integrated Circuits (ASICs), one or more Digital Signal Processors (DSPs), one or more Digital Signal Processing Devices (DSPDs), one or more Programmable Logic Devices (PLDs), or one or more Field Programmable Gate Arrays (FPGAs) may be included in one or more processors 102 , 202 . The functions, procedures, proposals and/or methods disclosed in this document may be implemented using firmware or software, and the firmware or software may be implemented to include modules, procedures, functions, and the like. Firmware or software configured to perform the functions, procedures, proposals, and/or methods disclosed herein is included in one or more processors 102, 202, or stored in one or more memories 104, 204, to one or more processors 102, 202) can be driven. The functions, procedures, proposals and/or methods disclosed in this document may be implemented using firmware or software in the form of code, instructions, and/or a set of instructions.
하나 이상의 메모리(104, 204)는 하나 이상의 프로세서(102, 202)와 연결될 수 있고, 다양한 형태의 데이터, 신호, 메시지, 정보, 프로그램, 코드, 지시 및/또는 명령을 저장할 수 있다. 하나 이상의 메모리(104, 204)는 ROM, RAM, EPROM, 플래시 메모리, 하드 드라이브, 레지스터, 캐쉬 메모리, 컴퓨터 판독 저장 매체 및/또는 이들의 조합으로 구성될 수 있다. 하나 이상의 메모리(104, 204)는 하나 이상의 프로세서(102, 202)의 내부 및/또는 외부에 위치할 수 있다. 또한, 하나 이상의 메모리(104, 204)는 유선 또는 무선 연결과 같은 다양한 기술을 통해 하나 이상의 프로세서(102, 202)와 연결될 수 있다.One or more memories 104 , 204 may be coupled with one or more processors 102 , 202 , and may store various forms of data, signals, messages, information, programs, code, instructions, and/or instructions. The one or more memories 104 and 204 may be comprised of ROM, RAM, EPROM, flash memory, hard drives, registers, cache memory, computer readable storage media, and/or combinations thereof. One or more memories 104 , 204 may be located inside and/or external to one or more processors 102 , 202 . Additionally, one or more memories 104 , 204 may be coupled to one or more processors 102 , 202 through various technologies, such as wired or wireless connections.
하나 이상의 송수신기(106, 206)는 하나 이상의 다른 장치에게 본 문서의 방법들 및/또는 동작 순서도 등에서 언급되는 사용자 데이터, 제어 정보, 무선 신호/채널 등을 전송할 수 있다. 하나 이상의 송수신기(106, 206)는 하나 이상의 다른 장치로부터 본 문서에 개시된 기능, 절차, 제안, 방법 및/또는 동작 순서도 등에서 언급되는 사용자 데이터, 제어 정보, 무선 신호/채널 등을 수신할 수 있다. 예를 들어, 하나 이상의 송수신기(106, 206)는 하나 이상의 프로세서(102, 202)와 연결될 수 있고, 무선 신호를 송수신할 수 있다. 예를 들어, 하나 이상의 프로세서(102, 202)는 하나 이상의 송수신기(106, 206)가 하나 이상의 다른 장치에게 사용자 데이터, 제어 정보 또는 무선 신호를 전송하도록 제어할 수 있다. 또한, 하나 이상의 프로세서(102, 202)는 하나 이상의 송수신기(106, 206)가 하나 이상의 다른 장치로부터 사용자 데이터, 제어 정보 또는 무선 신호를 수신하도록 제어할 수 있다. 또한, 하나 이상의 송수신기(106, 206)는 하나 이상의 안테나(108, 208)와 연결될 수 있고, 하나 이상의 송수신기(106, 206)는 하나 이상의 안테나(108, 208)를 통해 본 문서에 개시된 기능, 절차, 제안, 방법 및/또는 동작 순서도 등에서 언급되는 사용자 데이터, 제어 정보, 무선 신호/채널 등을 송수신하도록 설정될 수 있다. 본 문서에서, 하나 이상의 안테나는 복수의 물리 안테나이거나, 복수의 논리 안테나(예, 안테나 포트)일 수 있다. 하나 이상의 송수신기(106, 206)는 수신된 사용자 데이터, 제어 정보, 무선 신호/채널 등을 하나 이상의 프로세서(102, 202)를 이용하여 처리하기 위해, 수신된 무선 신호/채널 등을 RF 밴드 신호에서 베이스밴드 신호로 변환(Convert)할 수 있다. 하나 이상의 송수신기(106, 206)는 하나 이상의 프로세서(102, 202)를 이용하여 처리된 사용자 데이터, 제어 정보, 무선 신호/채널 등을 베이스밴드 신호에서 RF 밴드 신호로 변환할 수 있다. 이를 위하여, 하나 이상의 송수신기(106, 206)는 (아날로그) 오실레이터 및/또는 필터를 포함할 수 있다.One or more transceivers 106 , 206 may transmit user data, control information, radio signals/channels, etc. referred to in the methods and/or operational flowcharts of this document to one or more other devices. The one or more transceivers 106, 206 may receive user data, control information, radio signals/channels, etc., referred to in the functions, procedures, proposals, methods, and/or flowcharts of operations disclosed herein, or the like, from one or more other devices. For example, one or more transceivers 106 , 206 may be coupled to one or more processors 102 , 202 and may transmit and receive wireless signals. For example, one or more processors 102 , 202 may control one or more transceivers 106 , 206 to transmit user data, control information, or wireless signals to one or more other devices. In addition, one or more processors 102 , 202 may control one or more transceivers 106 , 206 to receive user data, control information, or wireless signals from one or more other devices. Further, one or more transceivers 106, 206 may be coupled to one or more antennas 108, 208, and the one or more transceivers 106, 206 may be coupled to one or more of the transceivers 106, 206 via the one or more antennas 108, 208 for the functions, procedures, and procedures disclosed herein. , may be set to transmit and receive user data, control information, radio signals/channels, etc. mentioned in a proposal, a method and/or an operation flowchart. In this document, one or more antennas may be a plurality of physical antennas or a plurality of logical antennas (eg, antenna ports). The one or more transceivers 106, 206 convert the received radio signal/channel, etc. from the RF band signal to process the received user data, control information, radio signal/channel, etc. using the one or more processors 102, 202. It can be converted into a baseband signal. One or more transceivers 106 and 206 may convert user data, control information, radio signals/channels, etc. processed using one or more processors 102 and 202 from baseband signals to RF band signals. To this end, one or more transceivers 106 , 206 may include (analog) oscillators and/or filters.
도 21는 전송 신호를 위한 신호 처리 회로를 예시한다.21 illustrates a signal processing circuit for a transmission signal.
도 21를 참조하면, 신호 처리 회로(1000)는 스크램블러(1010), 변조기(1020), 레이어 매퍼(1030), 프리코더(1040), 자원 매퍼(1050), 신호 생성기(1060)를 포함할 수 있다. 이로 제한되는 것은 아니지만, 도 21의 동작/기능은 도 20의 프로세서(102, 202) 및/또는 송수신기(106, 206)에서 수행될 수 있다. 도 21의 하드웨어 요소는 도 20의 프로세서(102, 202) 및/또는 송수신기(106, 206)에서 구현될 수 있다. 예를 들어, 블록 1010~1060은 도 20의 프로세서(102, 202)에서 구현될 수 있다. 또한, 블록 1010~1050은 도 20의 프로세서(102, 202)에서 구현되고, 블록 1060은 도 20의 송수신기(106, 206)에서 구현될 수 있다.Referring to FIG. 21 , the signal processing circuit 1000 may include a scrambler 1010 , a modulator 1020 , a layer mapper 1030 , a precoder 1040 , a resource mapper 1050 , and a signal generator 1060 . there is. Although not limited thereto, the operations/functions of FIG. 21 may be performed by the processors 102 and 202 and/or the transceivers 106 and 206 of FIG. 20 . The hardware elements of FIG. 21 may be implemented in the processors 102 , 202 and/or transceivers 106 , 206 of FIG. 20 . For example, blocks 1010 to 1060 may be implemented in the processors 102 and 202 of FIG. 20 . Further, blocks 1010 to 1050 may be implemented in the processors 102 and 202 of FIG. 20 , and block 1060 may be implemented in the transceivers 106 and 206 of FIG. 20 .
코드워드는 도 21의 신호 처리 회로(1000)를 거쳐 무선 신호로 변환될 수 있다. 여기서, 코드워드는 정보블록의 부호화된 비트 시퀀스이다. 정보블록은 전송블록(예, UL-SCH 전송블록, DL-SCH 전송블록)을 포함할 수 있다. 무선 신호는 도 A1의 다양한 물리 채널(예, PUSCH, PDSCH)을 통해 전송될 수 있다.The codeword may be converted into a wireless signal through the signal processing circuit 1000 of FIG. 21 . Here, the codeword is a coded bit sequence of an information block. The information block may include a transport block (eg, a UL-SCH transport block, a DL-SCH transport block). The radio signal may be transmitted through various physical channels (eg, PUSCH, PDSCH) of FIG. A1 .
구체적으로, 코드워드는 스크램블러(1010)에 의해 스크램블된 비트 시퀀스로 변환될 수 있다. 스크램블에 사용되는 스크램블 시퀀스는 초기화 값에 기반하여 생성되며, 초기화 값은 무선 기기의 ID 정보 등이 포함될 수 있다. 스크램블된 비트 시퀀스는 변조기(1020)에 의해 변조 심볼 시퀀스로 변조될 수 있다. 변조 방식은 pi/2-BPSK(pi/2-Binary Phase Shift Keying), m-PSK(m-Phase Shift Keying), m-QAM(m-Quadrature Amplitude Modulation) 등을 포함할 수 있다. 복소 변조 심볼 시퀀스는 레이어 매퍼(1030)에 의해 하나 이상의 전송 레이어로 매핑될 수 있다. 각 전송 레이어의 변조 심볼들은 프리코더(1040)에 의해 해당 안테나 포트(들)로 매핑될 수 있다(프리코딩). 프리코더(1040)의 출력 z는 레이어 매퍼(1030)의 출력 y를 N*M의 프리코딩 행렬 W와 곱해 얻을 수 있다. 여기서, N은 안테나 포트의 개수, M은 전송 레이어의 개수이다. 여기서, 프리코더(1040)는 복소 변조 심볼들에 대한 트랜스폼(transform) 프리코딩(예, DFT 변환)을 수행한 이후에 프리코딩을 수행할 수 있다. 또한, 프리코더(1040)는 트랜스폼 프리코딩을 수행하지 않고 프리코딩을 수행할 수 있다.Specifically, the codeword may be converted into a scrambled bit sequence by the scrambler 1010 . A scramble sequence used for scrambling is generated based on an initialization value, and the initialization value may include ID information of a wireless device, and the like. The scrambled bit sequence may be modulated by a modulator 1020 into a modulation symbol sequence. The modulation method may include pi/2-Binary Phase Shift Keying (pi/2-BPSK), m-Phase Shift Keying (m-PSK), m-Quadrature Amplitude Modulation (m-QAM), and the like. The complex modulation symbol sequence may be mapped to one or more transport layers by the layer mapper 1030 . Modulation symbols of each transport layer may be mapped to corresponding antenna port(s) by the precoder 1040 (precoding). The output z of the precoder 1040 may be obtained by multiplying the output y of the layer mapper 1030 by the precoding matrix W of N*M. Here, N is the number of antenna ports, and M is the number of transport layers. Here, the precoder 1040 may perform precoding after performing transform precoding (eg, DFT transform) on the complex modulation symbols. Also, the precoder 1040 may perform precoding without performing transform precoding.
자원 매퍼(1050)는 각 안테나 포트의 변조 심볼들을 시간-주파수 자원에 매핑할 수 있다. 시간-주파수 자원은 시간 도메인에서 복수의 심볼(예, CP-OFDMA 심볼, DFT-s-OFDMA 심볼)을 포함하고, 주파수 도메인에서 복수의 부반송파를 포함할 수 있다. 신호 생성기(1060)는 매핑된 변조 심볼들로부터 무선 신호를 생성하며, 생성된 무선 신호는 각 안테나를 통해 다른 기기로 전송될 수 있다. 이를 위해, 신호 생성기(1060)는 IFFT(Inverse Fast Fourier Transform) 모듈 및 CP(Cyclic Prefix) 삽입기, DAC(Digital-to-Analog Converter), 주파수 상향 변환기(frequency uplink converter) 등을 포함할 수 있다.The resource mapper 1050 may map modulation symbols of each antenna port to a time-frequency resource. The time-frequency resource may include a plurality of symbols (eg, a CP-OFDMA symbol, a DFT-s-OFDMA symbol) in the time domain and a plurality of subcarriers in the frequency domain. The signal generator 1060 generates a radio signal from the mapped modulation symbols, and the generated radio signal may be transmitted to another device through each antenna. To this end, the signal generator 1060 may include an Inverse Fast Fourier Transform (IFFT) module and a Cyclic Prefix (CP) inserter, a Digital-to-Analog Converter (DAC), a frequency uplink converter, and the like. .
무선 기기에서 수신 신호를 위한 신호 처리 과정은 도 21의 신호 처리 과정(1010~1060)의 역으로 구성될 수 있다. 예를 들어, 무선 기기(예, 도 20의 100, 200)는 안테나 포트/송수신기를 통해 외부로부터 무선 신호를 수신할 수 있다. 수신된 무선 신호는 신호 복원기를 통해 베이스밴드 신호로 변환될 수 있다. 이를 위해, 신호 복원기는 주파수 하향 변환기(frequency downlink converter), ADC(analog-to-digital converter), CP 제거기, FFT(Fast Fourier Transform) 모듈을 포함할 수 있다. 이후, 베이스밴드 신호는 자원 디-매퍼 과정, 포스트코딩(postcoding) 과정, 복조 과정 및 디-스크램블 과정을 거쳐 코드워드로 복원될 수 있다. 코드워드는 복호(decoding)를 거쳐 원래의 정보블록으로 복원될 수 있다. 따라서, 수신 신호를 위한 신호 처리 회로(미도시)는 신호 복원기, 자원 디-매퍼, 포스트코더, 복조기, 디-스크램블러 및 복호기를 포함할 수 있다.The signal processing process for the received signal in the wireless device may be configured in reverse of the signal processing process 1010 to 1060 of FIG. 21 . For example, the wireless device (eg, 100 and 200 in FIG. 20 ) may receive a wireless signal from the outside through an antenna port/transceiver. The received radio signal may be converted into a baseband signal through a signal restorer. To this end, the signal restorer may include a frequency downlink converter, an analog-to-digital converter (ADC), a CP remover, and a Fast Fourier Transform (FFT) module. Thereafter, the baseband signal may be restored to a codeword through a resource de-mapper process, a postcoding process, a demodulation process, and a descrambling process. The codeword may be restored to the original information block through decoding. Accordingly, the signal processing circuit (not shown) for the received signal may include a signal restorer, a resource de-mapper, a post coder, a demodulator, a descrambler, and a decoder.
도 22은 본 발명에 적용되는 무선 기기의 다른 예를 나타낸다. 무선 기기는 사용-예/서비스에 따라 다양한 형태로 구현될 수 있다(도 19, 도 23~26 참조).22 shows another example of a wireless device to which the present invention is applied. The wireless device may be implemented in various forms according to use-examples/services (refer to FIGS. 19 and 23 to 26 ).
도 22을 참조하면, 무선 기기(100, 200)는 도 20의 무선 기기(100,200)에 대응하며, 다양한 요소(element), 성분(component), 유닛/부(unit), 및/또는 모듈(module)로 구성될 수 있다. 예를 들어, 무선 기기(100, 200)는 통신부(110), 제어부(120), 메모리부(130) 및 추가 요소(140)를 포함할 수 있다. 통신부는 통신 회로(112) 및 송수신기(들)(114)을 포함할 수 있다. 예를 들어, 통신 회로(112)는 도 20의 하나 이상의 프로세서(102,202) 및/또는 하나 이상의 메모리(104,204) 를 포함할 수 있다. 예를 들어, 송수신기(들)(114)는 도 20의 하나 이상의 송수신기(106,206) 및/또는 하나 이상의 안테나(108,208)을 포함할 수 있다. 제어부(120)는 통신부(110), 메모리부(130) 및 추가 요소(140)와 전기적으로 연결되며 무선 기기의 제반 동작을 제어한다. 예를 들어, 제어부(120)는 메모리부(130)에 저장된 프로그램/코드/명령/정보에 기반하여 무선 기기의 전기적/기계적 동작을 제어할 수 있다. 또한, 제어부(120)는 메모리부(130)에 저장된 정보를 통신부(110)을 통해 외부(예, 다른 통신 기기)로 무선/유선 인터페이스를 통해 전송하거나, 통신부(110)를 통해 외부(예, 다른 통신 기기)로부터 무선/유선 인터페이스를 통해 수신된 정보를 메모리부(130)에 저장할 수 있다.Referring to FIG. 22 , wireless devices 100 and 200 correspond to wireless devices 100 and 200 of FIG. 20 , and include various elements, components, units/units, and/or modules. ) may consist of For example, the wireless devices 100 and 200 may include a communication unit 110 , a control unit 120 , a memory unit 130 , and an additional element 140 . The communication unit may include communication circuitry 112 and transceiver(s) 114 . For example, communication circuitry 112 may include one or more processors 102,202 and/or one or more memories 104,204 of FIG. 20 . For example, the transceiver(s) 114 may include one or more transceivers 106 , 206 and/or one or more antennas 108 , 208 of FIG. 20 . The control unit 120 is electrically connected to the communication unit 110 , the memory unit 130 , and the additional element 140 , and controls general operations of the wireless device. For example, the controller 120 may control the electrical/mechanical operation of the wireless device based on the program/code/command/information stored in the memory unit 130 . In addition, the control unit 120 transmits information stored in the memory unit 130 to the outside (eg, other communication device) through the communication unit 110 through a wireless/wired interface, or externally (eg, through the communication unit 110 ) Information received through a wireless/wired interface from another communication device) may be stored in the memory unit 130 .
추가 요소(140)는 무선 기기의 종류에 따라 다양하게 구성될 수 있다. 예를 들어, 추가 요소(140)는 파워 유닛/배터리, 입출력부(I/O unit), 구동부 및 컴퓨팅부 중 적어도 하나를 포함할 수 있다. 이로 제한되는 것은 아니지만, 무선 기기는 로봇(도 19, 100a), 차량(도 19, 100b-1, 100b-2), XR 기기(도 19, 100c), 휴대 기기(도 19, 100d), 가전(도 19, 100e), IoT 기기(도 19, 100f), 디지털 방송용 단말, 홀로그램 장치, 공공 안전 장치, MTC 장치, 의료 장치, 핀테크 장치(또는 금융 장치), 보안 장치, 기후/환경 장치, AI 서버/기기(도 19, 400), 기지국(도 19, 200), 네트워크 노드 등의 형태로 구현될 수 있다. 무선 기기는 사용-예/서비스에 따라 이동 가능하거나 고정된 장소에서 사용될 수 있다.The additional element 140 may be configured in various ways according to the type of the wireless device. For example, the additional element 140 may include at least one of a power unit/battery, an input/output unit (I/O unit), a driving unit, and a computing unit. Although not limited thereto, the wireless device includes a robot ( FIGS. 19 and 100a ), a vehicle ( FIGS. 19 , 100b-1 , 100b-2 ), an XR device ( FIGS. 19 and 100c ), a mobile device ( FIGS. 19 and 100d ), and a home appliance. (FIG. 19, 100e), IoT device (FIG. 19, 100f), digital broadcasting terminal, hologram device, public safety device, MTC device, medical device, fintech device (or financial device), security device, climate/environment device, It may be implemented in the form of an AI server/device ( FIGS. 19 and 400 ), a base station ( FIGS. 19 and 200 ), and a network node. The wireless device may be mobile or used in a fixed location depending on the use-example/service.
도 22에서 무선 기기(100, 200) 내의 다양한 요소, 성분, 유닛/부, 및/또는 모듈은 전체가 유선 인터페이스를 통해 상호 연결되거나, 적어도 일부가 통신부(110)를 통해 무선으로 연결될 수 있다. 예를 들어, 무선 기기(100, 200) 내에서 제어부(120)와 통신부(110)는 유선으로 연결되며, 제어부(120)와 제1 유닛(예, 130, 140)은 통신부(110)를 통해 무선으로 연결될 수 있다. 또한, 무선 기기(100, 200) 내의 각 요소, 성분, 유닛/부, 및/또는 모듈은 하나 이상의 요소를 더 포함할 수 있다. 예를 들어, 제어부(120)는 하나 이상의 프로세서 집합으로 구성될 수 있다. 예를 들어, 제어부(120)는 통신 제어 프로세서, 어플리케이션 프로세서(Application processor), ECU(Electronic Control Unit), 그래픽 처리 프로세서, 메모리 제어 프로세서 등의 집합으로 구성될 수 있다. 다른 예로, 메모리부(130)는 RAM(Random Access Memory), DRAM(Dynamic RAM), ROM(Read Only Memory), 플래시 메모리(flash memory), 휘발성 메모리(volatile memory), 비-휘발성 메모리(non-volatile memory) 및/또는 이들의 조합으로 구성될 수 있다.In FIG. 22 , various elements, components, units/units, and/or modules in the wireless devices 100 and 200 may be entirely interconnected through a wired interface, or at least some of them may be wirelessly connected through the communication unit 110 . For example, in the wireless devices 100 and 200 , the control unit 120 and the communication unit 110 are connected by wire, and the control unit 120 and the first unit (eg, 130 , 140 ) are connected to the communication unit 110 through the communication unit 110 . It can be connected wirelessly. In addition, each element, component, unit/unit, and/or module within the wireless device 100 , 200 may further include one or more elements. For example, the controller 120 may be configured with one or more processor sets. For example, the control unit 120 may be configured as a set of a communication control processor, an application processor, an electronic control unit (ECU), a graphic processing processor, a memory control processor, and the like. As another example, the memory unit 130 may include random access memory (RAM), dynamic RAM (DRAM), read only memory (ROM), flash memory, volatile memory, and non-volatile memory. volatile memory) and/or a combination thereof.
이하, 도 22의 구현 예에 대해 도면을 참조하여 보다 자세히 설명한다.Hereinafter, the embodiment of FIG. 22 will be described in more detail with reference to the drawings.
도 23는 본 발명에 적용되는 휴대 기기를 예시한다. 휴대 기기는 스마트폰, 스마트패드, 웨어러블 기기(예, 스마트워치, 스마트글래스), 휴대용 컴퓨터(예, 노트북 등)을 포함할 수 있다. 휴대 기기는 MS(Mobile Station), UT(user terminal), MSS(Mobile Subscriber Station), SS(Subscriber Station), AMS(Advanced Mobile Station) 또는 WT(Wireless terminal)로 지칭될 수 있다.23 illustrates a portable device to which the present invention is applied. The portable device may include a smart phone, a smart pad, a wearable device (eg, a smart watch, smart glasses), and a portable computer (eg, a laptop computer). A mobile device may be referred to as a mobile station (MS), a user terminal (UT), a mobile subscriber station (MSS), a subscriber station (SS), an advanced mobile station (AMS), or a wireless terminal (WT).
도 23를 참조하면, 휴대 기기(100)는 안테나부(108), 통신부(110), 제어부(120), 메모리부(130), 전원공급부(140a), 인터페이스부(140b) 및 입출력부(140c)를 포함할 수 있다. 안테나부(108)는 통신부(110)의 일부로 구성될 수 있다. 블록 110~130/140a~140c는 각각 도 22의 블록 110~130/140에 대응한다.Referring to FIG. 23 , the portable device 100 includes an antenna unit 108 , a communication unit 110 , a control unit 120 , a memory unit 130 , a power supply unit 140a , an interface unit 140b , and an input/output unit 140c . ) may be included. The antenna unit 108 may be configured as a part of the communication unit 110 . Blocks 110 to 130/140a to 140c respectively correspond to blocks 110 to 130/140 of FIG. 22 .
통신부(110)는 다른 무선 기기, 기지국들과 신호(예, 데이터, 제어 신호 등)를 송수신할 수 있다. 제어부(120)는 휴대 기기(100)의 구성 요소들을 제어하여 다양한 동작을 수행할 수 있다. 제어부(120)는 AP(Application Processor)를 포함할 수 있다. 메모리부(130)는 휴대 기기(100)의 구동에 필요한 데이터/파라미터/프로그램/코드/명령을 저장할 수 있다. 또한, 메모리부(130)는 입/출력되는 데이터/정보 등을 저장할 수 있다. 전원공급부(140a)는 휴대 기기(100)에게 전원을 공급하며, 유/무선 충전 회로, 배터리 등을 포함할 수 있다. 인터페이스부(140b)는 휴대 기기(100)와 다른 외부 기기의 연결을 지원할 수 있다. 인터페이스부(140b)는 외부 기기와의 연결을 위한 다양한 포트(예, 오디오 입/출력 포트, 비디오 입/출력 포트)를 포함할 수 있다. 입출력부(140c)는 영상 정보/신호, 오디오 정보/신호, 데이터, 및/또는 사용자로부터 입력되는 정보를 입력 받거나 출력할 수 있다. 입출력부(140c)는 카메라, 마이크로폰, 사용자 입력부, 디스플레이부(140d), 스피커 및/또는 햅틱 모듈 등을 포함할 수 있다.The communication unit 110 may transmit and receive signals (eg, data, control signals, etc.) with other wireless devices and base stations. The controller 120 may control components of the portable device 100 to perform various operations. The controller 120 may include an application processor (AP). The memory unit 130 may store data/parameters/programs/codes/commands necessary for driving the portable device 100 . Also, the memory unit 130 may store input/output data/information. The power supply unit 140a supplies power to the portable device 100 and may include a wired/wireless charging circuit, a battery, and the like. The interface unit 140b may support a connection between the portable device 100 and other external devices. The interface unit 140b may include various ports (eg, an audio input/output port and a video input/output port) for connection with an external device. The input/output unit 140c may receive or output image information/signal, audio information/signal, data, and/or information input from a user. The input/output unit 140c may include a camera, a microphone, a user input unit, a display unit 140d, a speaker, and/or a haptic module.
일 예로, 데이터 통신의 경우, 입출력부(140c)는 사용자로부터 입력된 정보/신호(예, 터치, 문자, 음성, 이미지, 비디오)를 획득하며, 획득된 정보/신호는 메모리부(130)에 저장될 수 있다. 통신부(110)는 메모리에 저장된 정보/신호를 무선 신호로 변환하고, 변환된 무선 신호를 다른 무선 기기에게 직접 전송하거나 기지국에게 전송할 수 있다. 또한, 통신부(110)는 다른 무선 기기 또는 기지국으로부터 무선 신호를 수신한 뒤, 수신된 무선 신호를 원래의 정보/신호로 복원할 수 있다. 복원된 정보/신호는 메모리부(130)에 저장된 뒤, 입출력부(140c)를 통해 다양한 형태(예, 문자, 음성, 이미지, 비디오, 헵틱)로 출력될 수 있다. For example, in the case of data communication, the input/output unit 140c obtains information/signals (eg, touch, text, voice, image, video) input from the user, and the obtained information/signals are stored in the memory unit 130 . can be saved. The communication unit 110 may convert the information/signal stored in the memory into a wireless signal, and transmit the converted wireless signal directly to another wireless device or to a base station. Also, after receiving a radio signal from another radio device or base station, the communication unit 110 may restore the received radio signal to original information/signal. After the restored information/signal is stored in the memory unit 130 , it may be output in various forms (eg, text, voice, image, video, haptic) through the input/output unit 140c.
도 24는 본 발명에 적용되는 차량 또는 자율 주행 차량을 예시한다. 차량 또는 자율 주행 차량은 이동형 로봇, 차량, 기차, 유/무인 비행체(Aerial Vehicle, AV), 선박 등으로 구현될 수 있다.24 illustrates a vehicle or an autonomous driving vehicle to which the present invention is applied. The vehicle or autonomous driving vehicle may be implemented as a mobile robot, a vehicle, a train, an aerial vehicle (AV), a ship, and the like.
도 24를 참조하면, 차량 또는 자율 주행 차량(100)은 안테나부(108), 통신부(110), 제어부(120), 구동부(140a), 전원공급부(140b), 센서부(140c) 및 자율 주행부(140d)를 포함할 수 있다. 안테나부(108)는 통신부(110)의 일부로 구성될 수 있다. 블록 110/130/140a~140d는 각각 도 22의 블록 110/130/140에 대응한다.Referring to FIG. 24 , the vehicle or autonomous driving vehicle 100 includes an antenna unit 108 , a communication unit 110 , a control unit 120 , a driving unit 140a , a power supply unit 140b , a sensor unit 140c and autonomous driving. It may include a part 140d. The antenna unit 108 may be configured as a part of the communication unit 110 . Blocks 110/130/140a-140d correspond to blocks 110/130/140 of FIG. 22, respectively.
통신부(110)는 다른 차량, 기지국(e.g. 기지국, 노변 기지국(Road Side unit) 등), 서버 등의 외부 기기들과 신호(예, 데이터, 제어 신호 등)를 송수신할 수 있다. 제어부(120)는 차량 또는 자율 주행 차량(100)의 요소들을 제어하여 다양한 동작을 수행할 수 있다. 제어부(120)는 ECU(Electronic Control Unit)를 포함할 수 있다. 구동부(140a)는 차량 또는 자율 주행 차량(100)을 지상에서 주행하게 할 수 있다. 구동부(140a)는 엔진, 모터, 파워 트레인, 바퀴, 브레이크, 조향 장치 등을 포함할 수 있다. 전원공급부(140b)는 차량 또는 자율 주행 차량(100)에게 전원을 공급하며, 유/무선 충전 회로, 배터리 등을 포함할 수 있다. 센서부(140c)는 차량 상태, 주변 환경 정보, 사용자 정보 등을 얻을 수 있다. 센서부(140c)는 IMU(inertial measurement unit) 센서, 충돌 센서, 휠 센서(wheel sensor), 속도 센서, 경사 센서, 중량 감지 센서, 헤딩 센서(heading sensor), 포지션 모듈(position module), 차량 전진/후진 센서, 배터리 센서, 연료 센서, 타이어 센서, 스티어링 센서, 온도 센서, 습도 센서, 초음파 센서, 조도 센서, 페달 포지션 센서 등을 포함할 수 있다. 자율 주행부(140d)는 주행중인 차선을 유지하는 기술, 어댑티브 크루즈 컨트롤과 같이 속도를 자동으로 조절하는 기술, 정해진 경로를 따라 자동으로 주행하는 기술, 목적지가 설정되면 자동으로 경로를 설정하여 주행하는 기술 등을 구현할 수 있다.The communication unit 110 may transmit/receive signals (eg, data, control signals, etc.) to and from external devices such as other vehicles, base stations (e.g., base stations, roadside units, etc.), servers, and the like. The controller 120 may control elements of the vehicle or the autonomous driving vehicle 100 to perform various operations. The controller 120 may include an Electronic Control Unit (ECU). The driving unit 140a may cause the vehicle or the autonomous driving vehicle 100 to run on the ground. The driving unit 140a may include an engine, a motor, a power train, a wheel, a brake, a steering device, and the like. The power supply unit 140b supplies power to the vehicle or the autonomous driving vehicle 100 , and may include a wired/wireless charging circuit, a battery, and the like. The sensor unit 140c may obtain vehicle status, surrounding environment information, user information, and the like. The sensor unit 140c includes an inertial measurement unit (IMU) sensor, a collision sensor, a wheel sensor, a speed sensor, an inclination sensor, a weight sensor, a heading sensor, a position module, and a vehicle forward movement. / may include a reverse sensor, a battery sensor, a fuel sensor, a tire sensor, a steering sensor, a temperature sensor, a humidity sensor, an ultrasonic sensor, an illuminance sensor, a pedal position sensor, and the like. The autonomous driving unit 140d includes a technology for maintaining a driving lane, a technology for automatically adjusting speed such as adaptive cruise control, a technology for automatically driving along a predetermined route, and a technology for automatically setting a route when a destination is set. technology can be implemented.
일 예로, 통신부(110)는 외부 서버로부터 지도 데이터, 교통 정보 데이터 등을 수신할 수 있다. 자율 주행부(140d)는 획득된 데이터를 기반으로 자율 주행 경로와 드라이빙 플랜을 생성할 수 있다. 제어부(120)는 드라이빙 플랜에 따라 차량 또는 자율 주행 차량(100)이 자율 주행 경로를 따라 이동하도록 구동부(140a)를 제어할 수 있다(예, 속도/방향 조절). 자율 주행 도중에 통신부(110)는 외부 서버로부터 최신 교통 정보 데이터를 비/주기적으로 획득하며, 주변 차량으로부터 주변 교통 정보 데이터를 획득할 수 있다. 또한, 자율 주행 도중에 센서부(140c)는 차량 상태, 주변 환경 정보를 획득할 수 있다. 자율 주행부(140d)는 새로 획득된 데이터/정보에 기반하여 자율 주행 경로와 드라이빙 플랜을 갱신할 수 있다. 통신부(110)는 차량 위치, 자율 주행 경로, 드라이빙 플랜 등에 관한 정보를 외부 서버로 전달할 수 있다. 외부 서버는 차량 또는 자율 주행 차량들로부터 수집된 정보에 기반하여, AI 기술 등을 이용하여 교통 정보 데이터를 미리 예측할 수 있고, 예측된 교통 정보 데이터를 차량 또는 자율 주행 차량들에게 제공할 수 있다.For example, the communication unit 110 may receive map data, traffic information data, and the like from an external server. The autonomous driving unit 140d may generate an autonomous driving route and a driving plan based on the acquired data. The controller 120 may control the driving unit 140a to move the vehicle or the autonomous driving vehicle 100 along the autonomous driving path (eg, speed/direction adjustment) according to the driving plan. During autonomous driving, the communication unit 110 may obtain the latest traffic information data from an external server non/periodically, and may acquire surrounding traffic information data from surrounding vehicles. Also, during autonomous driving, the sensor unit 140c may acquire vehicle state and surrounding environment information. The autonomous driving unit 140d may update the autonomous driving route and driving plan based on the newly acquired data/information. The communication unit 110 may transmit information about a vehicle location, an autonomous driving route, a driving plan, and the like to an external server. The external server may predict traffic information data in advance using AI technology or the like based on information collected from the vehicle or autonomous driving vehicles, and may provide the predicted traffic information data to the vehicle or autonomous driving vehicles.
도 25은 본 발명에 적용되는 차량을 예시한다. 차량은 운송수단, 기차, 비행체, 선박 등으로도 구현될 수 있다.25 illustrates a vehicle to which the present invention is applied. The vehicle may also be implemented as a means of transportation, a train, an air vehicle, a ship, and the like.
도 25을 참조하면, 차량(100)은 통신부(110), 제어부(120), 메모리부(130), 입출력부(140a) 및 위치 측정부(140b)를 포함할 수 있다. 여기서, 블록 110~130/140a~140b는 각각 도 22의 블록 110~130/140에 대응한다.Referring to FIG. 25 , the vehicle 100 may include a communication unit 110 , a control unit 120 , a memory unit 130 , an input/output unit 140a , and a position measurement unit 140b . Here, blocks 110 to 130/140a to 140b correspond to blocks 110 to 130/140 of FIG. 22 , respectively.
통신부(110)는 다른 차량, 또는 기지국 등의 외부 기기들과 신호(예, 데이터, 제어 신호 등)를 송수신할 수 있다. 제어부(120)는 차량(100)의 구성 요소들을 제어하여 다양한 동작을 수행할 수 있다. 메모리부(130)는 차량(100)의 다양한 기능을 지원하는 데이터/파라미터/프로그램/코드/명령을 저장할 수 있다. 입출력부(140a)는 메모리부(130) 내의 정보에 기반하여 AR/VR 오브젝트를 출력할 수 있다. 입출력부(140a)는 HUD를 포함할 수 있다. 위치 측정부(140b)는 차량(100)의 위치 정보를 획득할 수 있다. 위치 정보는 차량(100)의 절대 위치 정보, 주행선 내에서의 위치 정보, 가속도 정보, 주변 차량과의 위치 정보 등을 포함할 수 있다. 위치 측정부(140b)는 GPS 및 다양한 센서들을 포함할 수 있다.The communication unit 110 may transmit and receive signals (eg, data, control signals, etc.) with other vehicles or external devices such as a base station. The controller 120 may control components of the vehicle 100 to perform various operations. The memory unit 130 may store data/parameters/programs/codes/commands supporting various functions of the vehicle 100 . The input/output unit 140a may output an AR/VR object based on information in the memory unit 130 . The input/output unit 140a may include a HUD. The position measuring unit 140b may acquire position information of the vehicle 100 . The location information may include absolute location information of the vehicle 100 , location information within a driving line, acceleration information, location information with a surrounding vehicle, and the like. The position measuring unit 140b may include a GPS and various sensors.
일 예로, 차량(100)의 통신부(110)는 외부 서버로부터 지도 정보, 교통 정보 등을 수신하여 메모리부(130)에 저장할 수 있다. 위치 측정부(140b)는 GPS 및 다양한 센서를 통하여 차량 위치 정보를 획득하여 메모리부(130)에 저장할 수 있다. 제어부(120)는 지도 정보, 교통 정보 및 차량 위치 정보 등에 기반하여 가상 오브젝트를 생성하고, 입출력부(140a)는 생성된 가상 오브젝트를 차량 내 유리창에 표시할 수 있다(1410, 1420). 또한, 제어부(120)는 차량 위치 정보에 기반하여 차량(100)이 주행선 내에서 정상적으로 운행되고 있는지 판단할 수 있다. 차량(100)이 주행선을 비정상적으로 벗어나는 경우, 제어부(120)는 입출력부(140a)를 통해 차량 내 유리창에 경고를 표시할 수 있다. 또한, 제어부(120)는 통신부(110)를 통해 주변 차량들에게 주행 이상에 관한 경고 메세지를 방송할 수 있다. 상황에 따라, 제어부(120)는 통신부(110)를 통해 관계 기관에게 차량의 위치 정보와, 주행/차량 이상에 관한 정보를 전송할 수 있다. For example, the communication unit 110 of the vehicle 100 may receive map information, traffic information, and the like from an external server and store it in the memory unit 130 . The position measuring unit 140b may acquire vehicle position information through GPS and various sensors and store it in the memory unit 130 . The controller 120 may generate a virtual object based on map information, traffic information, vehicle location information, and the like, and the input/output unit 140a may display the created virtual object on a window inside the vehicle ( 1410 and 1420 ). In addition, the controller 120 may determine whether the vehicle 100 is normally operating within the driving line based on the vehicle location information. When the vehicle 100 deviates from the driving line abnormally, the controller 120 may display a warning on the windshield of the vehicle through the input/output unit 140a. Also, the control unit 120 may broadcast a warning message regarding driving abnormality to surrounding vehicles through the communication unit 110 . Depending on the situation, the control unit 120 may transmit the location information of the vehicle and information on driving/vehicle abnormality to a related organization through the communication unit 110 .
도 26는 본 발명에 적용되는 XR 기기를 예시한다. XR 기기는 HMD, 차량에 구비된 HUD(Head-Up Display), 텔레비전, 스마트폰, 컴퓨터, 웨어러블 디바이스, 가전 기기, 디지털 사이니지(signage), 차량, 로봇 등으로 구현될 수 있다.26 illustrates an XR device applied to the present invention. The XR device may be implemented as an HMD, a head-up display (HUD) provided in a vehicle, a television, a smart phone, a computer, a wearable device, a home appliance, a digital signage, a vehicle, a robot, and the like.
도 26를 참조하면, XR 기기(100a)는 통신부(110), 제어부(120), 메모리부(130), 입출력부(140a), 센서부(140b) 및 전원공급부(140c)를 포함할 수 있다. 여기서, 블록 110~130/140a~140c은 각각 도 22의 블록 110~130/140에 대응한다.Referring to FIG. 26 , the XR device 100a may include a communication unit 110 , a control unit 120 , a memory unit 130 , an input/output unit 140a , a sensor unit 140b , and a power supply unit 140c . . Here, blocks 110 to 130/140a to 140c correspond to blocks 110 to 130/140 of FIG. 22 , respectively.
통신부(110)는 다른 무선 기기, 휴대 기기, 또는 미디어 서버 등의 외부 기기들과 신호(예, 미디어 데이터, 제어 신호 등)를 송수신할 수 있다. 미디어 데이터는 영상, 이미지, 소리 등을 포함할 수 있다. 제어부(120)는 XR 기기(100a)의 구성 요소들을 제어하여 다양한 동작을 수행할 수 있다. 예를 들어, 제어부(120)는 비디오/이미지 획득, (비디오/이미지) 인코딩, 메타데이터 생성 및 처리 등의 절차를 제어 및/또는 수행하도록 구성될 수 있다. 메모리부(130)는 XR 기기(100a)의 구동/XR 오브젝트의 생성에 필요한 데이터/파라미터/프로그램/코드/명령을 저장할 수 있다. 입출력부(140a)는 외부로부터 제어 정보, 데이터 등을 획득하며, 생성된 XR 오브젝트를 출력할 수 있다. 입출력부(140a)는 카메라, 마이크로폰, 사용자 입력부, 디스플레이부, 스피커 및/또는 햅틱 모듈 등을 포함할 수 있다. 센서부(140b)는 XR 기기 상태, 주변 환경 정보, 사용자 정보 등을 얻을 수 있다. 센서부(140b)는 근접 센서, 조도 센서, 가속도 센서, 자기 센서, 자이로 센서, 관성 센서, RGB 센서, IR 센서, 지문 인식 센서, 초음파 센서, 광 센서, 마이크로폰 및/또는 레이더 등을 포함할 수 있다. 전원공급부(140c)는 XR 기기(100a)에게 전원을 공급하며, 유/무선 충전 회로, 배터리 등을 포함할 수 있다.The communication unit 110 may transmit/receive signals (eg, media data, control signals, etc.) to/from external devices such as other wireless devices, portable devices, or media servers. Media data may include images, images, and sounds. The controller 120 may perform various operations by controlling the components of the XR device 100a. For example, the controller 120 may be configured to control and/or perform procedures such as video/image acquisition, (video/image) encoding, and metadata generation and processing. The memory unit 130 may store data/parameters/programs/codes/commands necessary for driving the XR device 100a/creating an XR object. The input/output unit 140a may obtain control information, data, and the like from the outside, and may output the generated XR object. The input/output unit 140a may include a camera, a microphone, a user input unit, a display unit, a speaker, and/or a haptic module. The sensor unit 140b may obtain an XR device state, surrounding environment information, user information, and the like. The sensor unit 140b may include a proximity sensor, an illuminance sensor, an acceleration sensor, a magnetic sensor, a gyro sensor, an inertial sensor, an RGB sensor, an IR sensor, a fingerprint recognition sensor, an ultrasonic sensor, an optical sensor, a microphone, and/or a radar. there is. The power supply unit 140c supplies power to the XR device 100a, and may include a wired/wireless charging circuit, a battery, and the like.
일 예로, XR 기기(100a)의 메모리부(130)는 XR 오브젝트(예, AR/VR/MR 오브젝트)의 생성에 필요한 정보(예, 데이터 등)를 포함할 수 있다. 입출력부(140a)는 사용자로부터 XR 기기(100a)를 조작하는 명령을 회득할 수 있으며, 제어부(120)는 사용자의 구동 명령에 따라 XR 기기(100a)를 구동시킬 수 있다. 예를 들어, 사용자가 XR 기기(100a)를 통해 영화, 뉴스 등을 시청하려고 하는 경우, 제어부(120)는 통신부(130)를 통해 컨텐츠 요청 정보를 다른 기기(예, 휴대 기기(100b)) 또는 미디어 서버에 전송할 수 있다. 통신부(130)는 다른 기기(예, 휴대 기기(100b)) 또는 미디어 서버로부터 영화, 뉴스 등의 컨텐츠를 메모리부(130)로 다운로드/스트리밍 받을 수 있다. 제어부(120)는 컨텐츠에 대해 비디오/이미지 획득, (비디오/이미지) 인코딩, 메타데이터 생성/처리 등의 절차를 제어 및/또는 수행하며, 입출력부(140a)/센서부(140b)를 통해 획득한 주변 공간 또는 현실 오브젝트에 대한 정보에 기반하여 XR 오브젝트를 생성/출력할 수 있다.For example, the memory unit 130 of the XR device 100a may include information (eg, data, etc.) necessary for generating an XR object (eg, AR/VR/MR object). The input/output unit 140a may obtain a command to operate the XR device 100a from the user, and the controller 120 may drive the XR device 100a according to the user's driving command. For example, when the user intends to watch a movie or news through the XR device 100a, the controller 120 transmits the content request information to another device (eg, the mobile device 100b) through the communication unit 130 or can be sent to the media server. The communication unit 130 may download/stream contents such as movies and news from another device (eg, the portable device 100b) or a media server to the memory unit 130 . The controller 120 controls and/or performs procedures such as video/image acquisition, (video/image) encoding, and metadata generation/processing for the content, and is acquired through the input/output unit 140a/sensor unit 140b It is possible to generate/output an XR object based on information about one surrounding space or a real object.
또한, XR 기기(100a)는 통신부(110)를 통해 휴대 기기(100b)와 무선으로 연결되며, XR 기기(100a)의 동작은 휴대 기기(100b)에 의해 제어될 수 있다. 예를 들어, 휴대 기기(100b)는 XR 기기(100a)에 대한 콘트롤러로 동작할 수 있다. 이를 위해, XR 기기(100a)는 휴대 기기(100b)의 3차원 위치 정보를 획득한 뒤, 휴대 기기(100b)에 대응하는 XR 개체를 생성하여 출력할 수 있다. In addition, the XR device 100a is wirelessly connected to the portable device 100b through the communication unit 110 , and the operation of the XR device 100a may be controlled by the portable device 100b. For example, the portable device 100b may operate as a controller for the XR device 100a. To this end, the XR device 100a may obtain 3D location information of the portable device 100b, and then generate and output an XR object corresponding to the portable device 100b.
도 27는 본 발명에 적용되는 로봇을 예시한다. 로봇은 사용 목적이나 분야에 따라 산업용, 의료용, 가정용, 군사용 등으로 분류될 수 있다.27 illustrates a robot applied to the present invention. Robots can be classified into industrial, medical, home, military, etc. depending on the purpose or field of use.
도 27를 참조하면, 로봇(100)은 통신부(110), 제어부(120), 메모리부(130), 입출력부(140a), 센서부(140b) 및 구동부(140c)를 포함할 수 있다. 여기서, 블록 110~130/140a~140c은 각각 도 22의 블록 110~130/140에 대응한다.Referring to FIG. 27 , the robot 100 may include a communication unit 110 , a control unit 120 , a memory unit 130 , an input/output unit 140a , a sensor unit 140b , and a driving unit 140c . Here, blocks 110 to 130/140a to 140c correspond to blocks 110 to 130/140 of FIG. 22 , respectively.
통신부(110)는 다른 무선 기기, 다른 로봇, 또는 제어 서버 등의 외부 기기들과 신호(예, 구동 정보, 제어 신호 등)를 송수신할 수 있다. 제어부(120)는 로봇(100)의 구성 요소들을 제어하여 다양한 동작을 수행할 수 있다. 메모리부(130)는 로봇(100)의 다양한 기능을 지원하는 데이터/파라미터/프로그램/코드/명령을 저장할 수 있다. 입출력부(140a)는 로봇(100)의 외부로부터 정보를 획득하며, 로봇(100)의 외부로 정보를 출력할 수 있다. 입출력부(140a)는 카메라, 마이크로폰, 사용자 입력부, 디스플레이부, 스피커 및/또는 햅틱 모듈 등을 포함할 수 있다. 센서부(140b)는 로봇(100)의 내부 정보, 주변 환경 정보, 사용자 정보 등을 얻을 수 있다. 센서부(140b)는 근접 센서, 조도 센서, 가속도 센서, 자기 센서, 자이로 센서, 관성 센서, IR 센서, 지문 인식 센서, 초음파 센서, 광 센서, 마이크로폰, 레이더 등을 포함할 수 있다. 구동부(140c)는 로봇 관절을 움직이는 등의 다양한 물리적 동작을 수행할 수 있다. 또한, 구동부(140c)는 로봇(100)을 지상에서 주행하거나 공중에서 비행하게 할 수 있다. 구동부(140c)는 액츄에이터, 모터, 바퀴, 브레이크, 프로펠러 등을 포함할 수 있다.The communication unit 110 may transmit/receive signals (eg, driving information, control signals, etc.) with external devices such as other wireless devices, other robots, or control servers. The controller 120 may perform various operations by controlling the components of the robot 100 . The memory unit 130 may store data/parameters/programs/codes/commands supporting various functions of the robot 100 . The input/output unit 140a may obtain information from the outside of the robot 100 and may output information to the outside of the robot 100 . The input/output unit 140a may include a camera, a microphone, a user input unit, a display unit, a speaker, and/or a haptic module. The sensor unit 140b may obtain internal information, surrounding environment information, user information, and the like of the robot 100 . The sensor unit 140b may include a proximity sensor, an illumination sensor, an acceleration sensor, a magnetic sensor, a gyro sensor, an inertial sensor, an IR sensor, a fingerprint recognition sensor, an ultrasonic sensor, an optical sensor, a microphone, a radar, and the like. The driving unit 140c may perform various physical operations such as moving a robot joint. In addition, the driving unit 140c may make the robot 100 travel on the ground or fly in the air. The driving unit 140c may include an actuator, a motor, a wheel, a brake, a propeller, and the like.
도 28은 본 발명에 적용되는 AI 기기를 예시한다. AI 기기는 TV, 프로젝터, 스마트폰, PC, 노트북, 디지털방송용 단말기, 태블릿 PC, 웨어러블 장치, 셋톱박스(STB), 라디오, 세탁기, 냉장고, 디지털 사이니지, 로봇, 차량 등과 같은, 고정형 기기 또는 이동 가능한 기기 등으로 구현될 수 있다.28 illustrates an AI device applied to the present invention. AI devices include TVs, projectors, smartphones, PCs, laptops, digital broadcasting terminals, tablet PCs, wearable devices, set-top boxes (STBs), radios, washing machines, refrigerators, digital signage, robots, vehicles, etc. It may be implemented in any possible device or the like.
도 28을 참조하면, AI 기기(100)는 통신부(110), 제어부(120), 메모리부(130), 입/출력부(140a/140b), 러닝 프로세서부(140c) 및 센서부(140d)를 포함할 수 있다. 블록 110~130/140a~140d는 각각 도 22의 블록 110~130/140에 대응한다.Referring to FIG. 28 , the AI device 100 includes a communication unit 110 , a control unit 120 , a memory unit 130 , input/output units 140a/140b , a learning processor unit 140c , and a sensor unit 140d). may include. Blocks 110 to 130/140a to 140d correspond to blocks 110 to 130/140 of FIG. 22, respectively.
통신부(110)는 유무선 통신 기술을 이용하여 다른 AI 기기(예, 도 19, 100x, 200, 400)나 AI 서버(200) 등의 외부 기기들과 유무선 신호(예, 센서 정보, 사용자 입력, 학습 모델, 제어 신호 등)를 송수신할 수 있다. 이를 위해, 통신부(110)는 메모리부(130) 내의 정보를 외부 기기로 전송하거나, 외부 기기로부터 수신된 신호를 메모리부(130)로 전달할 수 있다.The communication unit 110 uses wired/wireless communication technology to communicate with other AI devices (eg, FIGS. 19, 100x, 200, 400) or external devices such as the AI server 200 and wired/wireless signals (eg, sensor information, user input, learning). models, control signals, etc.). To this end, the communication unit 110 may transmit information in the memory unit 130 to an external device or transmit a signal received from the external device to the memory unit 130 .
제어부(120)는 데이터 분석 알고리즘 또는 머신 러닝 알고리즘을 사용하여 결정되거나 생성된 정보에 기초하여, AI 기기(100)의 적어도 하나의 실행 가능한 동작을 결정할 수 있다. 그리고, 제어부(120)는 AI 기기(100)의 구성 요소들을 제어하여 결정된 동작을 수행할 수 있다. 예를 들어, 제어부(120)는 러닝 프로세서부(140c) 또는 메모리부(130)의 데이터를 요청, 검색, 수신 또는 활용할 수 있고, 적어도 하나의 실행 가능한 동작 중 예측되는 동작이나, 바람직한 것으로 판단되는 동작을 실행하도록 AI 기기(100)의 구성 요소들을 제어할 수 있다. 또한, 제어부(120)는 AI 장치(100)의 동작 내용이나 동작에 대한 사용자의 피드백 등을 포함하는 이력 정보를 수집하여 메모리부(130) 또는 러닝 프로세서부(140c)에 저장하거나, AI 서버(도 19, 400) 등의 외부 장치에 전송할 수 있다. 수집된 이력 정보는 학습 모델을 갱신하는데 이용될 수 있다.The controller 120 may determine at least one executable operation of the AI device 100 based on information determined or generated using a data analysis algorithm or a machine learning algorithm. In addition, the controller 120 may control the components of the AI device 100 to perform the determined operation. For example, the control unit 120 may request, search, receive, or utilize the data of the learning processor unit 140c or the memory unit 130 , and may be predicted or preferred among at least one executable operation. Components of the AI device 100 may be controlled to execute the operation. In addition, the control unit 120 collects history information including user feedback on the operation contents or operation of the AI device 100 and stores it in the memory unit 130 or the learning processor unit 140c, or the AI server ( 19 and 400) may be transmitted to an external device. The collected historical information may be used to update the learning model.
메모리부(130)는 AI 기기(100)의 다양한 기능을 지원하는 데이터를 저장할 수 있다. 예를 들어, 메모리부(130)는 입력부(140a)로부터 얻은 데이터, 통신부(110)로부터 얻은 데이터, 러닝 프로세서부(140c)의 출력 데이터, 및 센싱부(140)로부터 얻은 데이터를 저장할 수 있다. 또한, 메모리부(130)는 제어부(120)의 동작/실행에 필요한 제어 정보 및/또는 소프트웨어 코드를 저장할 수 있다.The memory unit 130 may store data supporting various functions of the AI device 100 . For example, the memory unit 130 may store data obtained from the input unit 140a , data obtained from the communication unit 110 , output data of the learning processor unit 140c , and data obtained from the sensing unit 140 . Also, the memory unit 130 may store control information and/or software codes necessary for the operation/execution of the control unit 120 .
입력부(140a)는 AI 기기(100)의 외부로부터 다양한 종류의 데이터를 획득할 수 있다. 예를 들어, 입력부(120)는 모델 학습을 위한 학습 데이터, 및 학습 모델이 적용될 입력 데이터 등을 획득할 수 있다. 입력부(140a)는 카메라, 마이크로폰 및/또는 사용자 입력부 등을 포함할 수 있다. 출력부(140b)는 시각, 청각 또는 촉각 등과 관련된 출력을 발생시킬 수 있다. 출력부(140b)는 디스플레이부, 스피커 및/또는 햅틱 모듈 등을 포함할 수 있다. 센싱부(140)는 다양한 센서들을 이용하여 AI 기기(100)의 내부 정보, AI 기기(100)의 주변 환경 정보 및 사용자 정보 중 적어도 하나를 얻을 수 있다. 센싱부(140)는 근접 센서, 조도 센서, 가속도 센서, 자기 센서, 자이로 센서, 관성 센서, RGB 센서, IR 센서, 지문 인식 센서, 초음파 센서, 광 센서, 마이크로폰 및/또는 레이더 등을 포함할 수 있다.The input unit 140a may acquire various types of data from the outside of the AI device 100 . For example, the input unit 120 may obtain training data for model learning, input data to which the learning model is applied, and the like. The input unit 140a may include a camera, a microphone, and/or a user input unit. The output unit 140b may generate an output related to sight, hearing, or touch. The output unit 140b may include a display unit, a speaker, and/or a haptic module. The sensing unit 140 may obtain at least one of internal information of the AI device 100 , surrounding environment information of the AI device 100 , and user information by using various sensors. The sensing unit 140 may include a proximity sensor, an illuminance sensor, an acceleration sensor, a magnetic sensor, a gyro sensor, an inertial sensor, an RGB sensor, an IR sensor, a fingerprint recognition sensor, an ultrasonic sensor, an optical sensor, a microphone, and/or a radar. there is.
러닝 프로세서부(140c)는 학습 데이터를 이용하여 인공 신경망으로 구성된 모델을 학습시킬 수 있다. 러닝 프로세서부(140c)는 AI 서버(도 19, 400)의 러닝 프로세서부와 함께 AI 프로세싱을 수행할 수 있다. 러닝 프로세서부(140c)는 통신부(110)를 통해 외부 기기로부터 수신된 정보, 및/또는 메모리부(130)에 저장된 정보를 처리할 수 있다. 또한, 러닝 프로세서부(140c)의 출력 값은 통신부(110)를 통해 외부 기기로 전송되거나/되고, 메모리부(130)에 저장될 수 있다.The learning processor unit 140c may train a model composed of an artificial neural network by using the training data. The learning processor unit 140c may perform AI processing together with the learning processor unit of the AI server ( FIGS. 19 and 400 ). The learning processor unit 140c may process information received from an external device through the communication unit 110 and/or information stored in the memory unit 130 . In addition, the output value of the learning processor unit 140c may be transmitted to an external device through the communication unit 110 and/or stored in the memory unit 130 .
여기서, 본 명세서의 무선 기기(100, 200)에서 구현되는 무선 통신 기술은 LTE, NR 및 6G뿐만 아니라 저전력 통신을 위한 Narrowband Internet of Things를 포함할 수 있다. 이때, 예를 들어 NB-IoT 기술은 LPWAN(Low Power Wide Area Network) 기술의 일례일 수 있고, LTE Cat NB1 및/또는 LTE Cat NB2 등의 규격으로 구현될 수 있으며, 상술한 명칭에 한정되는 것은 아니다. 추가적으로 또는 대체적으로, 본 명세서의 무선 기기(100, 200)에서 구현되는 무선 통신 기술은 LTE-M 기술을 기반으로 통신을 수행할 수 있다. 이때, 일 예로, LTE-M 기술은 LPWAN 기술의 일례일 수 있고, eMTC(enhanced Machine Type Communication) 등의 다양한 명칭으로 불릴 수 있다. 예를 들어, LTE-M 기술은 1) LTE CAT 0, 2) LTE Cat M1, 3) LTE Cat M2, 4) LTE non-BL(non-Bandwidth Limited), 5) LTE-MTC, 6) LTE Machine Type Communication, 및/또는 7) LTE M 등의 다양한 규격 중 적어도 어느 하나로 구현될 수 있으며 상술한 명칭에 한정되는 것은 아니다. 추가적으로 또는 대체적으로, 본 명세서의 무선 기기(100, 200)에서 구현되는 무선 통신 기술은 저전력 통신을 고려한 지그비(ZigBee), 블루투스(Bluetooth) 및 저전력 광역 통신망(Low Power Wide Area Network, LPWAN) 중 적어도 어느 하나를 포함할 수 있으며, 상술한 명칭에 한정되는 것은 아니다. 일 예로 ZigBee 기술은 IEEE 802.15.4 등의 다양한 규격을 기반으로 소형/저-파워 디지털 통신에 관련된 PAN(personal area networks)을 생성할 수 있으며, 다양한 명칭으로 불릴 수 있다.Here, the wireless communication technology implemented in the wireless devices 100 and 200 of the present specification may include a narrowband Internet of Things for low-power communication as well as LTE, NR, and 6G. At this time, for example, NB-IoT technology may be an example of LPWAN (Low Power Wide Area Network) technology, and may be implemented in standards such as LTE Cat NB1 and/or LTE Cat NB2, and is limited to the above-mentioned names. not. Additionally or alternatively, the wireless communication technology implemented in the wireless devices 100 and 200 of the present specification may perform communication based on the LTE-M technology. In this case, as an example, the LTE-M technology may be an example of an LPWAN technology, and may be called various names such as enhanced machine type communication (eMTC). For example, LTE-M technology is 1) LTE CAT 0, 2) LTE Cat M1, 3) LTE Cat M2, 4) LTE non-BL (non-Bandwidth Limited), 5) LTE-MTC, 6) LTE Machine Type Communication, and/or 7) may be implemented in at least one of various standards such as LTE M, and is not limited to the above-described name. Additionally or alternatively, the wireless communication technology implemented in the wireless devices 100 and 200 of the present specification is at least one of ZigBee, Bluetooth, and Low Power Wide Area Network (LPWAN) in consideration of low-power communication. It may include any one, and is not limited to the above-mentioned names. For example, the ZigBee technology can create PAN (personal area networks) related to small/low-power digital communication based on various standards such as IEEE 802.15.4, and can be called by various names.
전술한 본 명세서는, 프로그램이 기록된 매체에 컴퓨터가 읽을 수 있는 코드로서 구현하는 것이 가능하다. 컴퓨터가 읽을 수 있는 매체는, 컴퓨터 시스템에 의하여 읽혀질 수 있는 데이터가 저장되는 모든 종류의 기록장치를 포함한다. 컴퓨터가 읽을 수 있는 매체의 예로는, HDD(Hard Disk Drive), SSD(Solid State Disk), SDD(Silicon Disk Drive), ROM, RAM, CD-ROM, 자기 테이프, 플로피 디스크, 광 데이터 저장 장치 등이 있으며, 또한 캐리어 웨이브(예를 들어, 인터넷을 통한 전송)의 형태로 구현되는 것도 포함한다. 따라서, 상기의 상세한 설명은 모든 면에서 제한적으로 해석되어서는 아니되고 예시적인 것으로 고려되어야 한다. 본 명세서의 범위는 첨부된 청구항의 합리적 해석에 의해 결정되어야 하고, 본 명세서의 등가적 범위 내에서의 모든 변경은 본 명세서의 범위에 포함된다.The above-described specification can be implemented as computer-readable code on a medium in which a program is recorded. The computer-readable medium includes all types of recording devices in which data readable by a computer system is stored. Examples of computer-readable media include Hard Disk Drive (HDD), Solid State Disk (SSD), Silicon Disk Drive (SDD), ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc. There is also a carrier wave (eg, transmission over the Internet) that is implemented in the form of. Accordingly, the above detailed description should not be construed as restrictive in all respects but as exemplary. The scope of this specification should be determined by a reasonable interpretation of the appended claims, and all modifications within the scope of equivalents of this specification are included in the scope of this specification.

Claims (13)

  1. 복수의 디바이스들의 학습 결과에 기초하여, 서버가 최종 학습 결과물을 도출하는 연합학습을 위한 통신방법에 있어서,In a communication method for federated learning in which a server derives a final learning result based on the learning result of a plurality of devices,
    상기 디바이스들이 상기 서버로부터 글로벌 모델 및 가중치 변화량에 기초한 스케일링 계수를 제공받는 단계;receiving, by the devices, a scaling factor based on a global model and a weight change amount from the server;
    상기 다바이스들이 상기 글로벌 모델에 기초하여 상기 가중치 변화량을 산출하는 단계;calculating, by the devices, the weight change amount based on the global model;
    상기 디바이스들이 상기 가중치 변화량에 기초하여 양자화를 수행하는 단계; 및performing, by the devices, quantization based on the weight change amount; and
    상기 디바이스들이 양자화 된 가중치 변화량을 서버로 전송하는 단계;를 포함하는 연합학습을 위한 통신방법.A communication method for federated learning comprising a; the devices transmitting the quantized weight change amount to the server.
  2. 제 1 항에 있어서,The method of claim 1,
    상기 양자화를 수행하는 단계는The step of performing the quantization is
    상기 디바이스들 각각이 동일한 값을 갖는 상기 스케일링 계수를 이용하는 것을 특징으로 하는 연합학습을 위한 통신방법.A communication method for joint learning, characterized in that each of the devices uses the scaling factor having the same value.
  3. 제 1 항에 있어서,The method of claim 1,
    상기 디바이스들이 상기 스케일링 계수를 제공받는 단계는,The step in which the devices are provided with the scaling factor comprises:
    (i-1)(i는 2 이상의 자연수) 번째 라운드에서 생성한 가중치 변화량 분포에 기초하여 생성된 상기 스케일링 계수를, i 번째 라운드에서 제공받는 것을 특징으로 하는 연합학습을 위한 통신방법.(i-1) (i is a natural number greater than or equal to 2) A communication method for joint learning, characterized in that the scaling factor generated based on the weight change distribution generated in the th round is provided in the i th round.
  4. 제 3 항에 있어서,4. The method of claim 3,
    상기 (i-1)번째 라운드에서 상기 가중치 변화량 분포를 생성하는 단계는,The step of generating the weight change amount distribution in the (i-1)-th round comprises:
    상기 (i-1)번째 라운드 까지의 라운드 중에서 적어도 하나 이상의 라운드의 상기 가중치 변화량의 절대값을 누적한 누적 분포 함수를 생성하는 단계; 및generating a cumulative distribution function accumulating absolute values of the weight change amount of at least one round among rounds up to the (i-1)-th round; and
    상기 누적 분포 함수에서 미리 설정된 경계값 이상의 상기 가중치 변화량의 절대값의 누적 분포는 클립핑하는 단계를 포함하는 것을 특징으로 하는 연합학습을 위한 통신방법.and clipping the cumulative distribution of the absolute value of the weight change amount greater than or equal to a preset boundary value in the cumulative distribution function.
  5. 제 4 항에 있어서,5. The method of claim 4,
    상기 스케일링 계수를 결정하는 단계는The step of determining the scaling factor is
    상기 (i-1) 번째 라운드의 손실과 상기 (i-2) 번째 라운드의 손실의 차이값을 계산한 것에 기초하여 손실 변화량을 산출하는 단계; 및calculating a change in loss based on a difference between the loss in the (i-1)-th round and the loss in the (i-2)-th round; and
    상기 손실 변화량의 크기에 기초하여 스케일링 계수의 크기를 결정하는 단계를 포함하는 것을 특징으로 하는 연합학습을 위한 통신방법.and determining a size of a scaling factor based on the magnitude of the change in loss.
  6. 제 5 항에 있어서,6. The method of claim 5,
    상기 스케일링 계수를 결정하는 단계는The step of determining the scaling factor is
    미리 설정된 최대 양자화 비트에 대비한 상기 경계값의 크기 보다 작은 범위 내에서 결정하는 것을 특징으로 하는 연합학습을 위한 통신방법.A communication method for federated learning, characterized in that the determination is made within a range smaller than the size of the boundary value with respect to a preset maximum quantization bit.
  7. 제 6 항에 있어서,7. The method of claim 6,
    상기 스케일링 계수를 결정하는 단계는The step of determining the scaling factor is
    상기 경계값의 크기 보다 작은 범위를 둘 이상의 구간으로 구분하고, 구분된 구간에서 각각 서로 다른 스케일링 계수들을 생성하는 것을 특징으로 하는 연합학습을 위한 통신방법.A communication method for joint learning, characterized in that a range smaller than the size of the boundary value is divided into two or more sections, and different scaling coefficients are generated in the divided sections.
  8. 제 1 항에 있어서,The method of claim 1,
    상기 양자화를 수행하는 단계는The step of performing the quantization is
    상기 가중치 변화량을 상기 스케일링 계수 및 가변 가중치 변화량의 곱으로 산출하는 단계; 및calculating the weight change amount as a product of the scaling factor and the variable weight change amount; and
    상기 가변 가중치 변화량의 양자화 범위를 산출하는 단계;를 포함하는 것을 특징으로 하는 연합학습을 위한 통신방법.Calculating a quantization range of the variable weight change amount; a communication method for federated learning comprising: a.
  9. 제 8 항에 있어서,9. The method of claim 8,
    상기 양자화 된 가중치 변화량을 서버로 전송하는 단계는The step of transmitting the quantized weight change amount to the server is
    상기 스케일링 계수에 기초하여 가변된 가변 양자화 정보를 전송하는 단계를 더 포함하는 것을 특징으로 하는 연합학습을 위한 통신방법.Communication method for joint learning, characterized in that it further comprises the step of transmitting the variable quantization information changed based on the scaling coefficient.
  10. 서버로부터 제공받은 글로벌 모델에 기초하여 연합학습을 수행하는 디바이스에 있어서,In a device for performing federated learning based on a global model provided from a server,
    상기 서버와의 통신을 위한 트랜시버; 및a transceiver for communication with the server; and
    상기 글로벌 모델에 기초하여 연합학습을 수행하는 프로세서를 포함하고,A processor for performing federated learning based on the global model,
    상기 프로세서는 the processor is
    상기 서버로부터 글로벌 모델 및 가중치 변화량에 기초한 스케일링 계수를 제공받고, 상기 글로벌 모델에 기초하여 상기 가중치 변화량을 산출하며, 상기 가중치 변화량에 기초하여 양자화를 수행하고, 양자화 된 가중치 변화량을 서버로 전송하는 포함하는 연합학습을 수행하는 디바이스.Receiving a global model and a scaling factor based on a weight change amount from the server, calculating the weight change amount based on the global model, performing quantization based on the weight change amount, and transmitting the quantized weight change amount to the server A device that performs federated learning.
  11. 제 10 항에 있어서,11. The method of claim 10,
    상기 프로세서는 (i-1)(i는 2 이상의 자연수) 번째 라운드에서 생성한 상기 가중치 변화량 분포에 기초하여 생성된 상기 스케일링 계수를, i번째 라운드에서 제공받는 것을 특징으로 하는 연합학습을 수행하는 디바이스.wherein the processor receives, in the i-th round, the scaling factor generated based on the weight change distribution generated in the (i-1) (i is a natural number equal to or greater than 2)-th round .
  12. 제 11 항에 있어서,12. The method of claim 11,
    상기 (i-1)번째 라운드에서 생성된 상기 가중치 변화량 분포는The weight change amount distribution generated in the (i-1)-th round is
    상기 (i-1)번째 라운드 이전까지의 라운드 중에서 적어도 하나 이상의 라운드의 상기 가중치 변화량의 절대값을 누적한 누적 분포 함수에서, 미리 설정된 경계값 이상의 상기 가중치 변화량 절대값의 누적 분포를 클립핑하여 생성되는 것을 특징으로 하는 연합학습을 수행하는 디바이스.Generated by clipping the cumulative distribution of the absolute value of the weight change greater than or equal to a preset threshold in the cumulative distribution function accumulating the absolute value of the weight change amount of at least one round among the rounds before the (i-1)th round A device for performing federated learning, characterized in that.
  13. 제 12 항에 있어서,13. The method of claim 12,
    상기 스케일링 계수는The scaling factor is
    상기 (i-1) 번째 라운드의 손실과 상기 (i-2) 번째 라운드의 손실의 차이값을 계산한 것에 기초하여 크기가 결정되는 것을 특징으로 하는 연합학습을 수행하는 디바이스.A device for performing joint learning, characterized in that the size is determined based on a difference between the loss of the (i-1)-th round and the loss of the (i-2)-th round.
PCT/KR2020/011878 2020-09-03 2020-09-03 Communication method for federated learning and device for performing same WO2022050444A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/KR2020/011878 WO2022050444A1 (en) 2020-09-03 2020-09-03 Communication method for federated learning and device for performing same
KR1020237007045A KR20230060505A (en) 2020-09-03 2020-09-03 Communication method for federated learning and device performing the same

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/KR2020/011878 WO2022050444A1 (en) 2020-09-03 2020-09-03 Communication method for federated learning and device for performing same

Publications (1)

Publication Number Publication Date
WO2022050444A1 true WO2022050444A1 (en) 2022-03-10

Family

ID=80491172

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2020/011878 WO2022050444A1 (en) 2020-09-03 2020-09-03 Communication method for federated learning and device for performing same

Country Status (2)

Country Link
KR (1) KR20230060505A (en)
WO (1) WO2022050444A1 (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR102644669B1 (en) * 2023-08-14 2024-03-06 재단법인 서울디지털재단 Federated learning method and system for enhanced learning converges speed

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190378016A1 (en) * 2018-06-07 2019-12-12 International Business Machines Corporation Distributed computing architecture for large model deep learning

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190378016A1 (en) * 2018-06-07 2019-12-12 International Business Machines Corporation Distributed computing architecture for large model deep learning

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
AMIRHOSSEIN REISIZADEH , ARYAN MOKHTARI, HAMED HASSANI , ALI JADBABAIE , RAMTIN PEDARSANI: "FedPAQ: A Communication-Efficient Federated Learning Method with Periodic Averaging and Quantization", PROCEEDINGS OF THE 23RDINTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS (AISTATS) 2020, vol. 108, 28 August 2020 (2020-08-28), Palermo, Italy, pages 1 - 10, XP055907840 *
LIN YUJUN, HAN SONG, MAO HUIZI, WANG YU, DALLY WILLIAM J: "Deep Gradient Compression: Reducing the Communication Bandwidth for Distributed Training", ARXIV.ORG, 5 December 2017 (2017-12-05), pages 1 - 13, XP055907853 *
XU HANG, HO CHEN-YU, ABDELMONIEM AHMED M, DUTTA ARITRA, BERGOU EL HOUCINE, KARATSENIDIS KONSTANTINOS, CANINI MARCO, KALNIS PANOS: "Compressed Communication for Distributed Deep Learning: Survey and Quantitative Evaluation", TECHNICAL REPORT FROM KING ABDULLAH UNIVERSITY OF SCIENCE AND TECHNOLOGY., 16 July 2020 (2020-07-16), pages 1 - 16, XP055907852 *
YONI CHOUKROUN; ELI KRAVCHIK; FAN YANG; PAVEL KISILEV: "Low-bit Quantization of Neural Networks for Efficient Inference", ARXIV.ORG, 18 February 2019 (2019-02-18), pages 1 - 10, XP081121348 *

Also Published As

Publication number Publication date
KR20230060505A (en) 2023-05-04

Similar Documents

Publication Publication Date Title
WO2021112360A1 (en) Method and device for estimating channel in wireless communication system
WO2022050432A1 (en) Method and device for performing federated learning in wireless communication system
WO2021256584A1 (en) Method for transmitting or receiving data in wireless communication system and apparatus therefor
WO2022050468A1 (en) Method for performing federated learning in wireless communication system, and apparatus therefor
WO2022039295A1 (en) Method for preprocessing downlink in wireless communication system and apparatus therefor
WO2022045399A1 (en) Federated learning method based on selective weight transmission and terminal therefor
WO2022054981A1 (en) Method and device for executing compression federated learning
WO2022014732A1 (en) Method and apparatus for performing federated learning in wireless communication system
WO2022025321A1 (en) Signal randomization method and device of communication apparatus
WO2022054985A1 (en) Method and apparatus for transmitting and receiving signal by user equipment and base station in wireless communication system
WO2022059808A1 (en) Method for performing reinforcement learning by v2x communication device in autonomous driving system
WO2021251511A1 (en) Method for transmitting/receiving uplink signal of high frequency band in wireless communication system, and device therefor
WO2022050444A1 (en) Communication method for federated learning and device for performing same
WO2022030664A1 (en) Communication method based on similarity of inter-frequency band spatial information for channel in wireless communication system and apparatus therefor
WO2022039302A1 (en) Method for controlling calculations of deep neural network in wireless communication system, and apparatus therefor
WO2022025303A1 (en) Method for performing joint reasoning in wireless communication system, and apparatus therefor
WO2022004927A1 (en) Method for transmitting or receiving signal in wireless communication system using auto encoder, and apparatus therefor
WO2022119021A1 (en) Method and device for adapting learning class-based system to ai mimo
WO2022050434A1 (en) Method and apparatus for performing handover in wireless communication system
WO2022045377A1 (en) Method by which terminal and base station transmit/receive signals in wireless communication system, and apparatus
WO2022014735A1 (en) Method and device for terminal and base station to transmit and receive signals in wireless communication system
WO2022054980A1 (en) Encoding method and neural network encoder structure usable in wireless communication system
WO2022050440A1 (en) Communication method and server for distributed learning by which server derives final learning results on basis of learning results of plurality of devices
WO2022050441A1 (en) Wireless signal digital conversion device
WO2022050443A1 (en) Communication method and communication system for reducing overhead of reference signal

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: 20952543

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20952543

Country of ref document: EP

Kind code of ref document: A1