Configurable Neural Networks for Channel State Feedback (CSF) Learning

Abstract

A method for wireless communication by a user equipment (UE) includes receiving a plurality of neural network training configurations for channel state feedback (CSF). Each configuration corresponds to a different neural network framework. The method also includes training each of a population of neural network decoder / encoder pairs according to the received training configurations. A method for wireless communication by a base station includes transmitting a plurality of neural network training configurations for channel state feedback (CSF) to a user equipment (UE). Each configuration corresponds to a different neural network framework. The method also includes receiving neural network decoder / encoder pairs trained according to the training configurations.

Description

[0001] public domain

[0002] Aspects of the present disclosure relate generally to wireless communications, and more particularly to techniques and apparatus for configurable 5G New Radio (NR) channel state feedback (CSF) learning.

[0003] background

[0004] Wireless communication systems are widely deployed to provide a variety of telecommunication services, such as telephony, video, data, messaging, and broadcasts. Typical wireless communication systems may employ multiple access technologies capable of supporting communication with multiple users by sharing available system resources (e.g., bandwidth, transmit power, etc.). Examples of such multiple access technologies include code division multiple access (CDMA) systems, time division multiple access (TDMA) systems, frequency division multiple access (FDMA) systems, orthogonal frequency division multiple access (OFDMA) systems, single-carrier frequency division multiple access (SC-FDMA) systems, time division synchronous code division multiple access (TD-SCDMA) systems, and long term evolution (LTE). LTE / LTE-Advanced is a set of enhancements to the Universal Mobile Telecommunications System (UMTS) mobile standard promulgated by the Third Generation Partnership Project (3GPP).

[0005] A wireless communication network may include several base stations (BSs) that can support communications for several user equipment (UEs). User equipment (UEs) can communicate with the base stations (BSs) via downlinks and uplinks. The downlink (or forward link) refers to the communication link from the BS to the UE, while the uplink (or reverse link) refers to the communication link from the UE to the BS. As will be described in more detail herein, a BS may be referred to as a Node B, gNB, access point (AP), radio head, transmit / receive point (TRP), new radio (NR) BS, 5G Node B, and so on.

[0006] The above multiple access technologies have been adopted in various telecommunication standards to provide a common protocol that enables different user equipment to communicate at the city, country, region, and even global levels. New Radio (NR) (which may also be referred to as 5G) is a set of enhancements to the LTE mobile standard promulgated by the Third Generation Partnership Project (3GPP). NR is designed to better support mobile broadband Internet access by using orthogonal frequency division multiplexing (OFDM) (CP-OFDM) with a cyclic prefix (CP) on the downlink (DL), using CP-OFDM and / or SC-FDM (e.g., also known as discrete Fourier transform spread OFDM (DFT-s-OFDM)) on the uplink (UL), and supporting beamforming, multiple-input multiple-output (MIMO) antenna technology, and carrier aggregation to improve spectrum efficiency, reduce costs, improve services, utilize new spectrum, and better integrate with other open standards. However, as the demand for mobile broadband access continues to grow, there is a need for further improvements to NR and LTE technologies. Preferably, these improvements should be applicable to other multiple access technologies and telecommunication standards that adopt these technologies.

[0007] An artificial neural network may include a group of interconnected artificial neurons (e.g., a neuron model). An artificial neural network may be a computing device or represented as a method executed by a computing device. A convolutional neural network (such as a deep convolutional neural network) is a feedforward artificial neural network. A convolutional neural network may include layers of neurons that may be configured in a tiled receptive field. It would be desirable to apply neural network processing to wireless communications to achieve higher efficiency.

[0008] Overview

[0009] In one aspect of the present disclosure, a method for wireless communication by a user equipment (UE) includes receiving a plurality of neural network training configurations for channel state feedback (CSF). Each configuration corresponds to a different neural network framework. The method also includes training each of a population of neural network decoder / encoder pairs according to the received training configurations.

[0010] In another aspect of the present disclosure, a method for wireless communication by a base station includes transmitting a plurality of neural network training configurations for channel state feedback (CSF) to a user equipment (UE). Each configuration corresponds to a different neural network framework. The method also includes receiving a neural network decoder / encoder pair trained according to the training configurations.

[0011] In another aspect of the present disclosure, a UE includes a memory and at least one processor operatively coupled to the memory. The memory and the at least one processor are configured to receive a plurality of neural network training configurations for channel state feedback (CSF). Each configuration corresponds to a different neural network framework. The UE is further configured to train each of a set of neural network decoder / encoder pairs according to the received training configurations.

[0012] In another aspect of the present disclosure, a base station includes a memory and at least one processor operatively coupled to the memory. The memory and the at least one processor are configured to transmit a plurality of neural network training configurations for channel state feedback (CSF) to a user equipment (UE). Each configuration corresponds to a different neural network framework. The base station is further configured to receive neural network decoder / encoder pairs trained according to the training configurations.

[0013] In another aspect of the present disclosure, a UE includes means for receiving a plurality of neural network training configurations for channel state feedback (CSF). Each configuration corresponds to a different neural network framework. The UE also includes means for training each of a set of neural network decoder / encoder pairs according to the received training configurations.

[0014] In another aspect of the present disclosure, a base station includes means for transmitting a plurality of neural network training configurations for channel state feedback (CSF) to a user equipment (UE). Each configuration corresponds to a different neural network framework. The base station also includes means for receiving neural network decoder / encoder pairs trained according to the training configurations.

[0015] In another aspect of the present disclosure, a non-transitory computer-readable medium having program code recorded thereon is disclosed. The program code is executed by a UE processor and includes program code for receiving a plurality of neural network training configurations for channel state feedback (CSF). Each configuration corresponds to a different neural network framework. The program code also includes program code for training each of a group of neural network decoder / encoder pairs according to the received training configurations.

[0016] In another aspect of the present disclosure, a non-transitory computer-readable medium having program code recorded thereon is disclosed. The program code is executed by a base station processor and includes program code for transmitting a plurality of neural network training configurations for channel state feedback (CSF) to a user equipment (UE). Each configuration corresponds to a different neural network framework. The program code also includes program code for receiving neural network decoder / encoder pairs trained according to the training configurations.

[0017] Aspects generally include methods, apparatus (devices), systems, computer program products, non-transitory computer-readable media, user equipment, base stations, wireless communication devices, and processing systems as substantially described herein with reference to and as illustrated in the accompanying figures and description.

[0018] The foregoing has broadly outlined the features and technical advantages of examples according to the present disclosure in an effort to make the following detailed description better understood. Additional features and advantages will be described hereinafter. The concepts and specific examples disclosed can be readily used as a basis for modifying or designing other structures for implementing the same purposes as the present disclosure. Such equivalent constructions do not depart from the scope of the appended claims. The characteristics of the concepts disclosed herein, both in terms of their organization and method of operation, as well as the associated advantages, will be better understood by considering the following description in conjunction with the accompanying drawings. Each of the figures is provided for illustration and description purposes and is not intended to define limitations on the claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to understand in detail the features of the present disclosure set forth above, a more particular description of the content briefly summarized above may be obtained with reference to various aspects, some of which are illustrated in the accompanying drawings. It should be noted, however, that the accompanying drawings illustrate only certain typical aspects of the present disclosure and are not to be considered limiting of its scope, as the description may admit to other equally effective aspects. The same reference numerals in different drawings may identify the same or similar elements.

[0021] Figure 1 is a block diagram conceptually illustrating an example of a wireless communication network in accordance with various aspects of the present disclosure.

[0022] Figure 2 is a block diagram conceptually illustrating an example of a base station and a user equipment (UE) in communication in a wireless communication network according to various aspects of the present disclosure.

[0023] Figure 3 An example implementation of designing a neural network using a system on a chip (SOC) including a general-purpose processor in accordance with certain aspects of the present disclosure is illustrated.

[0024] Figure 4A 、 4B 4C are diagrams illustrating neural networks according to aspects of the present disclosure.

[0025] Figure 4D is a diagram illustrating an exemplary deep convolutional network (DCN) according to aspects of the present disclosure.

[0026] Figure 5 is a block diagram illustrating an exemplary deep convolutional network (DCN) according to aspects of the present disclosure.

[0027] Figure 6is a block diagram illustrating an exemplary autoencoder according to aspects of the present disclosure.

[0028] Figure 7 is a diagram illustrating example processes performed, for example, by a user equipment (UE) in accordance with various aspects of the present disclosure.

[0029] Figure 8 is a diagram illustrating example processes performed, for example, by a base station, according to various aspects of the present disclosure.

[0030] Detailed description

[0031] The various aspects of the present disclosure are described more fully below with reference to the accompanying drawings. However, the present disclosure can be implemented in many different forms and should not be interpreted as being limited to any specific structure or function given throughout the present disclosure. On the contrary, these aspects are provided to make the present disclosure thorough and complete, and they will fully convey the scope of the present disclosure to those skilled in the art. Based on the teachings herein, those skilled in the art will appreciate that the scope of the present disclosure is intended to cover any aspect of the present disclosure disclosed herein, whether it is implemented independently of any other aspect of the present disclosure or implemented in combination. For example, any number of aspects described can be used to implement a device or practice method. In addition, the scope of the present disclosure is intended to cover such devices or methods that are practiced using a supplement to the various aspects of the present disclosure described or other other structures, functionality, or structure and functionality. It should be understood that any aspect of the disclosed disclosure can be implemented by one or more elements of the claims.

[0032] Several aspects of telecommunications systems will now be presented with reference to various devices and techniques. These devices and techniques are described in the following detailed description and illustrated in the accompanying drawings by various blocks, modules, components, circuits, steps, processes, algorithms, etc. (collectively referred to as "elements"). These elements can be implemented using hardware, software, or a combination thereof. Whether such elements are implemented as hardware or software depends on the specific application and the design constraints imposed on the overall system.

[0033] It should be noted that while various aspects may be described using terminology generally associated with 5G and later generation wireless technologies, various aspects of the present disclosure may be employed in communication systems based on other generations, such as and including 3G and / or 4G technologies.

[0034] Figure 11 is a diagram illustrating a network 100 in which various aspects of the present disclosure may be practiced. The network 100 may be a 5G or NR network or some other wireless network, such as an LTE network. The wireless network 100 may include several BSs 110 (shown as BS 110a, BS 110b, BS 110c, and BS 110d) and other network entities. A BS is an entity that communicates with user equipment (UE) and may also be referred to as a base station, NR BS, Node B, gNB, 5G Node B (NB), access point, transmit reception point (TRP), etc. Each BS may provide communication coverage for a particular geographic area. In 3GPP, the term "cell" may refer to the coverage area of a BS and / or a BS subsystem serving that coverage area, depending on the context in which the term is used.

[0035] A BS may provide communication coverage for a macro cell, a pico cell, a femto cell, and / or another type of cell. A macro cell may cover a relatively large geographic area (e.g., several kilometers in radius) and may allow unrestricted access by UEs with service subscriptions. A pico cell may cover a relatively small geographic area and may allow unrestricted access by UEs with service subscriptions. A femto cell may cover a relatively small geographic area (e.g., a residence) and may allow restricted access by UEs associated with the femto cell (e.g., UEs in a closed subscriber group (CSG)). A BS for a macro cell may be referred to as a macro BS. A BS for a pico cell may be referred to as a pico BS. A BS for a femto cell may be referred to as a femto BS or a home BS. In Figure 1 In the example shown in FIG, BS 110a may be a macro BS for macro cell 102a, BS 110b may be a pico BS for pico cell 102b, and BS 110c may be a femto BS for femto cell 102c. A BS may support one or more (e.g., three) cells. The terms "eNB," "base station," "NR BS," "gNB," "TRP," "AP," "Node B," "5G NB," and "cell" may be used interchangeably herein.

[0036] In some aspects, the cells may not necessarily be stationary, and the geographic area of the cells may move depending on the location of the mobile BS. In some aspects, the BSs may interconnect with each other and / or with one or more other BSs or network nodes (not shown) in the wireless network 100 via various types of backhaul interfaces (such as direct physical connections, virtual networks, etc.) using any suitable transport network.

[0037] The wireless network 100 may also include a relay station. A relay station is an entity that can receive transmissions of data from an upstream station (e.g., a BS or a UE) and send transmissions of the data to a downstream station (e.g., a UE or a BS). A relay station may also be a UE that can relay transmissions for other UEs. Figure 1 In the example shown in , relay station 110d may communicate with macro BS 110a and UE 120d to facilitate communication between BS 110a and UE 120d. A relay station may also be referred to as a relay BS, relay base station, relay, or the like.

[0038] The wireless network 100 may be a heterogeneous network including different types of BSs (e.g., macro BSs, pico BSs, femto BSs, relay BSs, etc.). These different types of BSs may have different transmit power levels, different coverage areas, and different impacts on interference in the wireless network 100. For example, a macro BS may have a high transmit power level (e.g., 5 to 40 watts), while a pico BS, a femto BS, and a relay BS may have a lower transmit power level (e.g., 0.1 to 2 watts).

[0039] The network controller 130 may be coupled to a set of BSs and may provide coordination and control of these BSs. The network controller 130 may communicate with each BS via a backhaul. These BSs may also communicate with each other directly or indirectly, for example, via a wireless or wired backhaul.

[0040] UEs 120 (e.g., 120a, 120b, 120c) may be dispersed throughout the wireless network 100, and each UE may be stationary or mobile. A UE may also be referred to as an access terminal, terminal, mobile station, subscriber unit, station, etc. A UE may be a cellular phone (e.g., a smartphone), a personal digital assistant (PDA), a wireless modem, a wireless communication device, a handheld device, a laptop computer, a cordless phone, a wireless local loop (WLL) station, a tablet, a camera, a gaming device, a netbook, a smartbook, an ultrabook, a medical device or equipment, a biometric sensor / device, a wearable device (e.g., a smart watch, smart clothing, smart glasses, a smart wristband, smart jewelry (e.g., a smart ring, a smart bracelet)), an entertainment device (e.g., a music or video device, or a satellite radio), a vehicle component or sensor, a smart meter / sensor, industrial manufacturing equipment, a global positioning system device, or any other suitable device configured to communicate via a wireless or wired medium.

[0041] Some UEs may be considered machine type communication (MTC) UEs, or evolved or enhanced machine type communication (eMTC) UEs. MTC and eMTC UEs include, for example, robots, drones, remote devices, sensors, meters, monitors, location tags, and the like, which may communicate with a base station, another device (e.g., a remote device), or some other entity. A wireless node may provide connectivity to or to a network (e.g., a wide area network (such as the Internet) or a cellular network), for example, via a wired or wireless communication link. Some UEs may be considered Internet of Things (IoT) devices, and / or may be implemented as NB-IoT (narrowband Internet of Things) devices. Some UEs may be considered customer premises equipment (CPE). UE 120 may be included inside a housing that houses components of UE 120, such as a processor component, a memory component, and the like.

[0042] In general, any number of wireless networks may be deployed in a given geographic area. Each wireless network may support a specific RAT and may operate on one or more frequencies. A RAT may also be referred to as a radio technology, air interface, etc. A frequency may also be referred to as a carrier, frequency channel, etc. Each frequency may support a single RAT in a given geographic area to avoid interference between wireless networks of different RATs. In some cases, NR or 5G RAT networks may be deployed.

[0043] In some aspects, two or more UEs 120 (e.g., shown as UE 120a and UE 120e) may communicate directly (e.g., without using base station 110 as an intermediary) using one or more sidelink channels. For example, the UEs 120 may communicate using peer-to-peer (P2P) communication, device-to-device (D2D) communication, vehicle-to-everything (V2X) protocols (e.g., which may include vehicle-to-vehicle (V2V) protocols, vehicle-to-infrastructure (V2I) protocols, etc.), mesh networks, etc. In this scenario, the UEs 120 may perform scheduling operations, resource selection operations, and / or other operations described elsewhere herein as being performed by the base station 110.

[0044] As indicated above, Figure 1 These are provided as examples only. Other examples may differ from those regarding Figure 1 The content described.

[0045] Figure 2 A block diagram shows a design 200 of a base station 110 and a UE 120, which may be Figure 1One for each base station and one for each UE in . Base station 110 may be equipped with T antennas 234a through 234t, and UE 120 may be equipped with R antennas 252a through 252r, where in general T≧1 and R≧1.

[0046] At base station 110, transmit processor 220 may receive data for one or more UEs from data source 212, select one or more modulation and coding schemes (MCS) for each UE based at least in part on a channel quality indicator (CQI) received from the UE, process (e.g., encode and modulate) the data for each UE based at least in part on the MCS selected for the UE, and provide data symbols for all UEs. Transmit processor 220 may also process system information (e.g., for semi-static resource partitioning information (SRPI)) and control information (e.g., CQI requests, grants, upper layer signaling, etc.), and provide overhead symbols and control symbols. Transmit processor 220 may also generate reference symbols for reference signals (e.g., a cell-specific reference signal (CRS)) and synchronization signals (e.g., a primary synchronization signal (PSS) and a secondary synchronization signal (SSS)). A transmit (TX) multiple-input, multiple-output (MIMO) processor 230 may perform spatial processing (e.g., precoding) on data symbols, control symbols, overhead symbols, and / or reference symbols, as applicable, and may provide T output symbol streams to T modulators (MODs) 232a through 232t. Each modulator 232 may process a respective output symbol stream (e.g., for OFDM, etc.) to obtain an output sample stream. Each modulator 232 may further process (e.g., convert to analog, amplify, filter, and frequency upconvert) the output sample stream to obtain a downlink signal. The T downlink signals from modulators 232a through 232t may be transmitted via T antennas 234a through 234t, respectively. According to various aspects described in greater detail below, position coding may be utilized to generate synchronization signals to convey additional information.

[0047] At UE 120, antennas 252a through 252r may receive downlink signals from base station 110 and / or other base stations and may provide received signals to demodulators (DEMODs) 254a through 254r, respectively. Each demodulator 254 may condition (e.g., filter, amplify, downconvert, and digitize) the received signal to obtain input samples. Each demodulator 254 may further process the input samples (e.g., for OFDM, etc.) to obtain received symbols. A MIMO detector 256 may obtain received symbols from all R demodulators 254a through 254r, perform MIMO detection on the received symbols where applicable, and provide detected symbols. A receive processor 258 may process (e.g., demodulate and decode) the detected symbols, provide decoded data for UE 120 to a data sink 260, and provide decoded control information and system information to a controller / processor 280. The channel processor may determine reference signal received power (RSRP), received signal strength indicator (RSSI), reference signal received quality (RSRQ), channel quality indicator (CQI), etc. In some aspects, one or more components of UE 120 may be included in a housing.

[0048] On the uplink, at the UE 120, a transmit processor 264 may receive and process data from a data source 262 and control information (e.g., for reports including RSRP, RSSI, RSRQ, CQI, etc.) from the controller / processor 280. The transmit processor 264 may also generate reference symbols for one or more reference signals. The symbols from the transmit processor 264 may be precoded by a TX MIMO processor 266, if applicable, further processed by modulators 254a through 254r (e.g., for DFT-s-OFDM, CP-OFDM, etc.), and transmitted to the base station 110. At the base station 110, uplink signals from the UE 120 and other UEs may be received by the antennas 234, processed by the demodulators 254, detected by the MIMO detector 236, if applicable, and further processed by the receive processor 238 to obtain decoded data and control information transmitted by the UE 120. Receive processor 238 may provide decoded data to data sink 239 and decoded control information to controller / processor 240. Base station 110 may include communication unit 244 and communicate with network controller 130 via communication unit 244. Network controller 130 may include communication unit 294, controller / processor 290, and memory 292.

[0049] The controller / processor 240 of the base station 110, the controller / processor 280 of the UE 120, and / or Figure 2Any other component(s) of the controller / processor 240 of the base station 110, the controller / processor 280 of the UE 120, and / or Figure 2 Any other component(s) may perform or direct e.g. Figure 7 and 8 The operations of processes 700, 800 and / or other processes as described herein may be performed. Memories 242 and 282 may store data and program codes for base station 110 and UE 120, respectively. Scheduler 246 may schedule UEs for data transmission on the downlink and / or uplink.

[0050] In some aspects, UE 120 may include means for receiving, means for transmitting, means for instructing, and means for training. Such means may include in conjunction with Figure 2 One or more components of a UE 120 or base station 110 are depicted.

[0051] As indicated above, Figure 2 These are provided as examples only. Other examples may differ from those regarding Figure 2 The content described.

[0052] In some cases, different types of devices supporting different types of applications and / or services may coexist in a cellular cell. Examples of different types of devices include UE handsets, customer premises equipment (CPE), vehicles, Internet of Things (IoT) devices, and the like. Examples of different types of applications include ultra-reliable low-latency communications (URLLC) applications, massive machine-type communications (mMTC) applications, enhanced mobile broadband (eMBB) applications, vehicle-to-everything (V2X) applications, and the like. Furthermore, in some cases, a single device may support different applications or services simultaneously.

[0053] Figure 3An example implementation of a system on a chip (SOC) 300 according to certain aspects of the present disclosure is illustrated, which may include a central processing unit (CPU) 302 or a multi-core CPU configured for channel state information (CSI) learning. The SOC 300 may be included in a base station 110 or a UE 120. Variables (e.g., neural signals and synaptic weights), system parameters associated with a computing device (e.g., a neural network with weights), delays, frequency bin information, and task information may be stored in a memory block associated with a neural processing unit (NPU) 308, a memory block associated with the CPU 302, a memory block associated with a graphics processing unit (GPU) 304, a memory block associated with a digital signal processor (DSP) 306, a memory block 318, or may be distributed across multiple blocks. Instructions executed at the CPU 302 may be loaded from a program memory associated with the CPU 302 or may be loaded from the memory block 318.

[0054] The SOC 300 may also include additional processing blocks tailored for specific functions, such as a GPU 304, a DSP 306, a connectivity block 310 (which may include fifth generation (5G) connectivity, fourth generation long term evolution (4G LTE) connectivity, Wi-Fi connectivity, USB connectivity, Bluetooth connectivity, etc.), and a multimedia processor 312 that may, for example, detect and recognize gestures. In one implementation, the NPU is implemented in the CPU, DSP, and / or GPU. The SOC 300 may also include a sensor processor 314, an image signal processor (ISP) 316, and / or a navigation module 320 (which may include a global positioning system).

[0055] The SOC 300 may be based on the ARM instruction set. In one aspect of the present disclosure, the instructions loaded into the general purpose processor 302 may include code for transmitting, code for receiving, code for indicating, and code for training.

[0056] Deep learning architectures perform object recognition tasks by learning to represent inputs at successively higher levels of abstraction at each layer, thereby constructing useful feature representations of the input data. In this way, deep learning addresses a major bottleneck of traditional machine learning. Before the advent of deep learning, machine learning approaches to object recognition problems might rely heavily on human-engineered features, perhaps combined with shallow classifiers. A shallow classifier might be, for example, a two-class linear classifier, where the weighted sum of the feature vector components is compared to a threshold to predict which class the input belongs to. Human-engineered features might be templates or kernels customized for a specific problem domain by engineers with domain expertise. In contrast, deep learning architectures learn to represent features similar to those that a human engineer might design, but they do so through training. Furthermore, deep networks can learn to represent and recognize new types of features that humans might not have considered.

[0057] Deep learning architectures can learn hierarchies of features. For example, if a first layer is presented with visual data, it can learn to recognize relatively simple features (such as edges) in the input stream. In another example, if a first layer is presented with auditory data, it can learn to recognize spectral power in specific frequencies. A second layer, taking the output of the first layer as input, can learn to recognize combinations of features, such as simple shapes for visual data or combinations of sounds for auditory data. For example, higher layers can learn to represent complex shapes in visual data or words in auditory data. Even higher layers can learn to recognize common visual objects or spoken phrases.

[0058] Deep learning architectures can perform particularly well when applied to problems with a natural hierarchical structure. For example, the classification of motor vehicles can benefit from first learning to recognize wheels, windshields, and other features. These features can then be combined in different ways at higher levels to identify cars, trucks, and airplanes.

[0059] Neural networks can be designed with various connectivity patterns. In a feedforward network, information is passed from lower layers to higher layers, with each neuron in a given layer communicating to neurons in a higher layer. As described above, hierarchical representations can be constructed in successive layers of a feedforward network. Neural networks can also have reflow or feedback (also known as top-down) connections. In a reflow connection, the output from a neuron in a given layer can be communicated to another neuron in the same layer. The reflow architecture can help identify patterns that span more than one chunk of input data delivered sequentially to the neural network. The connections from neurons in a given layer to neurons in lower layers are called feedback (or top-down) connections. A network with many feedback connections can be helpful when the recognition of high-level concepts can assist in discerning specific low-level features of the input.

[0060] The connections between layers of a neural network can be fully connected or partially connected. Figure 4A Illustrated is an example of a fully connected neural network 402. In the fully connected neural network 402, a neuron in a first layer may communicate its output to every neuron in a second layer, such that every neuron in the second layer will receive input from every neuron in the first layer. Figure 4B An example of a locally connected neural network 404 is illustrated. In the locally connected neural network 404, neurons in a first layer may be connected to a limited number of neurons in a second layer. More generally, the locally connected layers of the locally connected neural network 404 may be configured such that each neuron in a layer will have the same or similar connectivity pattern, but their connection strengths may have different values (e.g., 410, 412, 414, and 416). The locally connected connectivity pattern may produce spatially distinct receptive fields in higher layers because higher layer neurons in a given area may receive inputs that are tuned through training to properties of a limited portion of the total input to the network.

[0061] An example of a locally connected neural network is a convolutional neural network. Figure 4C An example of a convolutional neural network 406 is illustrated. The convolutional neural network 406 can be configured such that the connection strengths associated with the inputs to each neuron in the second layer are shared (e.g., 408). Convolutional neural networks may be well suited for problems where the spatial location of the inputs is meaningful.

[0062] One type of convolutional neural network is a deep convolutional network (DCN). Figure 4D A detailed example of a DCN 400 designed to recognize visual features from an image 426 input from an image capture device 430 (such as an onboard camera) is illustrated. The DCN 400 of the current example can be trained to identify traffic signs and the numbers provided on traffic signs. Of course, the DCN 400 can be trained for other tasks, such as identifying lane markings or identifying traffic lights.

[0063] DCN 400 can be trained using supervised learning. During training, an image (such as image 426 of a speed limit sign) can be presented to DCN 400, and a "forward pass" can then be calculated to produce output 422. DCN 400 can include a feature extraction section and a classification section. Upon receiving image 426, convolution layer 432 can apply a convolution kernel (not shown) to image 426 to generate a first set of feature maps 418. As an example, the convolution kernel of convolution layer 432 can be a 5x5 kernel that generates a 28x28 feature map. In this example, since four different feature maps are generated in first set of feature maps 418, four different convolution kernels are applied to image 426 at convolution layer 432. Convolution kernels can also be referred to as filters or convolution filters.

[0064] The first set of feature maps 418 may be subsampled by a max pooling layer (not shown) to generate a second set of feature maps 420. The max pooling layer reduces the size of the first set of feature maps 418. That is, the size of the second set of feature maps 420 (such as 14x14) is smaller than the size of the first set of feature maps 418 (such as 28x28). The reduced size provides similar information to subsequent layers while reducing memory consumption. The second set of feature maps 420 may be further convolved via one or more subsequent convolutional layers (not shown) to generate one or more subsequent sets of feature maps (not shown).

[0065] exist Figure 4D In the example of , the second set of feature maps 420 is convolved to generate a first feature vector 424. In addition, the first feature vector 424 is further convolved to generate a second feature vector 428. Each feature of the second feature vector 428 may include a number corresponding to a possible feature of the image 426 (such as "sign," "60," and "100"). A softmax function (not shown) may convert the numbers in the second feature vector 428 into probabilities. Thus, the output 422 of the DCN 400 is the probability that the image 426 includes one or more features.

[0066] In this example, the probabilities for "logo" and "60" in output 422 are higher than the probabilities for other features of output 422 (such as "30," "40," "50," "70," "80," "90," and "100"). Before training, output 422 generated by DCN 400 is likely incorrect. Thus, the error between output 422 and the target output can be calculated. The target output is the true value of image 426 (e.g., "logo" and "60"). The weights of DCN 400 can then be adjusted to more closely align output 422 of DCN 400 with the target output.

[0067] To adjust the weights, the learning algorithm can calculate a gradient vector for the weights. This gradient can indicate the amount by which the error will increase or decrease if the weights are adjusted. At the top layer, this gradient can directly correspond to the value of the weight connecting the activated neurons in the penultimate layer to the neurons in the output layer. In lower layers, this gradient can depend on the value of the weights and the calculated error gradient for the higher layers. The weights can then be adjusted to reduce the error. This method of adjusting weights can be called "backpropagation" because it involves a "backward pass" through the neural network.

[0068] In practice, the error gradient of the weights may be calculated over a small number of examples so that the calculated gradient approximates the true error gradient. This approximation method may be called stochastic gradient descent. Stochastic gradient descent may be repeated until the error rate achievable by the entire system has stopped decreasing or until the error rate has reached a target level. After learning, the DCN may be presented with a new image (e.g., image 426 of a speed limit sign) and a forward pass through the network may produce output 422, which may be considered an inference or prediction of the DCN.

[0069] Deep Belief Network (DBN) is a probabilistic model comprising multiple layers of hidden nodes. DBN can be used to extract hierarchical representations of training data sets. DBN can be obtained by stacking multiple layers of restricted Boltzmann machines (RBM). RBM is a type of artificial neural network that can learn probability distributions on input sets. Since RBM can learn probability distributions without information about which class each input should be classified into, RBM is often used in unsupervised learning. Using a hybrid unsupervised and supervised paradigm, the bottom RBM of the DBN can be trained in an unsupervised manner and can be used as a feature extractor, while the top RBM can be trained in a supervised manner (on the joint distribution of the input and target class from the previous layer) and can be used as a classifier.

[0070] A deep convolutional network (DCN) is a network of convolutional networks with additional pooling and normalization layers. DCNs have achieved state-of-the-art performance on many tasks. DCNs can be trained using supervised learning, where both the input and output targets are known for many examples and are used to modify the network weights using gradient descent.

[0071] The DCN can be a feedforward network. In addition, as described above, the connections from neurons in the first layer of the DCN to the neuron groups in the next higher layer are shared across the neurons in the first layer. The feedforward and shared connections of the DCN can be used to perform fast processing. The computational burden of the DCN can be much smaller than, for example, the computational burden of a similarly sized neural network that includes recurrent or feedback connections.

[0072] The processing of each layer of the convolutional network can be thought of as a spatially invariant template or basis projection. If the input is first decomposed into multiple channels, such as the red, green, and blue channels of a color image, then the convolutional network trained on this input can be thought of as three-dimensional, with two spatial dimensions along the axes of the image and a third dimension that captures color information. The output of the convolutional connection can be thought of as forming a feature map in the subsequent layer, each element in which receives input from a range of neurons in the previous layer (e.g., feature map 218) and from each of the multiple channels. The values in the feature map can be further processed with nonlinearities (such as rectification, max(0, x)). The values from adjacent neurons can be further pooled (which corresponds to downsampling) and can provide additional local invariance and dimensionality reduction. Normalization can also be applied by lateral inhibition between neurons in the feature map, which corresponds to whitening.

[0073] The performance of deep learning architectures improves as more labeled data points become available or as computing power increases. Modern deep neural networks are routinely trained using thousands of times more computing resources than were available to typical researchers just fifteen years ago. New architectures and training paradigms can further boost deep learning performance. Rectified linear units can alleviate the training problem known as vanishing gradients. New training techniques can reduce overfitting and thus enable larger models to achieve better generalization. Encapsulation techniques can abstract the data within a given receptive field and further improve overall performance.

[0074] Figure 5 is a block diagram illustrating a deep convolutional network 550. The deep convolutional network 550 may include multiple layers of different types based on connectivity and weight sharing. Figure 5 As shown in FIG, the deep convolutional network 550 includes convolution blocks 554A and 554B. Each of the convolution blocks 554A and 554B may be configured with a convolution layer (CONV) 356, a normalization layer (LNorm) 558, and a maximum pooling layer (MAX POOL) 560.

[0075] The convolution layer 556 may include one or more convolution filters that can be applied to the input data to generate a feature map. Although only two convolution blocks 554A and 554B are shown, the present disclosure is not limited thereto, and instead any number of convolution blocks 554A and 554B may be included in the deep convolutional network 550 according to design preferences. The normalization layer 558 may normalize the output of the convolution filter. For example, the normalization layer 558 may provide whitening or lateral suppression. The maximum pooling layer 560 may provide spatial downsampling aggregation to achieve local invariance and dimensionality reduction.

[0076] For example, the parallel filter banks of the deep convolutional network can be offloaded onto the CPU 302 or GPU 304 of the SOC 300 to achieve high performance and low power consumption. In alternative embodiments, the parallel filter banks can be offloaded onto the DSP 306 or ISP 316 of the SOC 300. In addition, the deep convolutional network 550 can access other processing blocks that may be present on the SOC 300, such as the sensor processor 314 and the navigation module 320 dedicated to sensors and navigation, respectively.

[0077] The deep convolutional network 550 may also include one or more fully connected layers 562 (FC1 and FC2). The deep convolutional network 550 may further include a logistic regression (LR) layer 564. Between each layer 556, 558, 560, 562, 564 of the deep convolutional network 550 are weights (not shown) to be updated. The output of each layer (e.g., 556, 558, 560, 562, 564) can be used as input to a subsequent layer (e.g., 556, 558, 560, 562, 564) in the deep convolutional network 550 to learn a hierarchical feature representation from the input data 552 (e.g., image, audio, video, sensor data, and / or other input data) supplied from the first convolutional block 554A. The output of the deep convolutional network 550 is a classification score 566 for the input data 552. The classification score 566 can be a set of probabilities, where each probability is a probability that the input data includes a feature from a feature set.

[0078] As indicated above, Figure 3-5 are provided as examples. Other examples may differ from those described in Figure 3-5 The content described.

[0079] In machine learning (ML)-based channel state information (CSI) compression and feedback, a user equipment (UE) trains an encoder / decoder neural network (NN) pair and sends the trained decoder model to a base station (e.g., gNB). The UE uses the encoder NN to create and feed back channel state feedback (CSF). The encoder NN can take the original channel as input. The base station uses the decoder NN to recover the original channel from the CSF. The UE trains the encoder / decoder NN using a loss metric.

[0080] Figure 6 is a block diagram illustrating an exemplary autoencoder 600 according to aspects of the present disclosure. The autoencoder 600 includes an encoder 610 having a neural network (NN). The encoder 610 receives a channel realization and / or an interference realization as input and compresses the channel / interference realization. The channel realization may also be referred to as a channel estimate. The interference realization may also be referred to as an interference estimate. Interference depends on the environment and may address uplink interference or inter-stream interference in MIMO scenarios.

[0081] Compressed channel state feedback is output from the encoder 610. The autoencoder 600 also has a decoder 620 that receives the compressed channel state feedback output from the encoder 610. The decoder 620 passes the received information through a fully connected layer and a series of convolutional layers to recover the channel state (e.g., an approximate channel state).

[0082] The UE trains the encoder 610 and decoder 620 and transmits the decoder coefficients to the base station from time to time. At a higher frequency, the UE sends the output of the encoder 610 (e.g., the channel state feedback or compressed output of the encoder 610) to the base station. As the UE moves around, the weights of the decoder 620 may change. That is, when the channel environment changes, the decoder weights (e.g., coefficients) may also change. The updated decoder coefficients can thus be fed back from the UE to the base station to reflect the ever-changing environment. In other words, the UE can train the decoder based on the existing environment rather than just the encoder. The coefficients can be sent from the UE according to a timeline configured by radio resource control (RRC) signaling. In one configuration, the coefficients are sent at a lower frequency than the frequency at which the channel state feedback is sent.

[0083] Figure 7 is a diagram illustrating an example process 700, performed, for example, by a UE, in accordance with various aspects of the present disclosure. The example process 700 is an example of receiving and training a configurable neural network for channel state feedback (CSF) learning.

[0084] like Figure 7 As shown in FIG, in some aspects, process 700 may include receiving a plurality of neural network training configurations for channel state feedback (CSF), each configuration corresponding to a different neural network framework (block 702). For example, a UE (e.g., using antenna 252, DEMOD 254, MIMO detector 256, receive processor 258, controller / processor 280, memory 282, etc.) may receive the neural network training configurations.

[0085] like Figure 7 As shown in , in some aspects, process 700 may include training each of a population of neural network decoder / encoder pairs according to the received training configuration (block 704). For example, the UE (e.g., using controller / processor 280, memory 282, etc.) may train each neural network decoder / encoder pair.

[0086] Figure 8 is a diagram illustrating an example process 800, performed, for example, by a base station, in accordance with various aspects of the present disclosure. The example process 800 is an example of processing a configurable neural network for channel state feedback (CSF) learning.

[0087] like Figure 8As shown in FIG, in some aspects, process 800 may include transmitting a plurality of neural network training configurations for channel state feedback (CSF) to a user equipment (UE), each configuration corresponding to a different neural network framework (block 802). For example, a base station (e.g., using antennas 234, MOD 232, TX MIMO processor 230, transmit processor 220, controller / processor 240, memory 242, etc.) may transmit the neural network training configurations.

[0088] like Figure 8 As shown in , in some aspects, process 800 may include receiving a neural network decoder / encoder pair trained according to the training configurations (block 804). For example, a base station (e.g., using antennas 234, DEMOD 232, MIMO detector 236, receive processor 238, controller / processor 240, memory 242, etc.) may receive the neural network decoder / encoder pair.

[0089] In baseline federated learning, the base station maintains an encoder / decoder NN for each UE. The UE downloads the model maintained at the base station. Additionally, the UE trains and updates the model based on the channel / interference observed at the UE. The UE then sends the encoder and decoder networks to the gNB. The gNB then aggregates the models from multiple UEs and generates a new model for these UEs. This process is then repeated.

[0090] Different encoder / decoder neural network frameworks (e.g., architectures) can be used by the UE to perform neural network training for channel state feedback (CSF) depending on how the base station expects to use the CSF. For example, different base station antenna structures (such as 1D / 2D cross-polarization or 1D / 2D vertical / horizontal polarization) will benefit from different neural networks with different frameworks. Similarly, different UE mobility scenarios (such as low Doppler (e.g., 3km / h, 30km / h) or high Doppler (e.g., 120km / h or higher speeds)) can benefit from different frameworks. Other reasons for different neural network frameworks include: supporting different UEs of different computational complexity (e.g., shallow neural networks or deep neural networks for CSF); and different CSF quantities, such as original channel, channel quality indicator (CQI) / rank indicator (RI), or interference.

[0091] According to aspects of the present disclosure, a base station may configure a UE with multiple neural network frameworks for encoder / decoder neural network training for a CSF. Each configuration may be associated with a reference signal set and may specify one or more reporting quantities. The configuration may be indicated via higher layer messages, dynamic signaling, etc. (such as downlink control information (DCI), radio resource control (RRC) signaling, and / or medium access control-control element (MAC-CE)).

[0092] Different hyperparameters can be configured for different encoder / decoder neural network configurations. The hyperparameters for different UEs can be based on the desired feedback accuracy, feedback overhead, UE computational power, and base station antenna configuration. Different hyperparameters can indicate the number of layers in the neural network, the type of layer (e.g., convolutional or fully connected), the number of hidden units / kernel size for each layer (e.g., kernel size), and / or the loss function / loss metric of the neural network. For example, different loss metrics can be predefined and configured for different CSI quantities (e.g., RI, CQI, interference, etc.).

[0093] Hyperparameters can also specify the activation function (e.g., sigmoid, ReLU, tanh, etc.) and / or compression ratio used for each layer. For example, for a fully connected layer, the ratio of the number of neurons in the output layer to the number of neurons in the input layer can be defined. Hyperparameters can also specify the learning rate configuration and optimizer configuration, such as stochastic gradient descent, ADAM, Adadelta, etc.

[0094] The hyperparameters may further indicate the desired channel sample structure for different neural network frameworks. For example, in the case of 16 cross-polarized (+-45°) TxRUs (4 vertical and 4 horizontal, 4X4=16) on a base station with 2D deployment, the total number of antennas is 2X16=32 with +45 or -45 polarization. If only 32 taps are considered in the time domain, the channel sample size for training may be [I / Q][Pol][V][H][tap]=2X2X4X4X32 (complex values for each polarized antenna, I / Q for real and imaginary parts respectively). The structure of the channel sample may also be [I / Q][ant][tap]=2X32X32. Different input samples will then also be used for different frameworks of the neural network.

[0095] Hyperparameters can also indicate whether single-timestep analysis or multi-timestep analysis occurs. For example, the neural network can take into account the time-domain correlation of the wireless channel. That is, the hyperparameters can specify whether each channel sample is learned independently or jointly across different time slots.

[0096] In some aspects of the present disclosure, the hyperparameters are divided into multiple subsets. Some subsets include hyperparameters that are common to all UEs or all CSFs. Other subsets include hyperparameters that are specific to each UE or all CSFs.

[0097] According to a further aspect of the present disclosure, the base station indicates different CSF configurations with different neural network frameworks to the UE(s) via an RRC message. The message may include the number of neural network pairs to be trained at the UE and which neural network pair to use for a specific CSI quantity feedback instance. The message may also indicate the neural network architecture with hyperparameters used for CSF training for each CSI learning instance.

[0098] In yet another aspect of the present disclosure, new UE capabilities are introduced to support a maximum number of neural networks for simultaneous channel state compression and feedback.

[0099] The foregoing disclosure provides illustration and description, but is not intended to be exhaustive or to limit the aspects to the precise form disclosed. Modifications and variations may be made in light of the above disclosure or may be acquired from practice of the aspects.

[0100] As used herein, the term "component" is intended to be broadly interpreted as hardware, firmware, and / or a combination of hardware and software. As used herein, a processor is implemented using hardware, firmware, and / or a combination of hardware and software.

[0101] Some aspects are described herein in conjunction with thresholds. As used herein, satisfying a threshold may refer to a value being greater than a threshold, greater than or equal to a threshold, less than a threshold, less than or equal to a threshold, equal to a threshold, not equal to a threshold, etc., depending on the context.

[0102] It will be apparent that the systems and / or methods described herein can be implemented in various forms of hardware, firmware, and / or a combination of hardware and software. The actual dedicated control hardware or software code used to implement these systems and / or methods does not limit the aspects. Thus, the operation and behavior of these systems and / or methods are described herein without reference to specific software code—it is understood that software and hardware can be designed to implement these systems and / or methods based, at least in part, on the description herein.

[0103] Although specific feature combinations are described in the claims and / or disclosed in the specification, these combinations are not intended to limit the disclosure of the various aspects. In fact, many of these features can be combined in a manner not specifically described in the claims and / or not disclosed in the specification. Although each dependent claim listed below can be directly subordinate to only one claim, the disclosure of the various aspects includes that each dependent claim is combined with each other claim in this group of claims. The phrase quoting "at least one of" a list of items refers to any combination of these items, including single members. As an example, "at least one of a, b or c" is intended to cover: a, b, c, ab, ac, bc, and abc, as well as any combination with multiple identical elements (e.g., aa, aaa, aab, aac, abb, acc, bb, bbb, bbc, cc, and ccc, or any other arrangement of a, b and c).

[0104] The elements, actions or instructions used herein should not be interpreted as critical or necessary unless explicitly described as such. Moreover, as used herein, the articles "one" and "a" are intended to include one or more items and can be used interchangeably with "one or more". Furthermore, as used herein, the terms "set" and "group" are intended to include one or more items (e.g., related items, non-related items, a combination of related and non-related items, etc.) and can be used interchangeably with "one or more". Where intended to have only one item, the phrase "only one" or similar language is used. Furthermore, as used herein, the terms "having", "containing", "comprising" etc. are intended to be open terms. Furthermore, the phrase "based on" is intended to mean "based at least in part on", unless otherwise explicitly stated.

Claims

1. A method for wireless communication by a user equipment (UE), comprising: receiving a plurality of neural network training configurations for channel state feedback (CSF), each configuration corresponding to a different neural network framework; training each of a plurality of neural network decoder or encoder pairs according to the received training configuration; as well as A message is received indicating either activation or deactivation of a number of neural network pairs to be trained and which neural network pair to use for a particular amount of channel state information (CSI). The method of claim 1 , wherein each configuration is associated with a reference signal set. The method of claim 1 , wherein each configuration is associated with a reported quantity. The method of claim 1 , wherein each configuration corresponds to a different set of hyperparameters.

5. The method of claim 4, wherein each set of hyperparameters is based on desired feedback accuracy, feedback overhead, UE computing capability, and network entity antenna configuration.

6. The method of claim 4, wherein the hyperparameters include the number of neural network layers, the type of each neural network layer, the number of hidden units or kernel size used for each layer, and / or the activation function used for each layer.

7. The method of claim 4, wherein the hyperparameters include a compression ratio, a learning rate configuration, an optimizer configuration, and / or a loss metric associated with a CSI number.

8. The method of claim 4, wherein the hyperparameter indicates whether the structure of the encoder or decoder pair is based on antenna deployment characteristics and / or whether time-domain correlation of a wireless channel is to be considered.

9. The method of claim 4, wherein the super parameter sets include first multiple subsets containing super parameters common to UEs or CSFs and second multiple subsets containing super parameters that vary among UEs or vary among the number of CSFs.

10. The method of claim 1, further comprising indicating a maximum number of neural networks supported.

11. A method for wireless communication by a network entity, comprising: transmitting a plurality of neural network training configurations for channel state feedback (CSF) to a user equipment (UE), each configuration corresponding to a different neural network framework; receiving a neural network decoder or encoder pair trained according to the neural network training configuration; as well as A message is transmitted indicating either activation or deactivation of a number of neural network pairs to be trained and which neural network pair to use for a particular amount of channel state information (CSI). The method of claim 11 , wherein each configuration is associated with a reference signal set. The method of claim 11 , wherein each configuration is associated with a reported quantity. The method of claim 11 , wherein each configuration corresponds to a different set of hyperparameters.

15. The method of claim 14, wherein each set of hyperparameters is based on desired feedback accuracy, feedback overhead, UE computing capability, and network entity antenna configuration.

16. The method of claim 14, wherein the hyperparameters include the number of neural network layers, the type of each neural network layer, the number of hidden units or kernel size used for each layer, and / or the activation function used for each layer.

17. The method of claim 14, wherein the hyperparameters include a compression ratio, a learning rate configuration, an optimizer configuration, and / or a loss metric associated with a CSI number.

18. The method of claim 14, wherein the hyperparameter indicates whether the structure of an encoder or decoder pair is based on antenna deployment characteristics, and / or whether time-domain correlation of a wireless channel is to be considered.

19. The method of claim 14, wherein the super parameter sets include first multiple subsets containing super parameters common to UEs or common to CSFs and second multiple subsets containing super parameters that vary among UEs or vary among the number of CSFs.

20. The method of claim 11, further comprising receiving an indication of a maximum number of neural networks supported.

21. A UE (User Equipment) for wireless communication, comprising: Memory, and at least one processor operatively coupled to the memory, the memory and the at least one processor configured to: receiving a plurality of neural network training configurations for channel state feedback (CSF), each configuration corresponding to a different neural network framework; training each neural network decoder or encoder pair according to the received training configuration; as well as A message is received indicating either activation or deactivation of a number of neural network pairs to be trained and which neural network pair to use for a particular amount of channel state information (CSI).

22. The UE of claim 21, wherein each configuration is associated with a reference signal set.

23. The UE of claim 21, wherein each configuration is associated with a number of reports.

24. The UE of claim 21, wherein each configuration corresponds to a different set of hyperparameters.

25. The UE of claim 24, wherein each set of hyperparameters is based on desired feedback accuracy, feedback overhead, UE computational capability, and network entity antenna configuration.

26. The UE of claim 24, wherein the hyperparameters include the number of neural network layers, the type of each neural network layer, the number of hidden units or kernel size used for each layer, and / or the activation function used for each layer.

27. The UE of claim 24, wherein the hyperparameters include a compression ratio, a learning rate configuration, an optimizer configuration, and / or a loss metric associated with a CSI number.

28. The UE of claim 24, wherein the hyperparameter indicates whether the structure of an encoder or decoder pair is based on antenna deployment characteristics and / or whether time-domain correlation of a wireless channel is to be considered.

29. The UE of claim 24, wherein the super parameter set comprises a first plurality of subsets including super parameters common to UEs or common to CSFs and a second plurality of subsets including super parameters that vary among UEs or vary among the number of CSFs.

30. The UE of claim 21, wherein the at least one processor is further configured to indicate a maximum number of neural networks supported.

31. A network entity for wireless communication, comprising: Memory, and at least one processor operatively coupled to the memory, the memory and the at least one processor configured to: transmitting a plurality of neural network training configurations for channel state feedback (CSF) to a user equipment (UE), each configuration corresponding to a different neural network framework; receiving a neural network decoder or encoder pair trained according to the neural network training configuration; as well as A message is transmitted indicating either activation or deactivation of a number of neural network pairs to be trained and which neural network pair to use for a particular amount of channel state information (CSI).

32. The network entity of claim 31, wherein each configuration is associated with a reference signal set.

33. The network entity of claim 31, wherein each configuration is associated with a number of reports.

34. The network entity of claim 31, wherein each configuration corresponds to a different set of hyperparameters.

35. The network entity of claim 34, wherein each set of hyperparameters is based on desired feedback accuracy, feedback overhead, UE computational capability, and network entity antenna configuration.

36. The network entity of claim 34, wherein the hyperparameters include the number of neural network layers, the type of each neural network layer, the number of hidden units or kernel size used for each layer, and / or the activation function used for each layer.

37. The network entity of claim 34, wherein the hyperparameters include a compression ratio, a learning rate configuration, an optimizer configuration, and / or a loss metric associated with a CSI number.

38. The network entity of claim 34, wherein the hyperparameter indicates whether the structure of an encoder or decoder pair is based on antenna deployment characteristics, and / or whether time-domain correlation of a wireless channel is to be considered.

39. The network entity of claim 34, wherein the super parameter set comprises a first plurality of subsets including super parameters common to UEs or common to CSFs and a second plurality of subsets including super parameters that vary among UEs or vary among the number of CSFs.

40. The network entity of claim 31 , wherein the at least one processor is further configured to receive an indication of a maximum number of neural networks supported.

41. A UE (user equipment) for wireless communication, comprising: means for receiving a plurality of neural network training configurations for channel state feedback (CSF), each configuration corresponding to a different neural network framework; means for training each neural network decoder or encoder pair according to the received training configuration; as well as Means for receiving a message indicating either activation or deactivation of a number of neural network pairs to be trained and which neural network pair to use for a particular amount of channel state information (CSI).

42. The UE of claim 41, wherein each configuration is associated with a reference signal set.

43. The UE of claim 41 , wherein each configuration is associated with a number of reports.

44. The UE of claim 41, wherein each configuration corresponds to a different set of hyperparameters.

45. The UE of claim 44, wherein each set of hyperparameters is based on desired feedback accuracy, feedback overhead, UE computational capability, and network entity antenna configuration.

46. The UE of claim 44, wherein the hyperparameters include the number of neural network layers, the type of each neural network layer, the number of hidden units or kernel size used for each layer, and / or the activation function used for each layer.

47. The UE of claim 44, wherein the hyperparameters include a compression ratio, a learning rate configuration, an optimizer configuration (e.g., stochastic gradient descent, ADAM), and / or a loss metric associated with the number of CSIs.

48. The UE of claim 44, wherein the hyperparameter indicates whether the structure of an encoder or decoder pair is based on antenna deployment characteristics, and / or whether time-domain correlation of a wireless channel is to be considered.

49. The UE of claim 44, wherein the super parameter set comprises a first plurality of subsets containing super parameters common to UEs or common to CSFs and a second plurality of subsets containing super parameters that vary among UEs or vary among the number of CSFs.

50. The UE of claim 41 , further comprising means for indicating a maximum number of neural networks supported.

51. A network entity for wireless communication, comprising: means for transmitting a plurality of neural network training configurations for channel state feedback (CSF) to a user equipment (UE), each configuration corresponding to a different neural network framework; means for receiving a neural network decoder or encoder pair trained according to the neural network training configuration; as well as Means for transmitting a message indicating either activation or deactivation of a number of neural network pairs to be trained and which neural network pair to use for a particular amount of channel state information (CSI).

52. The network entity of claim 51, wherein each configuration is associated with a reference signal set.

53. The network entity of claim 51, wherein each configuration is associated with a number of reports.

54. The network entity of claim 51, wherein each configuration corresponds to a different set of hyperparameters.

55. The network entity of claim 54, wherein each set of hyperparameters is based on desired feedback accuracy, feedback overhead, UE computational capability, and network entity antenna configuration.

56. The network entity of claim 54, wherein the hyperparameters include the number of neural network layers, the type of each neural network layer, the number of hidden units or kernel size used for each layer, and / or the activation function used for each layer.

57. The network entity of claim 54, wherein the hyperparameters include a compression ratio, a learning rate configuration, an optimizer configuration, and / or a loss metric associated with a CSI number.

58. The network entity of claim 54, wherein the hyperparameter indicates whether the structure of an encoder or decoder pair is based on antenna deployment characteristics, and / or whether time-domain correlation of a wireless channel is to be considered.

59. The network entity of claim 54, wherein the super parameter set comprises first multiple subsets containing super parameters common to UEs or common to CSFs and second multiple subsets containing super parameters that vary among UEs or vary among the number of CSFs.

60. The network entity of claim 51 , further comprising means for receiving an indication of a maximum number of neural networks supported.

61. A non-transitory computer-readable medium having program code recorded thereon, the program code being executed by a UE (user equipment) and comprising: program code for receiving a plurality of neural network training configurations for channel state feedback (CSF), each configuration corresponding to a different neural network framework; program code for training each neural network decoder or encoder pair according to the received training configuration; as well as Program code for receiving a message indicating either activation or deactivation of a number of neural network pairs to be trained and which neural network pair to use for a particular amount of channel state information (CSI).

62. A non-transitory computer-readable medium having program code recorded thereon, the program code being executed by a network entity and comprising: program code for transmitting a plurality of neural network training configurations for channel state feedback (CSF) to a user equipment (UE), each configuration corresponding to a different neural network framework; program code for receiving a neural network decoder or encoder pair trained according to the neural network training configuration; as well as Program code for transmitting a message indicating either activation or deactivation of a number of neural network pairs to be trained and which neural network pair to use for a particular amount of channel state information (CSI).

Images