Neural network and antenna configuration instructions

Abstract

Disclosed are wireless communication systems and methods related to indicating antenna configuration information and neural network information. Neural network information can be selected for the encoder side based on the antenna configuration of the first device that houses the encoder. This information can be sent to a second device along with the antenna configuration information, and the second device can jointly train the neural network with the first device. The first device can also send one or more weights after training, which are also stored at the second device along with the antenna configuration information. When a third device with a similar antenna configuration as the first device establishes communication with the second device, the second device can send the neural network information and weights to the third device. The third device can use this information instead of default information to accelerate neural network initialization and training.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to and the benefit of U.S. Patent Application No. 17 / 247,664, filed on December 18, 2020, and U.S. Provisional Patent Application No. 62 / 951,889, filed on December 20, 2019, which are hereby incorporated by reference in their entirety as if fully set forth in their entirety below and for all applicable purposes. Technical Field

[0003] The present application relates to wireless communication systems, and more particularly, to indicating antenna configuration information and neural network information between devices. Background Art

[0004] To meet the growing demand for expanded mobile broadband connectivity, wireless communication technology is evolving from Long Term Evolution (LTE) technology to the next generation New Radio (NR) technology (which may be referred to as the fifth generation (5G)). In a wireless communication network implementing such wireless communication technology, a wireless multiple access communication system may include multiple base stations (BSs), each of which simultaneously supports communication for multiple communication devices (which may also be referred to as user equipment (UE)).

[0005] As wireless communication technology continues to advance, neural networks have been used to address some of the challenges associated with complexity, performance, and other factors. However, training neural networks can be a time- and computationally intensive process. This can introduce inefficiencies and power consumption, among other issues, as devices must perform the necessary processing to train and otherwise implement the neural networks used to send and receive data. Therefore, there is a need for more efficient methods to facilitate the training of neural networks and their use in wireless communication systems. Summary of the Invention

[0006] The following summarizes some aspects of the present disclosure to provide a basic understanding of the technology discussed. This summary is not an exhaustive overview of all anticipated features of the present disclosure and is not intended to identify key or important elements of all aspects of the present disclosure, nor is it intended to delineate the scope of any or all aspects of the present disclosure. Its sole purpose is to provide some concepts of one or more aspects of the present disclosure in a summarized form as a prelude to more detailed embodiments that will be presented later.

[0007] For example, in one aspect of the present disclosure, a method of wireless communication includes establishing, by a first wireless communication device, communication with a second wireless communication device. The method also includes determining, by the first wireless communication device, a neural network to be used for the communication with the second wireless communication device based on an antenna configuration of the first wireless communication device. The method also includes transmitting, by the first wireless communication device, antenna configuration information and at least one neural network parameter to the second wireless communication device based on the neural network.

[0008] In an additional aspect of the present disclosure, a method of wireless communication includes: establishing, by a first wireless communication device, communication with a second wireless communication device. The method also includes receiving, by the first wireless communication device, from the second wireless communication device, antenna configuration information for an antenna configuration at the second wireless communication device and at least one neural network parameter of a neural network at the second wireless communication device for use in the communication. The method also includes storing, by the first wireless communication device, the antenna configuration information and the at least one neural network parameter.

[0009] In an additional aspect of the present disclosure, a first wireless communication device includes a transceiver configured to establish communication with a second wireless communication device. The first wireless communication device also includes a processor configured to determine a neural network to use for the communication with the second wireless communication device based on an antenna configuration of the first wireless communication device. The first wireless communication device further includes wherein the transceiver is further configured to send antenna configuration information and at least one neural network parameter to the second wireless communication device based on the neural network.

[0010] In an additional aspect of the present disclosure, a first wireless communication device includes a transceiver configured to establish communication with a second wireless communication device. The transceiver is further configured to receive, from the second wireless communication device, antenna configuration information for an antenna configuration at the second wireless communication device and at least one neural network parameter for a neural network at the second wireless communication device for use in the communication. The first wireless communication device also includes a processor configured to store the antenna configuration information and the at least one neural network parameter at the first wireless communication device.

[0011] In an additional aspect of the present disclosure, a non-transitory computer-readable medium having program code recorded thereon includes code for causing a first wireless communication device to establish communication with a second wireless communication device. The program code also includes code for causing the first wireless communication device to determine a neural network to use for the communication with the second wireless communication device based on an antenna configuration of the first wireless communication device. The program code also includes code for causing the first wireless communication device to transmit antenna configuration information and at least one neural network parameter to the second wireless communication device based on the neural network.

[0012] In an additional aspect of the present disclosure, a non-transitory computer-readable medium having program code recorded thereon includes code for causing a first wireless communication device to establish communication with a second wireless communication device. The program code also includes code for causing the first wireless communication device to receive, from the second wireless communication device, antenna configuration information for an antenna configuration at the second wireless communication device and at least one neural network parameter for a neural network at the second wireless communication device for use in the communication. The program code also includes code for causing the first wireless communication device to store the antenna configuration information and the at least one neural network parameter.

[0013] In an additional aspect of the present disclosure, a first wireless communication device includes means for establishing communication with a second wireless communication device. The first wireless communication device further includes means for determining a neural network to use for the communication with the second wireless communication device based on an antenna configuration of the first wireless communication device. The first wireless communication device further includes means for sending antenna configuration information and at least one neural network parameter to the second wireless communication device based on the neural network.

[0014] In an additional aspect of the present disclosure, a first wireless communication device includes means for establishing communication with a second wireless communication device. The first wireless communication device also includes means for receiving, from the second wireless communication device, antenna configuration information for an antenna configuration at the second wireless communication device and at least one neural network parameter of a neural network at the second wireless communication device for use in the communication. The first wireless communication device also includes means for storing the antenna configuration information and the at least one neural network parameter.

[0015] Other aspects, features, and embodiments of the present invention will become apparent to those of ordinary skill in the art after reviewing the following description of specific, exemplary embodiments of the present invention in conjunction with the accompanying drawings. Although features of the present invention may be discussed below with respect to certain embodiments and drawings, all embodiments of the present invention may include one or more of the advantageous features discussed herein. In other words, although one or more embodiments may be discussed as having certain advantageous features, one or more of these features may also be used in accordance with the various embodiments of the present invention discussed herein. In a similar manner, although exemplary embodiments may be discussed below as device, system, or method embodiments, it should be understood that these exemplary embodiments may be implemented in a wide variety of devices, systems, and methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A wireless communication network according to some embodiments of the present disclosure is shown.

[0017] Figure 2 An exemplary autoencoder configuration for use in wireless communications is shown in accordance with some embodiments of the present disclosure.

[0018] Figure 3 is a block diagram of a user equipment (UE) according to some embodiments of the present disclosure.

[0019] Figure 4 is a block diagram of an exemplary base station (BS) according to some embodiments of the present disclosure.

[0020] Figure 5 A protocol diagram illustrating a neural network and antenna configuration indication scheme according to some embodiments of the present disclosure.

[0021] Figure 6 A flowchart of a wireless communication method according to some embodiments of the present disclosure is shown.

[0022] Figure 7A A flowchart of a wireless communication method according to some embodiments of the present disclosure is shown.

[0023] Figure 7B A flowchart of a wireless communication method according to some embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0024] The detailed description set forth below in conjunction with the accompanying drawings is intended as a description of various configurations and is not intended to represent the only configuration in which the concepts described herein may be practiced. In order to provide a comprehensive understanding of the various concepts, the detailed description includes specific details. However, it will be apparent to those skilled in the art that these concepts may be practiced without these specific details. In some instances, in order to avoid obscuring these concepts, well-known structures and components are shown in block diagram form.

[0025] In general, the present disclosure relates to wireless communication systems (also referred to as wireless communication networks). In various embodiments, the techniques and apparatuses may be used in wireless communication networks such as code division multiple access (CDMA) networks, time division multiple access (TDMA) networks, frequency division multiple access (FDMA) networks, orthogonal FDMA (OFDMA) networks, single carrier FDMA (SC-FDMA) networks, LTE networks, global system for mobile communications (GSM) networks, fifth generation (5G) or new radio (NR) networks, and other communication networks. As described herein, the terms "network" and "system" may be used interchangeably.

[0026] OFDMA networks can implement radio technologies such as Evolved UTRA (E-UTRA), Institute of Electrical and Electronics Engineers (IEEE) 802.11, IEEE 802.16, IEEE 802.20, Flash-OFDM, and the like. UTRA, E-UTRA, and GSM are part of the Universal Mobile Telecommunications System (UMTS). Specifically, Long Term Evolution (LTE) is a version of UMTS that uses E-UTRA. UTRA, E-UTRA, GSM, UMTS, and LTE are described in documents provided by an organization named "3rd Generation Partnership Project" (3GPP), and cdma2000 is described in documents provided by an organization named "3rd Generation Partnership Project 2" (3GPP2). These various radio technologies and standards are either known or under development. For example, the 3rd Generation Partnership Project (3GPP) is a collaboration between various telecommunications association groups with the goal of defining globally applicable third generation (3G) mobile phone specifications. 3GPP Long Term Evolution (LTE) is a 3GPP project aimed at improving the UMTS mobile phone standard. 3GPP may define specifications for next generation mobile networks, mobile systems, and mobile devices. This disclosure may describe certain aspects with reference to LTE, 4G, or 5G NR technology, and involves shared access to wireless spectrum between networks using a collection of new and different radio access technologies or radio air interfaces.

[0027] Specifically, 5G networks are expected to enable diverse deployments, diverse spectrum, and diverse services and devices using a unified air interface based on OFDM. To achieve these goals, in addition to developing new radio technologies for 5G NR networks, further enhancements to LTE and LTE-A are also being considered. 5G NR will be able to scale to: (1) provide coverage for the massive Internet of Things (IoT), which has ultra-high density (e.g., ~1M nodes / km) 2 (1) providing deep coverage with ultra-low complexity (e.g., ~10s of bits / second), ultra-low energy (e.g., ~10+ years of battery life), and the ability to reach challenging locations; (2) providing coverage including mission-critical control with strong security for protecting sensitive personal, financial, or confidential information, ultra-high reliability (e.g., ~99.9999% reliability), ultra-low latency (e.g., ~1ms), and providing coverage to users with a wide range of mobility or lack of mobility; and (3) providing coverage with enhanced mobile broadband, including very high capacity (e.g., ~10Tbps / km 2 ), extreme data rates (e.g., multi-Gbps rates, 100+Mbps user experience rates), and deep perception with improved discovery and optimization.

[0028] 5G NR can be implemented using an optimized OFDM-based waveform with scalable numerology and transmission time intervals (TTIs); a common, flexible framework to efficiently multiplex services and features using dynamic, low-latency time division duplex (TDD) / frequency division duplex (FDD) designs; and advanced wireless technologies such as massive multiple-input multiple-output (MIMO), robust millimeter wave (mmWave) transmission, advanced channel coding, and device-centric mobility. The scalability of the numerology in 5G NR (with scaling of subcarrier spacing) can efficiently address the operation of diverse services across diverse spectrums and diverse deployments. For example, in various outdoor and macro coverage deployments of less than 3 GHz FDD / TDD implementations, the subcarrier spacing can be, for example, 15 kHz on bandwidths of 5, 10, 20 MHz. For other various outdoor and small cell coverage deployments of TDD greater than 3 GHz, the subcarrier spacing can be 30 kHz on 80 / 100 MHz BW. For various other indoor broadband implementations, using TDD on the unlicensed portion of the 5 GHz band, subcarrier spacing may occur at 60 kHz over 160 MHz BW. Finally, for various deployments transmitting with the mmWave component at 28 GHz TDD, subcarrier spacing may occur at 120 kHz over 500 MHz BW.

[0029] 5G NR's scalable digital scheme facilitates scalable TTIs for different latency and quality of service (QoS) requirements. For example, shorter TTIs can be used for low latency and high reliability, while longer TTIs can be used for higher spectral efficiency. Efficient multiplexing of long and short TTIs allows transmissions to start on symbol boundaries. 5G NR also anticipates a self-contained integrated subframe design where UL / downlink scheduling information, data, and acknowledgments are in the same subframe. The self-contained integrated subframe supports communications in unlicensed or contention-based shared spectrum, adaptive UL / downlink (which can be flexibly configured on a per-cell basis to dynamically switch between UL and downlink to meet current traffic needs).

[0030] The various other aspects and features of the present disclosure are further described below. It should be apparent that the teachings herein can be embodied in a variety of forms, and any specific structure, function, or both disclosed herein are representative rather than limiting. Based on the teachings herein, it should be understood by those of ordinary skill in the art that the aspects disclosed herein can be implemented independently of any other aspects, and two or more of these aspects can be combined in various ways. For example, using any number of aspects set forth herein, a device can be implemented or a method can be implemented. In addition, using other structures, functions, or structures and functions other than or different from one or more of the aspects set forth herein, such a device can be implemented, or such a method can be implemented. For example, a method can be implemented as a part of a system, device, apparatus, and / or as instructions stored on a computer-readable medium for execution on a processor or computer. In addition, an aspect can include at least one element of a claim.

[0031] This application describes a mechanism for indicating antenna configuration information and neural network information between devices. According to an embodiment of the present disclosure, two wireless communication devices can use a machine learning / neural network framework to encode transmissions across a channel and decode them when received. For example, two wireless devices can jointly train an autoencoder. The autoencoder can be split between the transmitting side and the receiving side, that is, the encoder and decoder of the autoencoder can be implemented in different devices. When a subsequent device communicates with a device that has obtained one or more neural network parameters from a device with a matching or similar antenna configuration, embodiments of the present disclosure help to speed up the training process.

[0032] For example, a first wireless communication device may establish communication with a second wireless communication device. In one example, the first wireless communication device may be a user equipment (UE), and the second wireless communication device may be a base station (BS, also known as, for example, an evolved Node B or a next-generation eNB); alternatively, the first wireless communication device may be a BS, and the second wireless communication device may be a UE, or both devices may be UEs. For ease of discussion, although it is assumed that the first wireless communication device is a transmitter of information and therefore has an encoder-side residency, and the second wireless communication device is a receiver and therefore has a decoder-side residency, as described above, either device may be a transmitter or a receiver at different times and / or contexts.

[0033] Once communication is established, the first wireless communication device may determine a neural network to use for transmitting some or all types of data to the second wireless communication device. In some embodiments, the first wireless communication device may be pre-configured with a plurality of different neural network configuration options (also referred to herein as artificial intelligence (AI) modules), including different numbers of layers, numbers of nodes in a layer, and algorithms to use. Determining the neural network may include selecting one of the pre-configured options based on one or more antenna configuration parameters of the first wireless communication device. In some other embodiments, the first wireless communication device may not have pre-configured options, but instead may determine the number of layers to use when encoding, the number of nodes per layer, the algorithm to use per layer, etc. based on one or more antenna configuration parameters of the first wireless communication device.

[0034] According to an embodiment of the present disclosure, a first wireless communication device may send neural network information and antenna configuration information together to a second wireless communication device. Thereafter, the two devices jointly train the neural network to achieve an output at the receiving end that is sufficiently similar to the training input at the transmitting end. Upon completion of the training, compression of the input data will produce a codeword of reduced dimensionality, which improves transmission efficiency and reduces resource utilization, and the codeword can be restored on the decoder side. According to some further embodiments of the present disclosure, the first wireless communication device may send training weights from the trained neural network (e.g., in encoded form to reduce resource consumption in the channel) to the second wireless communication device, which may also store the information in association with the antenna configuration information of the first wireless communication device. The second wireless communication device is able to search its records based on the antenna configuration to identify the neural network information and weights.

[0035] Subsequently, when the second wireless communication device establishes communication with a third wireless communication device (e.g., a UE entering a new cell of a new BS, or a new UE entering a cell of a BS), the second wireless communication device may check whether the antenna configuration of the third wireless communication device matches (or passes a similarity threshold) any other antenna configuration information stored at the second wireless communication device. As an example for discussion purposes, sufficient similarity (up to and including a match) may be found between the antenna configuration of the third wireless communication device and the stored antenna configuration information of the first wireless communication device, whose antenna configuration information had previously been sent to the second wireless communication device along with the neural network indication.

[0036] In response, the second wireless communication device can transmit the neural network information received from the first wireless communication device to the third wireless communication device to assist in neural network selection and / or training at the third wireless communication device. Additionally, if the first wireless communication device transmits the neural network weights, the second wireless communication device can also transmit them to the third wireless communication device. The third wireless communication device can implement this information to accelerate the selection and / or training of the neural network used between the third wireless communication device and the second wireless communication device. This can save system resources by not requiring the new device to identify and train a new system (e.g., between the second and third wireless communication devices) based on default parameters.

[0037] Aspects of the present disclosure can provide several benefits. For example, as already mentioned, training a neural network can be a time-intensive and computationally intensive process, consuming significant system resources. If a deep neural network has already been used and trained for a particular antenna configuration pair, the same neural network can be used for future communication purposes between the second wireless communication device and other wireless communication devices. Thus, system resources can be saved by not identifying and training a new system from scratch (e.g., in this example, between the second wireless communication device and the third wireless communication device).

[0038] Figure 1 A wireless communication network 100 according to some aspects of the present disclosure is shown. The network 100 may be a 5G network. The network 100 includes a plurality of base stations (BSs) 105 (labeled 105a, 105b, 105c, 105d, 105e, and 105f, respectively) and other network entities. The BSs 105 may be stations that communicate with the UEs 115 and may also be referred to as evolved Node Bs (eNBs), next generation eNBs (gNBs), access points, and the like. Each BS 105 may provide communication coverage for a particular geographic area. In 3GPP, the term "cell" may refer to that particular geographic coverage area of ​​the BS 105 and / or a BS subsystem serving that coverage area, depending on the context in which the term is used.

[0039] The BS 105 may provide communication coverage for macro cells or small cells (e.g., pico cells or femto cells) and / or other types of cells. A macro cell typically covers a relatively large geographic area (e.g., a radius of several kilometers) and may allow unrestricted access by UEs with a service subscription with the network provider. A small cell (e.g., pico cell) will typically cover a relatively small geographic area and may allow unrestricted access by UEs with a service subscription with the network provider. A small cell (e.g., femto cell) will also typically cover a relatively small geographic area (e.g., a residence) and, in addition to unrestricted access, may also provide restricted access by UEs associated with the femto cell (e.g., UEs in a closed subscriber group (CSG), UEs for users in a residence, etc.). A BS for a macro cell may be referred to as a macro BS. A BS for a small cell may be referred to as a small cell BS, pico BS, femto BS, or home BS. In Figure 1 In the example shown in FIG, BS 105d and 105e may be conventional macro BSs, while BS 105a-105c may be macro BSs implemented using one of three-dimensional (3D), full-dimensional (FD), or massive MIMO. BS 105a-105c may utilize their higher-dimensional MIMO capabilities to utilize 3D beamforming in both elevation and azimuth beamforming to increase coverage and capacity. BS 105f may be a small cell BS, which may be a home node or a portable access point. BS 105 may support one or more (e.g., two, three, four, etc.) cells.

[0040] Network 100 may support synchronous operation or asynchronous operation. For synchronous operation, the BSs may have similar frame timing, and transmissions from different BSs may be approximately aligned in time. For asynchronous operation, the BSs may have different frame timing, and transmissions from different BSs may not be aligned in time.

[0041] UEs 115 are dispersed throughout wireless network 100, and each UE 115 may be stationary or mobile. UEs 115 may also be referred to as terminals, mobile stations, subscriber units, stations, and the like. UEs 115 may be cellular phones, personal digital assistants (PDAs), wireless modems, wireless communication devices, handheld devices, tablet computers, laptop computers, cordless phones, wireless local loop (WLL) stations, and the like. In one aspect, UEs 115 may be devices that include a Universal Integrated Circuit Card (UICC). In another aspect, UEs may be devices that do not include a UICC. In some aspects, UEs 115 that do not include a UICC may also be referred to as IoT devices or Internet of Everything (IoE) devices. UEs 115a-115d are examples of mobile smartphone-type devices that access network 100. UEs 115 may also be machines specifically configured for connected communications, including machine-type communications (MTC), enhanced MTC (eMTC), narrowband IoT (NB-IoT), and the like. UEs 115e-115h are examples of various machines configured for communication that access the network 100. UEs 115i-115k are examples of vehicles equipped with wireless communication devices configured for communication that access the network 100. UE 115 may be able to communicate with any type of BS (whether macro BS, small cell, etc.). Figure 1 In the figure, lightning (e.g., communication link) indicates wireless transmission between UE 115 and serving BS 105 (which is a BS designated to serve UE 115 on downlink (DL) and / or uplink (UL)), desired transmission between BSs 105, backhaul transmission between BSs, or sidelink transmission between UE 115.

[0042] In operation, BSs 105a-105c use 3D beamforming and coordinated spatial techniques (e.g., coordinated multipoint (CoMP) or multi-connectivity) to serve UEs 115a and 115b. Macro BS 105d can perform backhaul communications with BSs 105a-105c and small cell BS 105f. Macro BS 105d also transmits multicast services that UEs 115c and 115d subscribe to and receive. Such multicast services may include mobile television or streaming video, or may include other services for providing community information, such as weather emergencies or alerts (e.g., Amber Alerts or Gray Alerts).

[0043] The BSs 105 may also communicate with a core network. The core network may provide user authentication, access authorization, tracking, Internet Protocol (IP) connectivity, and other access, routing, or mobility functions. At least some of the BSs 105 (e.g., which may be examples of gNBs or access node controllers (ANCs)) may interface with the core network via a backhaul link (e.g., NG-C, NG-U, etc.) and may perform radio configuration and scheduling for communications with the UEs 115. In various examples, the BSs 105 may communicate with each other directly or indirectly (e.g., through the core network) over a backhaul link (e.g., X1, X2, etc.), which may be a wired or wireless communication link.

[0044] The network 100 may also support mission-critical communications using ultra-reliable and redundant links for mission-critical devices (e.g., UE 115e, which may be a drone). The redundant communication links with UE 115e may include links from macro BSs 105d and 105e, as well as a link from small cell BS 105f. Other machine-type devices (e.g., UE 115f (e.g., a thermometer), UE 115g (e.g., a smart meter), and UE 115h (e.g., a wearable device)) may communicate directly with a BS (e.g., small cell BS 105f and macro BS 105e) via the network 100, or in a multi-step configuration by communicating with another user device that relays its information to the network (e.g., UE 115f transmits temperature measurement information to the smart meter (UE 115g), which is then reported to the network via the small cell BS 105f). The network 100 may also provide additional network efficiency through dynamic, low-latency TDD / FDD communications, such as V2V, V2X, C-V2X communications between UE 115i, 115j, or 115k and other UEs 115 and / or vehicle-to-infrastructure (V2I) communications between UE 115i, 115j, or 115k and BS 105.

[0045] In some implementations, network 100 utilizes an OFDM-based waveform for communication. An OFDM-based system can divide the system BW into multiple (K) orthogonal subcarriers, which are also commonly referred to as subcarriers, tones, frequency bands, etc. Each subcarrier can be modulated with data. In some examples, the subcarrier spacing between adjacent subcarriers can be fixed, and the total number of subcarriers (K) can depend on the system BW. The system BW can also be divided into subbands. In other examples, the subcarrier spacing and / or the duration of the TTI can be scalable.

[0046] In some aspects, BS 105 may assign or schedule transmission resources (e.g., in the form of time-frequency resource blocks (RBs)) for downlink (DL) and uplink (UL) transmissions in network 100. DL refers to the transmission direction from BS 105 to UE 115, while UL refers to the transmission direction from UE 115 to BS 105. Communication may be in the form of radio frames. A radio frame may be divided into multiple subframes or time slots, e.g., approximately 10. Each time slot may be further divided into micro-slots. In FDD mode, simultaneous UL and DL transmissions may occur in different frequency bands. For example, each subframe includes a UL subframe in a UL frequency band and a DL subframe in a DL frequency band. In TDD mode, UL and DL transmissions occur at different time periods using the same frequency band. For example, a subset of subframes in a radio frame (e.g., DL subframes) may be used for DL ​​transmissions, and another subset of subframes in a radio frame (e.g., UL subframes) may be used for UL transmissions.

[0047] DL subframes and UL subframes can also be divided into several areas. For example, each DL or UL subframe can have a predefined area for the transmission of reference signals, control information, and data. Reference signals are predetermined signals that facilitate communication between BS 105 and UE 115. For example, reference signals can have a specific pilot pattern or structure, wherein pilot tones can span across an operational BW or frequency band, with each pilot tone located at a predefined time and a predefined frequency. For example, BS 105 can transmit a cell-specific reference signal (CRS) and / or a channel state information-reference signal (CSI-RS) to enable UE 115 to estimate the DL channel. Similarly, UE 115 can transmit a sounding reference signal (SRS) to enable BS 105 to estimate the UL channel. Control information can include resource assignments and protocol control. Data can include protocol data and / or operational data. In some aspects, BS 105 and UE 115 can communicate using self-contained subframes. Self-contained subframes can include a portion for DL ​​communication and a portion for UL communication. A self-contained subframe may be DL-centric or UL-centric. A DL-centric subframe may include a longer duration for DL ​​communications than for UL communications. A UL-centric subframe may include a longer duration for UL communications than for UL communications.

[0048] In some aspects, network 100 may be an NR network deployed on a licensed spectrum. BS 105 may transmit synchronization signals (e.g., including a primary synchronization signal (PSS) and a secondary synchronization signal (SSS)) in network 100 to facilitate synchronization. BS 105 may broadcast system information associated with network 100 (e.g., including a master information block (MIB), residual system information (RMSI), and other system information (OSI)) to facilitate initial network access. In some cases, BS 105 may broadcast the PSS, SSS, and / or MIB in the form of synchronization signal blocks (SSBs) on a physical broadcast channel (PBCH), and may broadcast the RMSI and / or OSI on a physical downlink shared channel (PDSCH).

[0049] In some aspects, a UE 115 attempting to access the network 100 may perform an initial cell search by detecting the PSS from the BS 105. The PSS may enable synchronization of period timing and may indicate a physical layer identification value. Subsequently, the UE 115 may receive the SSS. The SSS may enable radio frame synchronization and may provide a cell identification value that may be combined with the physical layer identification value to identify a cell. The PSS and SSS may be located in the center portion of a carrier or at any suitable frequency within the carrier.

[0050] After receiving the PSS and SSS, the UE 115 may receive the MIB. The MIB may include system information for initial network access and scheduling information for RMSI and / or OSI. After decoding the MIB, the UE 115 may receive the RMSI and / or OSI. The RMSI and / or OSI may include radio resource control (RRC) information related to the random access channel (RACH) procedure, paging, control resource set (CORESET) for physical downlink control channel (PDCCH) monitoring, physical UL control channel (PUCCH), physical UL shared channel (PUSCH), power control, and SRS.

[0051] After obtaining the MIB, RMSI, and / or OSI, UE 115 may perform a random access procedure to establish a connection with BS 105. In some examples, the random access procedure may be a four-step random access procedure. For example, UE 115 may send a random access preamble, and BS 105 may respond with a random access response. The random access response (RAR) may include a detected random access preamble identifier (ID) corresponding to the random access preamble, timing advance (TA) information, an UL grant, a temporary cell radio network temporary identifier (C-RNTI), and / or a fallback indicator. Upon receiving the random access response, UE 115 may send a connection request to BS 105, and BS 105 may respond with a connection response. The connection response may indicate contention resolution. In some examples, the random access preamble, RAR, connection request, and connection response may be referred to as message 1 (MSG1), message 2 (MSG2), message 3 (MSG3), and message 4 (MSG4), respectively. In some examples, the random access procedure may be a two-step random access procedure, where the UE 115 may send a random access preamble and a connection request in a single transmission, and the BS 105 may respond by sending a random access response and a connection response in a single transmission.

[0052] After establishing the connection, the UE 115 and the BS 105 may enter a normal operation phase, in which operational data may be exchanged. For example, the BS 105 may schedule the UE 115 for UL and / or DL ​​communications. The BS 105 may send an UL and / or DL ​​scheduling grant to the UE 115 via the PDCCH. The scheduling grant may be sent in the form of DL control information (DCI). The BS 105 may send a DL communication signal (e.g., carrying data) to the UE 115 via the PDSCH based on the DL scheduling grant. The UE 115 may send an UL communication signal to the BS 105 via the PUSCH and / or PUCCH based on the UL scheduling grant.

[0053] In some aspects, BS 105 may use HARQ technology to communicate with UE 115 to improve communication reliability, for example, to provide URLLC services. BS 105 may schedule UE 115 for PDSCH communication by sending a DL grant in the PDCCH. BS 105 may send DL data packets to UE 115 according to the schedule in the PDSCH. The DL data packets may be sent in the form of transport blocks (TBs). If UE 115 successfully receives the DL data packet, UE 115 may send a HARQ ACK to BS 105. Conversely, if UE 115 fails to successfully receive the DL transmission, UE 115 may send a HARQ NACK to BS 105. Upon receiving a HARQ NACK from UE 115, BS 105 may retransmit the DL data packet to UE 115. The retransmission may include a decoded version of the same DL data as the initial transmission. Alternatively, the retransmission may include a decoded version of the DL data that is different from the initial transmission. UE 115 may apply soft combining to combine the coded data received from the initial transmission and the retransmission for decoding.BS 105 and UE 115 may also apply HARQ for UL communications using substantially similar mechanisms as DL HARQ.

[0054] In some aspects, the network 100 may operate on a system BW or a component carrier (CC) BW. The network 100 may divide the system BW into multiple BWPs (e.g., portions). The BS 105 may dynamically assign the UE 115 to operate on a specific BWP (e.g., a specific portion of the system BW). The assigned BWP may be referred to as an active BWP. The UE 115 may monitor the active BWP for signaling information from the BS 105. The BS 105 may schedule the UE 115 to perform UL or DL ​​communications in the active BWP. In some aspects, the BS 105 may assign a pair of BWPs within a CC to the UE 115 for UL and DL communications. For example, the BWP pair may include one BWP for UL communications and one BWP for DL ​​communications.

[0055] In some aspects, network 100 may operate on a shared channel, which may include a shared frequency band or an unlicensed frequency band. For example, network 100 may be an NR license-free (NR-U) network operating on an unlicensed frequency band. In such aspects, BS 105 and UE 115 may be operated by multiple network operating entities. To avoid conflicts, BS 105 and UE 115 may use a listen-before-talk (LBT) process to monitor transmission opportunities (TXOPs) in a shared channel. For example, a transmitting node (e.g., BS 105 or UE 115) may perform LBT before transmitting in a channel. When LBT passes, the transmitting node may continue transmitting. When LBT fails, the transmitting node may avoid transmitting in the channel. In one example, LBT may be based on energy detection. For example, when the signal energy measured from the channel is below a threshold, LBT results in a pass. Conversely, when the signal energy measured from the channel exceeds a threshold, LBT results in a fail. In another example, LBT may be based on signal detection. For example, when a channel reservation signal (eg, a predetermined preamble signal) is not detected in the channel, LBT results in a pass. TXOP may also be referred to as a channel occupation time (COT).

[0056] In some aspects, the network 100 may provide sidelink communications to allow a UE 115 to communicate with another UE 115 without tunneling through the BS 105 and / or the core network. The BS 105 may configure certain resources in a licensed frequency band and / or an unlicensed frequency band for sidelink communications between the UE 115 and the other UE 115. During sidelink communications, the UE 115 may transmit physical sidelink shared channel (PSSCH) data, physical sidelink shared control channel (PSCCH) sidelink control information (SCI), sidelink COT shared SCI, sidelink scheduling SCI, and / or physical sidelink feedback channel (PSFCH) ACK / NACK feedback (e.g., for sidelink HARQ) to the other UE and / or receive PSSCH data, PSCCH SCI, sidelink COT shared SCI, sidelink scheduling SCI, and / or PSFCH ACK / NACK feedback from the other UE 115.

[0057] UE 115 (whether transmitting to BS 105 or to another UE 115) can implement artificial intelligence (AI) (such as machine learning (ML) and / or deep learning (DL)) to assist in communication, thereby achieving one or more benefits, including, for example, not requiring knowledge of the underlying data distribution or explicit identification of a certain structure to work, and higher compression efficiency. One example of such an implementation is an autoencoder that is split between the transmitting and receiving sides, that is, the encoder and decoder of the autoencoder can be implemented in different devices. In the example of UE 115 communicating with BS 105, the encoder can be located at UE 115, and BS 105 can have a decoder, which work together as an autoencoder to train and achieve end-to-end compression and reconstruction of data. In another example, BS 105 can be replaced by another UE 115, or BS 105 can be the source of the transmission to another device and therefore have an encoder, etc. More generally, in an autoencoder implementation, a first wireless communication device transmitting to another device can act as an encoder, and a second wireless communication device receiving from the first device can act as a decoder. The data sent after compression by an encoder may be referred to herein as encoded data, compressed data, code, etc.

[0058] Because training a neural network used in an autoencoder (between a first wireless communication device and a second wireless communication device) can consume significant resources, including computational resources and / or time resources (to name a few), it is desirable to mitigate this consumption. Figure 1 When the first wireless communication device trains a neural network for subsequent communication with the second wireless communication device, the first wireless communication device may transmit the neural network information to the second wireless communication device along with the antenna configuration information of the first wireless communication device.

[0059] In some embodiments, the first wireless communication device may be provided (e.g., previously, e.g., statically or dynamically, e.g., via RRC signaling, MAC signaling, or other signaling) a set of potential neural network configurations from which to select, also referred to herein for simplicity as AI modules. For example, a given AI module may specify different neural network characteristics, including, for example, the number of layers to be used in the neural network on the encoder side, the number of nodes to be used in each layer on the encoder side, etc. In other embodiments, the first wireless communication device may not be provided with potential neural network configurations; instead, the first wireless communication device may dynamically select the number of layers, the number of nodes per layer, etc. to use for encoding. Under any of the above methods, the first wireless communication device may select a neural network based on the antenna configuration of the first wireless communication device.

[0060] By sending neural network information to the second wireless communication device, the first wireless communication device and the second wireless communication device can train the neural network for subsequent use. This can be a collaborative and iterative process to achieve the same output as the input training sequence. For example, the first wireless communication device can modify one or more training weights and / or biases of one or more nodes in one or more layers of the neural network at the first wireless communication device.

[0061] According to some embodiments of the present disclosure, when a first wireless communication device collaborates with a second wireless communication device to complete training, the first wireless communication device may send one or more weights and / or one or more biases of the corresponding node generated by the training to the second wireless communication device. This may be sent without compression, or alternatively, after compression by the neural network after training. This information may also be maintained in association with the antenna configuration information for the first wireless communication device. The second wireless communication device may be able to search its records based on the antenna configuration to identify the neural network information and weights (or biases).

[0062] For example, if a second wireless communication device establishes communication with a third wireless communication device (e.g., BS 105 establishes communication with another UE 115, or UE 115 establishes communication with another BS 105 due to a change in location, etc.), the second wireless communication device can use the neural network configuration information and / or training bias to reduce the computational and / or time burden on the third wireless communication device. To do so, the second wireless communication device can first determine whether the antenna configuration information from the third wireless communication device corresponds to the antenna configuration information from another device that previously collaborated with the second wireless communication device to train a neural network. This can include, for example, determining whether the antenna configuration of the third wireless communication device is the same as the antenna configuration of the first wireless communication device, or whether it passes a similarity threshold, etc.

[0063] When the antenna configurations are sufficiently similar, the second wireless communication device may send neural network configuration information (e.g., information identifying the number of layers, number of nodes, etc.) to the third wireless communication device. For example, if the second wireless communication device is BS 105, this may be an instruction to use the same neural network information; if the second wireless communication device is UE 115 and the third wireless communication device is BS 105, this may be a suggestion. Along with the neural network information, the second wireless communication device may also send antenna configuration information to the third wireless communication device, including, for example, panel orientation, antenna array dimensions, antenna polarization, panel position, etc.

[0064] Furthermore, in some embodiments, when the antenna configurations are sufficiently similar, the second wireless communication device may also send trained weights / biases to the third wireless communication device. This weight information may be used at the third wireless communication device as a starting point for neural network training between the second and third wireless communication devices. Thus, if a deep neural network has already been used and trained for a certain antenna configuration pair (e.g., in this example, between the first and second wireless communication devices), the same neural network may be used for future communication purposes (e.g., in this example, between the second and third wireless communication devices). This may occur, for example, when the UE 115 moves to another cell with the same or similar antenna configuration, or when the BS 105 serves a new UE 115 with the same or similar antenna configuration as the old UE 115 for which the neural network information was used and trained. In this way, system resources may be saved by not identifying and training a new system from scratch (e.g., in this example, between the second and third wireless communication devices).

[0065] exist Figure 2 An example of an auto-encoder configuration 200 between wireless communication devices is shown in FIG. The auto-encoder configuration 200 may include an encoder 202 and a decoder 208 that communicate over a channel 206. The encoder 202 may be part of a first wireless communication device, and the decoder 208 may be part of a second wireless communication device. Figure 1 In the example of , the encoder 202 can be part of the UE 115 and communicate with the decoder 208, which is part of the BS 105 or another UE 115 in various examples. As another example, the encoder 202 can be part of the BS 105 and communicate with the decoder 208, which is part of the UE 115.

[0066] The encoder 202 may include one or more layers 204a, which include, for example, an input layer and one or more hidden layers. Each layer may include one or more nodes connected to nodes in another layer. Weights may be applied to the data as it passes through the one or more layers 204a. The encoder 202 may also include a normalization layer 204b. The decoder 208 may include one or more hidden layers 210a, followed by an activation layer 210b. For example, an input S may pass through the layer 204a and the normalization layer 204b, thereby generating a compressed codeword A. The compressed codeword A has a reduced dimension compared to the dimension of the input S. As the codeword A passes through the channel 206, it arrives at the receiving decoder 208 as the received codeword B. The hidden layers 210a and activation layers 210b can reconstruct the original input S into S'. The encoder 202 and decoder 208 are jointly trained to restore the input S at the output to S'.

[0067] Configuration 200 is exemplary; when implementing embodiments of the present disclosure, different numbers of layers may be included in an encoder or decoder, as well as additional layers and / or functionality and / or algorithms for implementing different types of neural networks. For example, according to some embodiments of the present disclosure, a first wireless communication device implementing encoder 202 may store multiple different AI modules, each having a different neural network configuration (e.g., layers 204a, 204b and the number of nodes in each layer, etc.). The AI ​​modules may be pre-configured, and in some examples may also be dynamically updated (e.g., via RRC signaling or other control signaling). In other examples, encoder 202 may not be configured with an AI module, but instead may determine the neural network parameters to be implemented on its own.

[0068] The AI ​​module (or more generally, neural network parameters) may be determined at the encoder 202 based on one or more parameters of the first wireless communication device housing the encoder 202. For example, the neural network parameters may be determined based on antenna configuration parameters at the first wireless communication device (including, for example, antenna array dimensions (including, for example, the number of antennas in the array), antenna polarization, panel orientation, combinations of parameters, etc.).

[0069] Figure 3 is a block diagram of an exemplary UE 300 according to some aspects of the present disclosure. The UE 300 may be Figure 1 1 . As shown, UE 300 may include a processor 302, a memory 304, a neural network communication module 308, a transceiver 310 including a modem subsystem 312 and a radio frequency (RF) unit 318, and one or more antennas 320. These elements may communicate with each other directly or indirectly, for example, via one or more buses.

[0070] The processor 302 may include a central processing unit (CPU), a digital signal processor (DSP), an application specific integrated circuit (ASIC), a controller, a field programmable gate array (FPGA) device, another hardware device, a firmware device, or any combination thereof configured to perform the operations described herein. The processor 302 may also be implemented as a combination of computing devices, for example, a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors in combination with a DSP core, or any other such configuration.

[0071] Memory 304 may include cache memory (e.g., cache memory of processor 302), random access memory (RAM), magnetoresistive RAM (MRAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), flash memory, a solid-state memory device, a hard drive, other forms of volatile and non-volatile memory, or a combination of different types of memory. In one aspect, memory 304 includes a non-transitory computer-readable medium. Memory 304 may store or have recorded thereon instructions 306. Instructions 306 may include, when executed by processor 302, causing processor 302 to perform aspects described herein in conjunction with the present disclosure (e.g., Figure 5-7B Instructions 306 may also be referred to as program code. Program code may be used to cause a wireless communication device to perform operations, such as by causing one or more processors (e.g., processor 302) to control or command the wireless communication device to do so. The terms "instructions" and "code" should be broadly interpreted to include any type of computer-readable statements. For example, the terms "instructions" and "code" may refer to one or more programs, routines, subroutines, functions, procedures, and the like. "Instructions" and "code" may include a single computer-readable statement or many computer-readable statements.

[0072] The neural network communication module 308 can be implemented via hardware, software, or a combination thereof. For example, the neural network communication module 308 can be implemented as a processor, circuitry, and / or instructions 306 stored in memory 304 and executed by processor 302. In some cases, the neural network communication module 308 can be integrated within the modem subsystem 312. For example, the neural network communication module 308 can be implemented as a neural network unit 314, which can implement one or more machine learning algorithms, shown as AI modules 316a-316n. In other examples, the neural network communication module 308 can be implemented by offloading neural network processing to an application processor, such as processor 302. Furthermore, the neural network communication module 308 can be implemented solely by a combination of software components (e.g., executed by a DSP or general-purpose processor) and hardware components (e.g., logic gates and circuits) within the modem subsystem 312, or in conjunction with an application processor, such as processor 302, or solely by an application processor, such as processor 302.

[0073] The neural network communication module 308 can be used in various aspects of the present disclosure, for example, Figure 5-7BAspects of the present disclosure. The neural network communication module 308 is configured to perform neural network configuration, training and / or compression using a trained neural network. In some examples, the neural network communication module 308 can operate as an encoder side in conjunction with a decoder side on a receiving end. In other examples, the neural network communication module 308 can operate as a decoder side in conjunction with an encoder side on a transmitting end. The neural network communication module 308 can determine the neural network parameters to be used based on the configuration of the antenna 320 of the UE 115 (according to a pre-configured AI module or generally). According to an embodiment of the present disclosure, the neural network communication module 308 can also signal the decoder side of the receiving end with antenna configuration information and neural network information and / or training weights. The signaled information can be used on the decoder side to speed up and reduce the computational burden of a new device that performs neural network training with the decoder side. In some examples, the neural network communication module 308 can be used as part of an encoder of a first wireless communication device (for example, according to a description of the present disclosure). Figure 1 In other examples, the neural network communication module 308 can be used as a Figure 1 A portion of an encoder of an example third wireless communication device that receives and implements previously provided neural network parameters and / or weights.

[0074] As shown, the transceiver 310 may include a modem subsystem 312 and an RF unit 318. The transceiver 310 may be configured to communicate bidirectionally with other devices, such as the BS 105 and other UEs 115. The modem subsystem 312 may be configured to modulate and / or encode data from the memory 304 and / or the neural network communication module 308 (whether implemented by the processor 302 and / or the neural network unit 314) according to a modulation and coding scheme (MCS) (e.g., a low-density parity check (LDPC) coding scheme, a turbo coding scheme, a convolutional coding scheme, a digital beamforming scheme, etc.). As described above, the modem subsystem 312 may include a neural network unit 314. The neural network unit 314 may include one or more AI modules 316a-316n, or implement parameters without an AI module.

[0075] The RF unit 318 can be configured to process (e.g., perform analog-to-digital conversion or digital-to-analog conversion, etc.) modulated / coded data from the modem subsystem 312 (for outbound transmissions) or transmissions originating from another source (such as the UE 115 or the BS 105). The RF unit 318 can also be configured to perform analog beamforming in conjunction with digital beamforming. Although shown as integrated with the transceiver 310, the modem subsystem 312 and the RF unit 318 can be separate devices that are coupled together at the UE 115 to enable the UE 115 to communicate with other devices.

[0076] The RF unit 318 may provide modulated and / or processed data (e.g., a data packet (or, more generally, a data message that may include one or more data packets and other information)) to the antenna 320 for transmission to one or more other devices. The antenna 320 may also receive data messages sent from other devices. The antenna 320 may provide the received data message for processing and / or demodulation at the transceiver 310. The antenna 320 may include multiple antennas of similar or different designs in order to maintain multiple transmission links. The RF unit 318 may configure the antenna 320.

[0077] In one aspect, the UE 300 may include multiple transceivers 310 that implement different RATs (e.g., NR and LTE). In one aspect, the UE 300 may include a single transceiver 310 that implements multiple RATs (e.g., NR and LTE). In one aspect, the transceiver 310 may include various components, where different combinations of components may implement different RATs.

[0078] Figure 4 is a block diagram of an exemplary BS 400 according to some aspects of the present disclosure. BS 400 may be Figure 1 4. BS 400 is a BS 105 in the network 100 as discussed above. As shown, BS 400 may include a processor 402, a memory 404, a neural network communication module 408, a transceiver 410 including a modem subsystem 412 and an RF unit 418, and one or more antennas 420. These elements may communicate with each other directly or indirectly, for example, via one or more buses.

[0079] The processor 402 may have various features specific to the type of processor. For example, these may include a CPU, a DSP, an ASIC, a controller, an FPGA device, another hardware device, a firmware device, or any combination thereof configured to perform the operations described herein. The processor 402 may also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors in combination with a DSP core, or any other such configuration.

[0080] Memory 404 may include cache memory (e.g., cache memory of processor 402), RAM, MRAM, ROM, PROM, EPROM, EEPROM, flash memory, solid-state memory devices, one or more hard disk drives, memristor-based arrays, other forms of volatile and non-volatile memory, or a combination of different types of memory. In some embodiments, memory 404 may include non-transitory computer-readable media. Memory 404 may store instructions 406. Instructions 406 may include instructions that, when executed by processor 402, cause processor 402 to perform the operations described herein. Instructions 406 may also be referred to as code, which may be broadly interpreted to include any type of computer-readable statements, as described above with respect to Figure 3 Discussed.

[0081] The neural network communication module 408 can be implemented via hardware, software, or a combination thereof. For example, the neural network communication module 408 can be implemented as a processor, circuitry, and / or instructions 406 stored in memory 404 and executed by the processor 402. In some cases, the neural network communication module 408 can be integrated within the modem subsystem 412. For example, the neural network communication module 408 can be implemented as a neural network unit 414, which can implement one or more machine learning algorithms, shown as AI modules 416a-416n. In other examples, the neural network communication module 408 can be implemented by offloading neural network processing to an application processor, such as the processor 402. Furthermore, the neural network communication module 408 can be implemented solely by a combination of software components (e.g., executed by a DSP or general-purpose processor) and hardware components (e.g., logic gates and circuits) within the modem subsystem 412, or in conjunction with an application processor, such as the processor 402, or solely by an application processor, such as the processor 402.

[0082] The neural network communication module 408 can be used in various aspects of the present disclosure, for example, Figure 5-7BThe neural network communication module 408 is configured to perform neural network configuration, training, decompression, and / or compression (depending on whether the neural network communication module 408 operates as a decoder or encoder with the BS 400). In some examples, the neural network communication module 408 can operate as a decoder side in conjunction with an encoder side on the transmitting end. In other examples, the neural network communication module 408 can operate as an encoder side in conjunction with a decoder side on the receiving end.

[0083] The neural network configuration module 408 can receive neural network parameters used on the encoding side (from a pre-configured AI module or generally) based on the configuration of the antennas on the encoding side. According to an embodiment of the present disclosure, the neural network communication module 408 can receive antenna configuration information and neural network information and / or training weights from the encoder side. When operating on the decoder side, the neural network communication module 408 can use the signaled information to speed up and reduce the computational burden of new devices performing neural network training with the decoder side. In some examples, the neural network communication module 408 can be used as part of a decoder of a second wireless communication device (e.g., based on information about Figure 1 In other examples, according to Figure 1 For example, the neural network communication module 308 can be used as part of an encoder of a first wireless communication device or a third wireless communication device, which determines and implements or receives and implements neural network parameters and / or weights, respectively.

[0084] As shown, the transceiver 410 may include a modem subsystem 412 and an RF unit 418. The transceiver 410 may be configured to communicate bidirectionally with other devices (such as UE 115 and / or 300 and / or another core network element). The modem subsystem 412 may be configured to modulate and / or encode data according to an MCS (e.g., an LDPC coding scheme, a turbo coding scheme, a convolutional coding scheme, a digital beamforming scheme, etc.). The modem subsystem 412 may include a neural network unit 414 (which may include one or more AI modules 416a-416n) and / or a storage device (e.g., memory 404) for storing antenna configuration information and neural network configuration information. The RF unit 418 may be configured to process (e.g., perform analog-to-digital conversion or digital-to-analog conversion, etc.) the modulated / encoded data from the modem subsystem 412 (for outbound transmissions) or transmissions originating from another source (such as UE 115 and / or 300). The RF unit 418 can also be configured to perform analog beamforming in conjunction with digital beamforming, for example, after processing codewords including channel state information (wherein a neural network is used on the transmit side to reduce its dimensionality) to estimate the channel. Although shown as integrated with the transceiver 410, the modem subsystem 412 and / or the RF unit 418 can be separate devices that are coupled together at the BS 400 to enable the BS 400 to communicate with other devices.

[0085] The RF unit 418 may provide modulated and / or processed data (e.g., data packets (or more generally, data messages containing one or more data packets and other information)) to the antenna 420 for transmission to one or more other devices. The antenna 420 may also receive data messages sent from other devices and provide the received data messages for processing and / or demodulation at the transceiver 410. The antenna 420 may include multiple antennas of similar or different designs to maintain multiple transmission links. In one aspect, the BS 400 may include multiple transceivers 410 that implement different RATs (e.g., NR and LTE). In one aspect, the BS 400 may include a single transceiver 410 that implements multiple RATs (e.g., NR and LTE). In one aspect, the transceiver 410 may include various components, where different combinations of components may implement different RATs.

[0086] Now go to Figure 5, which illustrates a protocol diagram of a neural network and antenna configuration indication scheme 500 according to some embodiments of the present disclosure. Communication may be performed between a first wireless communication device 502, a second wireless communication device 504, and / or a third wireless communication device 506. For example, the first wireless communication device 502 may be a UE, the second wireless communication device 504 may be a base station, and the third wireless communication device 506 may be another UE. As another example, the first wireless communication device 502 may be a base station, the second wireless communication device 504 may be a UE, and the third wireless communication device 506 may be another base station. In another example, all three wireless communication devices may be UEs.

[0087] At action 510, the first wireless communication device 502 and the second wireless communication device 504 establish communication with each other. This may include one or more of the aspects of establishing communication, as described above with respect to, for example, Figure 1 discussed.

[0088] At action 512, the first wireless communication device 502 (at Figure 5 is described as a source that uses a neural network for transmission and therefore includes an encoder, such as Figure 2 The encoder 202 discussed above determines a neural network to be used in autoencoder communications with the second wireless communication device 504. Figure 3 (wherein the first wireless communication device 502 is a UE) or Figure 4 (Where the first wireless communication device 502 is a BS) As further discussed, this can be performed by the neural network communication module 308 / 408 (respectively). This can be based on, for example, the antenna configuration at the first wireless communication device 502, including, for example, antenna array dimensions (including, for example, the number of antennas in the array), antenna polarization, panel orientation, combinations of parameters, etc.

[0089] In some embodiments, this involves the first wireless communication device 502 selecting an AI module from a plurality of different AI modules that may be configured at the first wireless communication device 502 (e.g., pre-installed on the device or received from the network via one or more configuration updates, and / or some combination thereof). For example, the first wireless communication device 502 may be pre-configured with a set of AI modules, each having different neural network parameters, such as the number of layers, the number of nodes per layer, the machine learning algorithm (overall and / or per layer, etc.). The possible AI modules may be pre-configured at the second wireless communication device 504 (or may have been the source of pre-configuration at the first wireless communication device 502, where, for example, the second wireless communication device 504 is the BS 105).

[0090] In other embodiments, this involves the first wireless communication device 502 not being provided with a pre-configured AI module. In this case, the first wireless communication device 502 can dynamically determine the number of layers, the number of nodes per layer, etc. to use when encoding. Thus, each aspect of the neural network to be used can be selected. Alternatively, the first wireless communication device can be provided with pre-configured layer options, but not the number of nodes per layer, so that the first wireless communication device 502 can select from the pre-configured layer options and dynamically determine the number of nodes. Alternatively, the node number option can be pre-configured, but the number of layers can be dynamically determined by the first wireless communication device 502.

[0091] In the case where the neural network is determined, at action 514 , the first wireless communication device 502 transmits the neural network information together with the antenna configuration information of the first wireless communication device 502 to the second wireless communication device 504 .

[0092] For example, where action 512 involves selecting from a plurality of pre-configured AI modules, action 514 may involve the first wireless communication device 502 sending an index identifying the selected AI module (e.g., implicit signaling of neural network parameters) along with the antenna configuration information to the second wireless communication device 504. As described above, in these cases, the second wireless communication device 504 may also use the index to identify the AI ​​module that has already been provided at the second wireless communication device 504. Thus, the second wireless communication device 504 may know the relevant neural network parameters via the AI ​​module. This helps reduce signaling overhead when sending the neural network information indication along with the antenna configuration information, thereby improving efficiency.

[0093] As another example, where act 512 involves determining explicit neural network parameters, act 514 may instead involve the first wireless communication device 502 sending the explicit neural network parameters (e.g., explicit signaling of the parameters) along with the antenna configuration information to the second wireless communication device 504. This may involve explicitly signaling the number of layers of the neural network, the number of nodes per layer, the algorithm or algorithms used, etc. Alternatively, where some parameters may be pre-provisioned and others are not, this may involve implicitly signaling the pre-provisioned portions and explicitly signaling the non-pre-provisioned portions.

[0094] In an example where the first wireless communication device 502 is a UE 115, the transmission 514 may be accomplished on uplink resources such as an uplink RRC message, MAC-CE, or PUCCH. In an example where the first wireless communication device 502 is a BS 105, the transmission 514 may be accomplished on downlink resources such as RRC, MAC-CE, or PDCCH (where RRC signaling is less dynamic and PDCCH is more dynamic). In an example where both the first wireless communication device 502 and the second wireless communication device 504 are UEs 115, the transmission 514 may be accomplished via sidelink control or sidelink higher layer signaling.

[0095] At action 516, the second wireless communication device 504 stores the received neural network information and corresponding antenna configuration information. This can be stored in a searchable manner so that the second wireless communication device 504 can search based on the antenna configuration parameter (or parameters) to see if the antenna configuration of the new wireless communication device is the same as at least one parameter, the same as multiple parameters, or within a similarity threshold of one or more antenna configuration parameters of a given entry in the storage of the second wireless communication device 504. This can be stored in a memory at the second wireless communication device 504, while in other examples, it can be stored in a memory accessible to the second wireless communication device 504 via a network.

[0096] At act 518, the first wireless communication device 502 and the second wireless communication device 504 participate in training the neural network determined at act 512. As previously mentioned, this can be a collaborative and iterative process in which the first wireless communication device 502 and the second wireless communication device jointly train the neural network to achieve an output that is the same as or sufficiently similar to the input training sequence. For example, the first wireless communication device 502 can modify one or more training weights and / or biases of one or more nodes in one or more layers of the neural network at the first wireless communication device 502. Upon completion of training, compression of the input data will result in a reduced-dimensional codeword that improves transmission efficiency and reduces resource utilization, which can be recovered at the decoder side (i.e., in this example, at the second wireless communication device 504).

[0097] At action 520, the first wireless communication device 502 may encode the results of the training from action 518 (e.g., trained weights and / or biases) into a codeword. Since this may be a very large data set, the first wireless communication device 504 may request authorization to reserve transmission resources, if necessary, so that the transmission can avoid conflicts with other transmissions. This may come from the second wireless communication device 504 (when it is a BS) or from a third device (when both wireless communication devices are UEs), to name a few examples.

[0098] At act 522, the first wireless communication device 502 sends the codeword (including the neural network training results) to the second wireless communication device 504. In some examples, the first wireless communication device 502 sends the training weights without encoding, and thus proceeds from act 518 to act 522, where the information is sent to the second wireless communication device 504. The second wireless communication device 504 receives the information and can store it along with the same information stored at act 516 so that the training data can be accessed when searching based on at least one antenna configuration parameter as described above.

[0099] At action 524, Figure 5 The second wireless communication device 504 is shown establishing communication with a third wireless communication device 506 (which can be a UE 115 or a BS 105). For example, the second wireless communication device 504 can be a BS 105, and the third wireless communication device 506 can be another UE 115 entering coverage of the BS 105. As another example, the second wireless communication device 504 can be a UE 115 entering coverage of a new BS 105 that is the third wireless communication device 506. As yet another example, both the second wireless communication device 504 and the third wireless communication device 506 can be UEs 115 establishing sidelink communications. Although shown as occurring after action 522, action 524 can occur before, during, or after the events discussed with respect to action 522.

[0100] At act 526, the third wireless communication device 506 sends antenna configuration information for the third wireless communication device 506 to the second wireless communication device 504. This may include one or more parameters of the antenna configuration of the third wireless communication device 506, such as panel orientation, antenna array dimensions, antenna polarization, panel position, etc.

[0101] At act 528, the second wireless communication device 504 determines a level of antenna configuration similarity of the third wireless communication device 506 to any antenna configurations stored at the second wireless communication device 504. This may involve finding a match for the plurality of antenna configuration parameters, a match for one of the plurality of antenna configuration parameters (e.g., array dimensions, as just one example) within a similarity threshold for the plurality of antenna configuration parameters, or within a similarity threshold for one of the plurality of antenna configuration parameters.

[0102] If similarity is determined at act 528, then at act 530, the second wireless communication device 504 may send the neural network information to the third wireless communication device 506 to assist in neural network selection and / or training at the third wireless communication device 506. For example, where one or more parameters of the antenna configuration at the third wireless communication device 506 match / are sufficiently similar to one or more antenna configuration parameters of the first wireless communication device 502, the second wireless communication device 504 may identify the neural network parameters (e.g., an index of the AI ​​module (if pre-provided) or explicit neural network parameters, such as the number / type of layers, the number of nodes, etc., or some combination thereof) from a storage device to send to the third wireless communication device 506.

[0103] Additionally, if the second wireless communication device 504 also stores neural network weights associated with the matching / similar antenna configuration parameters determined from act 528, the second wireless communication device 504 may also transmit these parameters. In some examples, this may be transmitted without compression, while in other examples, this may also be compressed.

[0104] In the example where the second wireless communication device 504 is BS 105 and the third wireless communication device 506 is UE 115, the transmission at action 530 may be an instruction for implementation by UE 115. In the example where the second wireless communication device 504 is UE 115 and the third wireless communication device 506 is UE 115, the transmission at action 530 may be a suggestion to BS 105.

[0105] At action 532, the third wireless communication device 506 implements the neural network information received from the second wireless communication device 504. Thus, if a neural network has already been utilized (e.g., as signaled using the antenna configuration information at action 514 above) and further trained (e.g., as signaled at action 522 above) for a certain antenna configuration pair (e.g., between the first wireless communication device 502 and the second wireless communication device 504), the same neural network can be used for future communication purposes (e.g., with the third wireless communication device 506 in the example above). This can be utilized, for example, when the UE 115 moves to another cell with the same or similar antenna configuration; as another example, when the BS 105 serves a new UE 115 with the same or similar antenna configuration as the previous UE 115 for which the neural network was already utilized and trained. This can save system resources by not requiring the new device to identify and train a new system based on default parameters (e.g., in this example, between the second wireless communication device 504 and the third wireless communication device 506).

[0106] Figure 6 A flow chart of a wireless communication method 600 is shown, in accordance with some embodiments of the present disclosure. Aspects of method 600 may be performed by a wireless communication device, such as a UE 115 and / or 300, utilizing one or more components, such as a processor 302, a memory 304, a neural network communication module 308, a transceiver 310, a modem 312, one or more antennas 316, and various combinations thereof. Alternatively, the wireless communication device may be a base station 105 and / or 400, utilizing one or more components, such as a processor 402, a memory 404, a neural network communication module 408, a transceiver 410, a modem 412, one or more antennas 416, and various combinations thereof. To simplify the discussion, general reference will again be made to a first wireless communication device and a second wireless communication device, either of which can be a UE or a base station, in accordance with embodiments of the present disclosure. As shown, method 600 includes many of the enumerated steps, but embodiments of method 600 may include additional steps before, during, after, and between the enumerated steps. For example, in some cases, one or more aspects of methods 700 and 730 may be implemented as part of method 600. Additionally, in some embodiments, one or more of the enumerated steps may be omitted or performed in a different order.

[0107] At block 602, a first wireless communication device establishes communication with a second wireless communication device, e.g., as described above with respect to Figure 5 Action 510 discussed above.

[0108] At decision block 604 , if the first wireless communication device has been pre-provisioned with a plurality of possible AI modules (each specifying a different set of neural network parameters for implementation), the method 600 may proceed to block 606 .

[0109] At block 606, the first wireless communication device selects a neural network option from the available AI module options. The first wireless communication device may do so based on one or more parameters of the antenna configuration at the first wireless communication device.

[0110] At block 608, the first wireless communication device transmits an identifier of the neural network option selected at block 606, i.e., an AI module identifier, to the second wireless communication device. This may be, for example, receiving an index value that the second wireless communication device identifies as being associated with a corresponding pre-provisioned AI module at the second wireless communication device. The first wireless communication device may transmit the identifier of the neural network option along with the antenna configuration information of the first wireless communication device. This may be via RRC signaling, MAC-CE, or PDCCH (if the first wireless communication device is BS 105), or via MAC-CE or PUCCH (if the first wireless communication device is UE 115).

[0111] Returning to decision block 604 , if the first wireless communication device has not been pre-provisioned with a plurality of possible AI modules, the method 600 instead proceeds to block 610 .

[0112] At block 610, the first wireless communication device determines parameters to be used for a neural network at the first wireless communication device. This may include dynamically determining the number of layers to use when encoding, the number of nodes per layer, and the like. Thus, each aspect of the neural network to be used may be selected. Alternatively, the layer options may be preconfigured, but not the number of nodes per layer, or the number of nodes may be preconfigured, but the number of layers may be dynamically determined.

[0113] At block 612, the first wireless communication device sends explicit information about the neural network determined from block 610 to the second wireless communication device. For example, this may involve explicitly signaling the number of layers of the neural network, the number of nodes per layer, the algorithm or algorithms used, etc. Alternatively, where some parameters may be pre-provisioned and others are not, this may involve implicitly signaling the pre-provisioned portion and explicitly signaling the non-pre-provisioned portion, as mentioned above with respect to action 514. Similar to block 608, the first wireless communication device may send the explicit information about the determined neural network to the second wireless communication device along with the antenna configuration information of the first wireless communication device. This may be via RRC signaling, MAC-CE, or PDCCH (if the first wireless communication device is BS 105), or via MAC-CE or PUCCH (if the first wireless communication device is UE 115).

[0114] From either block 608 or block 612, the method 600 proceeds to block 614. At block 614, the first wireless communication device collaborates (i.e., jointly) with the second wireless communication device to train the neural network selected at block 606 or determined at block 610. This may involve feed-forward and / or feedback operations to optimize a reconstructed estimate of the input at the first wireless communication device at the output at the second wireless communication device.

[0115] At block 616, the first wireless communication device may request one or more transmission resources to transmit one or more trained neural network weights to the second wireless communication device for storage in association with the antenna configuration information of the first wireless communication device. For example, where the first wireless communication device is a UE and the second wireless communication device is a BS, the request for transmission resources may be directed to the BS. As another example, where the second wireless communication device is also a UE, this may be to the BS or to a neighboring device to clear the channel.

[0116] At block 618, the first wireless communication device uses the neural network encoding side at the first wireless communication device to compress the trained neural network weights from block 614. This may produce a codeword for transmission.

[0117] At block 620, the first wireless communication device transmits the codeword generated from block 618 to the second wireless communication device on a channel. This may be via the PDSCH (in the case where the first wireless communication device is BS 105) or the PUSCH (in the case where the first wireless communication device is UE 115). Instead of compressing the trained neural network weights from block 614, the first wireless communication device may transmit the trained neural network weights to the second wireless communication device using conventional methods (again, for example, via the PDSCH (in the case where the first wireless communication device is BS 105) or the PUSCH (in the case where the first wireless communication device is UE 115). Thus, block 618 may be optional, and in the absence of block 618, block 620 includes transmitting the trained neural network weights without compression.

[0118] Figure 7A A flow chart of a wireless communication method 700 is shown in accordance with some embodiments of the present disclosure. Aspects of the method 700 may be performed by a wireless communication device, such as a UE 115 and / or 300 utilizing one or more components, such as a processor 302, a memory 304, a neural network communication module 308, a transceiver 310, a modem 312, one or more antennas 316, and various combinations thereof. Alternatively, the wireless communication device may be a BS 105 and / or 400 utilizing one or more components, such as a processor 402, a memory 404, a neural network communication module 408, a transceiver 410, a modem 412, one or more antennas 416, and various combinations thereof. To simplify the discussion, reference will again be generally made to a first wireless communication device and a second wireless communication device, where either device can be a UE or a BS in accordance with embodiments of the present disclosure. Thus, while as discussed with respect to Figure 6 As done, reference is made to a first wireless communication device and a second wireless communication device, but method 700 describes operations with reference to a device that receives the neural network and antenna configuration information.

[0119] As shown, method 700 includes many of the enumerated steps, but embodiments of method 700 may include additional steps before, during, after, and between the enumerated steps. For example, in some cases, one or more aspects of methods 600 and 730 may be implemented as part of method 700. Furthermore, in some embodiments, one or more of the enumerated steps may be omitted or performed in a different order.

[0120] At block 702, a first wireless communication device establishes communication with a second wireless communication device, e.g., as described above with respect to Figure 5 Action 510 and Figure 6 As discussed in block 602 of FIG.

[0121] At block 704, the first wireless communication device receives neural network information and antenna configuration information from the second wireless communication device. For example, where the second wireless communication device selects an AI module from a plurality of AI modules pre-provisioned at the second wireless communication device (and, in some examples, also pre-provisioned at the first wireless communication device in receive mode), the neural network information may be an index value or other small representation of which AI module has been selected. Alternatively, the neural network information may be explicitly signaled from the second wireless communication device, which may occur where the second wireless communication device determines parameters to be used for the neural network at the second wireless communication device (such as the number of layers, the number of nodes per layer, etc., dynamically for use in encoding).

[0122] At block 706, the first wireless communication device stores the neural network information and corresponding antenna configuration information received from the second wireless communication device, e.g., Figure 5 The first wireless communication device may then search its storage based on characteristics, such as one or more antenna configuration parameters, to determine whether the new device matches or meets a similarity threshold.

[0123] At block 708, the first wireless communication device coordinates with the second wireless communication device in training (ie, jointly training) the neural network identified at block 704 for use at the second wireless communication device.

[0124] At block 710, once the training from block 708 is complete, the first wireless communication device receives one or more weights and / or biases from the second wireless communication device. This may be via the PUSCH (if the first wireless communication device is the BS 105) or the PDSCH (if the first wireless communication device is the UE 115).

[0125] In addition to the above Figure 7A In addition to the aspects described, Figure 7B Other aspects shown, Figure 7BA flow chart of a wireless communication method 730 according to some embodiments of the present disclosure is shown. Various aspects of the method 730 may be performed by a wireless communication device, such as a UE 115 and / or 300, utilizing one or more components, such as a processor 302, a memory 304, a neural network communication module 308, a transceiver 310, a modem 312, one or more antennas 316, and various combinations thereof. Alternatively, the wireless communication device may be a BS 105 and / or 400, utilizing one or more components, such as a processor 402, a memory 404, a neural network communication module 408, a transceiver 410, a modem 412, one or more antennas 416, and various combinations thereof. To simplify the discussion, reference will again be generally made to a first wireless communication device and a second wireless communication device, wherein either device can be a UE or a BS according to embodiments of the present disclosure.

[0126] As shown, method 700 includes many of the enumerated steps, but embodiments of method 700 may include additional steps before, during, after, and between the enumerated steps. For example, in some cases, one or more aspects of methods 600, 700 may be implemented as part of method 730. For example, aspects of method 730 may occur after the events of method 700 discussed above. Thus, the first wireless communication device may have received neural network information, antenna configuration information, and / or neural network weights for the associated neural network information and antenna configuration information from the second wireless communication device. Furthermore, in some embodiments, one or more of the enumerated steps may be omitted or performed in a different order.

[0127] At block 732, the first wireless communication device establishes communication with the third wireless communication device, as described above with respect to Figure 5 For example, in the case where the first wireless communication device is BS 105, the third wireless communication device may be UE 115. As another example, in the case where the first wireless communication device is UE 115, the third wireless communication device may be another UE 115 or BS 105.

[0128] At block 734, the first wireless communication device receives antenna configuration information from the third wireless communication device, as described above with respect to Figure 5 Action 526 discussed above.

[0129] At block 736, upon receiving antenna configuration information regarding a configuration at the third wireless communication device, the first wireless communication device may compare the received antenna configuration information regarding the third wireless communication device with stored antenna configuration information (including antenna configuration information stored at the first wireless communication device previously received from the second wireless communication device (e.g., according to the above)). Figure 7A704 and 706)) for comparison.

[0130] At decision block 738, if the antenna configuration information of the third wireless communication device does not meet the similarity threshold for any stored antenna configuration information, then method 730 may proceed to block 740. For example, in some embodiments, the similarity threshold is a matching threshold for a plurality of antenna configuration parameters, while in other embodiments, the similarity threshold is a matching threshold for one of the plurality of antenna configuration parameters (e.g., array dimensions, as just one example). As yet another example, the similarity threshold comprises a level of similarity to the plurality of antenna configuration parameters (e.g., by a similarity threshold for each individual parameter from the plurality of parameters, or by comparing an average of the similarities to a combined threshold, or by a weighted average to give greater weight to one or more particular parameters than to other parameters). As another example, the similarity threshold comprises a similarity threshold to one of the plurality of antenna configuration parameters.

[0131] At block 740, the first wireless communication device participates in training a neural network with the third wireless communication device using default parameters.

[0132] Returning to decision block 738, if, alternatively, the antenna configuration information of the third wireless communication device satisfies the similarity threshold with the antenna configuration entry stored at the first wireless communication device, the first wireless communication device transmits the neural network information associated with the stored antenna configuration entry to the third wireless communication device. For example, the neural network information may include an identification of neural network parameters from the stored information (e.g., an index of an AI module (if pre-provisioned) or explicit neural network parameters (such as the number / type of layers, number of nodes, etc.), or some combination thereof). This may be used by the third wireless communication device to assist in neural network selection and / or training. Furthermore, if neural network weights are also stored with the antenna configuration entry, the first wireless communication device may also transmit these weights with or without compression. The neural network information may be transmitted as part of instructions for implementing the information (e.g., if the first wireless communication device is BS 105) or as part of a suggestion for implementation (e.g., if the first wireless communication device is UE 115 and the third wireless communication device is BS 105). This may conserve system resources by not requiring the new device to identify and train a new system based on default parameters (e.g., between the first and third wireless communication devices in this example).

[0133] Information and signals may be represented using any of a variety of different techniques and methods. For example, data, instructions, commands, information, signals, bits, symbols, and chips mentioned throughout the foregoing description may be represented by voltages, currents, electromagnetic waves, magnetic fields or particles, optical fields or particles, or any combination thereof.

[0134] The various illustrative blocks and modules described in conjunction with the disclosure herein may be implemented or executed using a general purpose processor, a DSP, an ASIC, an FPGA or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general purpose processor may be a microprocessor, but in the alternative, the processor may be any conventional processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices (e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, a combination of one or more microprocessors and a DSP core, or any other such configuration).

[0135] The functions described herein can be implemented in hardware, software executed by a processor, firmware, or any combination thereof. When implemented in software executed by a processor, the functions can be stored as one or more instructions or codes on a computer-readable medium, or sent on a computer-readable medium. Other examples and implementations are within the scope of protection of this disclosure and the appended claims. For example, due to the nature of software, the functions described above can be implemented using software executed by a processor, hardware, firmware, hardware wiring, or any combination thereof. The features used to implement the functions can be physically located at multiple locations, including being distributed so that the functions are implemented at different physical locations. In addition, as used herein (including in the claims), the "or" used in a list of items (e.g., the "or" used in a list of items ending with "at least one of" or "one or more of") indicates an inclusive list, so that, for example, a list of [at least one of A, B, or C] means: A or B or C or AB or AC or BC or ABC (i.e., A and B and C).

[0136] Examples of this disclosure may include the following terms:

[0137] 1. A wireless communication method, comprising:

[0138] establishing, by the first wireless communication device, communication with the second wireless communication device;

[0139] determining, by the first wireless communication device, a neural network to be used for the communication with the second wireless communication device based on an antenna configuration of the first wireless communication device; and

[0140] Antenna configuration information and at least one neural network parameter are sent by the first wireless communication device to the second wireless communication device based on the neural network.

[0141] 2. The method according to clause 1, wherein:

[0142] The determining further comprises: selecting the neural network from a plurality of neural networks, and

[0143] The at least one neural network parameter includes an identifier of the selected neural network.

[0144] 3. The method according to clause 1 or clause 2, wherein:

[0145] The first wireless communication device comprises a user equipment and the second wireless communication device comprises a base station, and

[0146] The transmission is on a physical uplink control channel.

[0147] 4. A method according to clause 1 or clause 2, wherein:

[0148] The first wireless communication device comprises a base station and the second wireless communication device comprises a user equipment, and

[0149] The transmission is on a physical downlink control channel.

[0150] 5. The method according to clause 1 or clause 2, wherein:

[0151] The first wireless communication device comprises a first user equipment (UE) and the second wireless communication device comprises a second UE, and

[0152] The transmission is on a physical sidelink control channel.

[0153] 6. The method according to any one of clauses 1 to 5, further comprising:

[0154] Coordinating training of the neural network by the first wireless communication device and the second wireless communication device; and

[0155] At least one training weight resulting from the training is transmitted by the first wireless communication device to the second wireless communication device.

[0156] 7. The method according to any one of clauses 1 to 6, wherein the antenna configuration information comprises a panel direction of an antenna array of the first wireless communication device, dimensions of the antenna array, and polarization information of the antenna array.

[0157] 8. A method of wireless communication, comprising:

[0158] establishing, by the first wireless communication device, communication with the second wireless communication device;

[0159] receiving, by the first wireless communication device from the second wireless communication device, antenna configuration information of an antenna configuration at the second wireless communication device and at least one neural network parameter of a neural network at the second wireless communication device for use in the communication; and

[0160] The antenna configuration information and the at least one neural network parameter are stored by the first wireless communication device.

[0161] 9. The method according to clause 8, further comprising:

[0162] A second communication is established by the first wireless communication device with a third wireless communication device.

[0163] 10. The method according to clause 9, further comprising:

[0164] determining, by the first wireless communication device, a sharing characteristic of the antenna configuration of the third wireless communication device and the antenna configuration information from the second wireless communication device; and

[0165] The at least one neural network parameter is sent, by the first wireless communication device to the third wireless communication device in response to the determination, for use in the second communication.

[0166] 11. The method according to clause 8 or clause 9, further comprising:

[0167] Coordinating, by the first wireless communication device and the second wireless communication device, to train the neural network; and

[0168] At least one training weight for the neural network is received by the first wireless communication device from the second wireless communication device in response to the coordination.

[0169] 12. The method according to clause 11, further comprising:

[0170] determining, by the first wireless communication device, that the antenna configuration of the third wireless communication device shares characteristics with the antenna configuration at the second wireless communication device; and

[0171] The at least one neural network parameter and the at least one training weight are sent, by the first wireless communication device, to the third wireless communication device in response to the determination, for use in the second communication.

[0172] 13. The method according to clause 11 or clause 12, further comprising:

[0173] An allocation of resources on which the at least one training weight is received is provided by the first wireless communication device to the second wireless communication device.

[0174] 14. A method according to any of clauses 8 to 13, wherein the neural network comprises one of a plurality of neural networks pre-configured at the first wireless communication device and the second wireless communication device.

[0175] 15. The method according to any of clauses 8 to 14, wherein the antenna configuration information comprises a panel orientation of an antenna array of the second wireless communication device, dimensions of the antenna array, and polarization information of the antenna array.

[0176] 16. A first wireless communication device, comprising:

[0177] a transceiver configured to: establish communication with a second wireless communication device; and

[0178] a processor configured to: determine a neural network to use for the communication with the second wireless communication device based on an antenna configuration of the first wireless communication device;

[0179] The transceiver is further configured to send antenna configuration information and at least one neural network parameter to the second wireless communication device based on the neural network.

[0180] 17. The first wireless communication device of clause 16, wherein:

[0181] For the determination, the processor is further configured to: select the neural network from a plurality of neural networks, and

[0182] The at least one neural network parameter includes an identifier of the selected neural network.

[0183] 18. A first wireless communication device according to clause 16 or clause 17, wherein:

[0184] The first wireless communication device comprises a user equipment and the second wireless communication device comprises a base station, and

[0185] The transmission is on a physical uplink control channel.

[0186] 19. A first wireless communications device as described in clause 16 or clause 17, wherein:

[0187] The first wireless communication device comprises a base station and the second wireless communication device comprises a user equipment, and

[0188] The transmission is on a physical downlink control channel.

[0189] 20. A first wireless communication device as described in clause 16 or clause 17, wherein:

[0190] The first wireless communication device comprises a first user equipment (UE) and the second wireless communication device comprises a second UE, and

[0191] The transmission is on a physical sidelink control channel.

[0192] 21. A first wireless communication device according to any of clauses 16 to 20, wherein:

[0193] The processor is further configured to coordinate training of the neural network with the second wireless communication device, and

[0194] The transceiver is further configured to transmit at least one training weight resulting from the training to the second wireless communication device.

[0195] 22. A first wireless communication device according to any of clauses 16 to 21, wherein the processor comprises a neural network unit integrated with a transceiver.

[0196] 23. A first wireless communication device as described in any of clauses 16 to 21, wherein the processor comprises an application processor separate from the transceiver.

[0197] 24. A first wireless communication device, comprising:

[0198] A transceiver, the transceiver being configured to:

[0199] establishing communication with a second wireless communication device; and

[0200] receiving, from the second wireless communication device, antenna configuration information of an antenna configuration at the second wireless communication device and at least one neural network parameter of a neural network at the second wireless communication device for use in the communication; and

[0201] A processor is configured to store the antenna configuration information and the at least one neural network parameter at the first wireless communication device.

[0202] 25. A first wireless communication device according to clause 24, wherein the transceiver is further configured to:

[0203] A second communication is established with a third wireless communication device.

[0204] 26. The first wireless communication device of clause 25, wherein:

[0205] The processor is further configured to: determine an antenna configuration of the third wireless communication device sharing characteristics with the antenna configuration information from the second wireless communication device; and

[0206] The transceiver is further configured to transmit the at least one neural network parameter to the third wireless communication device for use in the second communication in response to the determination.

[0207] 27. A first wireless communications device as described in clause 24 or clause 25, wherein:

[0208] The processor is further configured to coordinate with the second wireless communication device to train the neural network, and

[0209] The transceiver is further configured to receive at least one training weight for the neural network from the second wireless communication device in response to the coordination.

[0210] 28. The first wireless communication device of clause 27, wherein:

[0211] The processor is further configured to: determine that the antenna configuration of the third wireless communication device shares characteristics with the antenna configuration at the second wireless communication device; and

[0212] The transceiver is further configured to transmit the at least one neural network parameter and the at least one training weight to the third wireless communication device in response to the determination for use in the second communication.

[0213] 29. A first wireless communication device according to any of clauses 24 to 28, wherein the processor comprises a neural network unit integrated with the transceiver.

[0214] 30. A first wireless communication device as described in any of clauses 24 to 28, wherein the processor comprises an application processor separate from the transceiver.

[0215] Examples of this disclosure may include the following terms:

[0216] 1. A non-transitory computer-readable medium having program code recorded thereon, the program code comprising:

[0217] code for causing a first wireless communication device to establish communication with a second wireless communication device;

[0218] code for causing the first wireless communication device to determine a neural network to use for the communication with the second wireless communication device based on an antenna configuration of the first wireless communication device; and

[0219] Code for causing the first wireless communication device to transmit antenna configuration information and at least one neural network parameter to the second wireless communication device based on the neural network.

[0220] 2. The non-transitory computer-readable medium of clause 1, wherein the neural network comprises one of a plurality of neural networks pre-configured at the first wireless communication device.

[0221] 3. The non-transitory computer-readable medium of clause 2, wherein:

[0222] The code for causing the first wireless communication device to determine the neural network further includes code for causing the first wireless communication device to select the neural network from the plurality of neural networks, and

[0223] The at least one neural network parameter includes an identifier of the selected neural network.

[0224] 4. The non-transitory computer-readable medium of clause 1, wherein the at least one neural network parameter comprises one or more explicit parameters of the neural network.

[0225] 5. The non-transitory computer-readable medium of clause 1, wherein:

[0226] The first wireless communication device comprises a user equipment and the second wireless communication device comprises a base station, and

[0227] The transmission is on a physical uplink control channel.

[0228] 6. The non-transitory computer-readable medium of clause 1, wherein:

[0229] The first wireless communication device comprises a base station and the second wireless communication device comprises a user equipment, and

[0230] The transmission is on a physical downlink control channel.

[0231] 7. The non-transitory computer-readable medium of clause 1, wherein:

[0232] The first wireless communication device comprises a first user equipment (UE) and the second wireless communication device comprises a second UE, and

[0233] The transmission is on a physical sidelink control channel.

[0234] 8. The non-transitory computer-readable medium of clause 1, further comprising:

[0235] Code for causing the first wireless communication device to coordinate training of the neural network with the second wireless communication device.

[0236] 9. The non-transitory computer-readable medium of clause 8, further comprising:

[0237] Code for causing the first wireless communication device to transmit at least one training weight resulting from the training to the second wireless communication device.

[0238] 10. The non-transitory computer-readable medium of clause 9, further comprising:

[0239] Code for causing the first wireless communication device to request allocation of resources on which to transmit the at least one training weight

[0240] 11. The non-transitory computer-readable medium of clause 1, wherein the antenna configuration information comprises a panel orientation of an antenna array of the first wireless communication device, dimensions of the antenna array, and polarization information of the antenna array.

[0241] 12. A non-transitory computer-readable medium having program code recorded thereon, the program code comprising:

[0242] code for causing a first wireless communication device to establish communication with a second wireless communication device;

[0243] code for causing the first wireless communication device to receive, from the second wireless communication device, antenna configuration information of an antenna configuration at the second wireless communication device and at least one neural network parameter of a neural network at the second wireless communication device for use in the communication; and

[0244] Code for causing the first wireless communication device to store the antenna configuration information and the at least one neural network parameter.

[0245] 13. The non-transitory computer-readable medium of clause 12, further comprising:

[0246] Code for causing the first wireless communication device to establish a second communication with a third wireless communication device.

[0247] 14. The non-transitory computer-readable medium of clause 13, further comprising:

[0248] code for causing the first wireless communication device to determine a sharing characteristic of the antenna configuration of the third wireless communication device and the antenna configuration information from the second wireless communication device; and

[0249] Code for causing the first wireless communication device to transmit the at least one neural network parameter to the third wireless communication device for use in the second communication in response to the determination.

[0250] 15. The non-transitory computer-readable medium of clause 13, further comprising:

[0251] Code for causing the first wireless communication device to coordinate with the second wireless communication device to train the neural network.

[0252] 16. The non-transitory computer-readable medium of clause 15, further comprising:

[0253] Code for causing the first wireless communication device to receive, from the second wireless communication device in response to the coordination, at least one trained weight for the neural network.

[0254] 17. The non-transitory computer-readable medium of clause 16, further comprising:

[0255] code for causing the first wireless communication device to determine that the antenna configuration of the third wireless communication device shares characteristics with the antenna configuration at the second wireless communication device; and

[0256] Code for causing the first wireless communication device to transmit the at least one neural network parameter and the at least one training weight to the third wireless communication device for use in the second communication in response to the determination.

[0257] 18. The non-transitory computer-readable medium of clause 16, further comprising:

[0258] Code for causing the first wireless communication device to provide, to the second wireless communication device, an allocation of resources on which the at least one training weight was received.

[0259] 19. The non-transitory computer-readable medium of clause 12, wherein the neural network comprises one of a plurality of neural networks pre-configured at the first wireless communication device and the second wireless communication device.

[0260] 20. The non-transitory computer-readable medium of clause 19, wherein the at least one neural network parameter comprises an identifier of the neural network selected from the plurality of neural networks.

[0261] 21. The non-transitory computer-readable medium of clause 12, wherein the at least one neural network parameter comprises one or more explicit parameters of the neural network.

[0262] 22. The non-transitory computer-readable medium of clause 12, wherein:

[0263] The first wireless communication device comprises a user equipment and the second wireless communication device comprises a base station, and

[0264] The receiving is on a physical downlink control channel.

[0265] 23. The non-transitory computer-readable medium of clause 12, wherein:

[0266] The first wireless communication device comprises a base station and the second wireless communication device comprises a user equipment, and

[0267] The receiving is on a physical uplink control channel.

[0268] 24. The non-transitory computer-readable medium of clause 12, wherein:

[0269] The first wireless communication device comprises a first user equipment (UE) and the second wireless communication device comprises a second UE, and

[0270] The receiving is on a physical sidelink control channel.

[0271] 25. The non-transitory computer-readable medium of clause 12, wherein the antenna configuration information comprises a panel orientation of an antenna array of the second wireless communication device, dimensions of the antenna array, and polarization information of the antenna array.

[0272] 26. A first wireless communication device, comprising:

[0273] means for establishing communication with a second wireless communication device;

[0274] means for determining a neural network to use for said communication with said second wireless communication device based on an antenna configuration of said first wireless communication device; and

[0275] Means for sending antenna configuration information and at least one neural network parameter to the second wireless communication device based on the neural network.

[0276] 27. The first wireless communication device of clause 26, wherein the neural network comprises a neural network of a plurality of neural networks pre-configured at the first wireless communication device.

[0277] 28. The first wireless communication device of clause 27, wherein:

[0278] The means for determining the neural network further comprises: means for selecting the neural network from the plurality of neural networks, and

[0279] The at least one neural network parameter includes an identifier of the selected neural network.

[0280] 29. The first wireless communication device of clause 26, wherein the at least one neural network parameter comprises one or more explicit parameters of the neural network.

[0281] 30. The first wireless communication device of clause 26, wherein:

[0282] The first wireless communication device comprises a user equipment and the second wireless communication device comprises a base station, and

[0283] The transmission is on a physical uplink control channel.

[0284] 31. The first wireless communication device of clause 26, wherein:

[0285] The first wireless communication device comprises a base station and the second wireless communication device comprises a user equipment, and

[0286] The transmission is on a physical downlink control channel.

[0287] 32. The first wireless communication device of clause 26, wherein:

[0288] The first wireless communication device comprises a first user equipment (UE) and the second wireless communication device comprises a second UE, and

[0289] The transmission is on a physical sidelink control channel.

[0290] 33. The first wireless communication device of clause 26, further comprising:

[0291] Means for coordinating training of the neural network with the second wireless communication device.

[0292] 34. The first wireless communication device of clause 33, further comprising:

[0293] means for transmitting at least one training weight resulting from the training to the second wireless communication device.

[0294] 35. The first wireless communication device of clause 34, further comprising:

[0295] means for requesting allocation of a resource on which to transmit the at least one training weight

[0296] 36. The first wireless communication device of clause 26, wherein the antenna configuration information comprises a panel orientation of an antenna array of the first wireless communication device, dimensions of the antenna array, and polarization information of the antenna array.

[0297] 37. A first wireless communication device, comprising:

[0298] means for establishing communication with a second wireless communication device;

[0299] means for receiving, from the second wireless communication device, antenna configuration information of an antenna configuration at the second wireless communication device and at least one neural network parameter of a neural network at the second wireless communication device for use in the communication; and

[0300] means for storing the antenna configuration information and the at least one neural network parameter.

[0301] 38. The first wireless communication device of clause 37, further comprising:

[0302] Means for establishing a second communication with a third wireless communication device.

[0303] 39. The first wireless communication device of clause 38, further comprising:

[0304] means for determining a sharing characteristic of the antenna configuration of the third wireless communication device and the antenna configuration information from the second wireless communication device; and

[0305] means for transmitting, in response to the determining, the at least one neural network parameter to the third wireless communication device for use in the second communication.

[0306] 40. The first wireless communication device of clause 38, further comprising:

[0307] Means for coordinating with the second wireless communication device to train the neural network.

[0308] 41. The first wireless communication device of clause 40, further comprising:

[0309] Means for receiving, in response to the coordinating, at least one training weight for the neural network from the second wireless communication device.

[0310] 42. The first wireless communication device of clause 41, further comprising:

[0311] means for determining that the antenna configuration of the third wireless communication device shares characteristics with the antenna configuration at the second wireless communication device; and

[0312] means for transmitting, in response to the determining, the at least one neural network parameter and the at least one training weight to the third wireless communication device for use in the second communication.

[0313] 43. The first wireless communication device of clause 41, further comprising:

[0314] Means for providing, to the second wireless communication device, an allocation of resources on which the at least one training weight was received.

[0315] 44. The first wireless communication device of clause 37, wherein the neural network comprises one of a plurality of neural networks preconfigured at the first wireless communication device and the second wireless communication device.

[0316] 45. The first wireless communication device of clause 44, wherein the at least one neural network parameter comprises an identifier of the neural network selected from the plurality of neural networks.

[0317] 46. ​​The first wireless communication device of clause 37, wherein the at least one neural network parameter comprises one or more explicit parameters of the neural network.

[0318] 47. The first wireless communications device of clause 37, wherein:

[0319] The first wireless communication device comprises a user equipment and the second wireless communication device comprises a base station, and

[0320] The receiving is on a physical downlink control channel.

[0321] 48. The first wireless communications device of clause 37, wherein:

[0322] The first wireless communication device comprises a base station and the second wireless communication device comprises a user equipment, and

[0323] The receiving is on a physical uplink control channel.

[0324] 49. The first wireless communications device of clause 37, wherein:

[0325] The first wireless communication device comprises a first user equipment (UE) and the second wireless communication device comprises a second UE, and

[0326] The receiving is on a physical sidelink control channel.

[0327] 50. The first wireless communication device of clause 37, wherein the antenna configuration information comprises a panel orientation of an antenna array of the second wireless communication device, dimensions of the antenna array, and polarization information of the antenna array.

[0328] As will be apparent to those skilled in the art by now, and depending on the specific application at the time, many modifications, substitutions, and changes can be made in the materials, apparatus, configurations, and methods of use of the apparatus of the present disclosure, as well as in the apparatus of the present disclosure, without departing from the spirit and scope of the present disclosure. In view of this, the scope of the present disclosure should not be limited to the scope of the specific embodiments shown and described herein (as they are by way of some examples thereof), but should be fully commensurate with the claims appended hereto and their functional equivalents.

Claims

1. A method of wireless communication, comprising: establishing, by the first wireless communication device, communication with the second wireless communication device; determining, by the first wireless communication device, a neural network to be used for the communication with the second wireless communication device based on an antenna configuration of the first wireless communication device; as well as In response to determining the neural network, antenna configuration information and at least one neural network parameter are sent by the first wireless communication device to the second wireless communication device based on the neural network.

2. The method according to claim 1, wherein: The determining further comprises: selecting the neural network from a plurality of neural networks, and The at least one neural network parameter includes an identifier of the selected neural network.

3. The method according to claim 1, wherein: The first wireless communication device comprises a user equipment and the second wireless communication device comprises a base station, and The transmission is on a physical uplink control channel.

4. The method according to claim 1, wherein: The first wireless communication device comprises a base station and the second wireless communication device comprises a user equipment, and The transmission is on a physical downlink control channel.

5. The method according to claim 1, wherein: The first wireless communication device comprises a first user equipment (UE) and the second wireless communication device comprises a second UE, and The transmission is on a physical sidelink control channel.

6. The method according to claim 1, further comprising: Coordinating training of the neural network by the first wireless communication device and the second wireless communication device; as well as At least one training weight resulting from the training is transmitted by the first wireless communication device to the second wireless communication device.

7. The method according to claim 1, wherein The antenna configuration information includes a panel direction of an antenna array of the first wireless communication device, dimensions of the antenna array, and polarization information of the antenna array.

8. A method of wireless communication, comprising: establishing, by the first wireless communication device, communication with the second wireless communication device; receiving, by the first wireless communication device from the second wireless communication device, antenna configuration information at the second wireless communication device in response to determining a neural network and at least one neural network parameter of the neural network at the second wireless communication device for use in the communication; as well as The antenna configuration information and the at least one neural network parameter are stored by the first wireless communication device.

9. The method according to claim 8, wherein The neural network is used for the communication between the first wireless communication device and the second wireless communication device.

10. The method according to claim 8, further comprising: A second communication is established by the first wireless communication device with a third wireless communication device.

11. The method according to claim 10, further comprising: determining, by the first wireless communication device, a sharing characteristic of the antenna configuration of the third wireless communication device and the antenna configuration information from the second wireless communication device; as well as The at least one neural network parameter is sent, by the first wireless communication device to the third wireless communication device in response to the determination, for use in the second communication.

12. The method according to claim 10, further comprising: Coordinating, by the first wireless communication device and the second wireless communication device, to train the neural network; as well as At least one training weight for the neural network is received by the first wireless communication device from the second wireless communication device in response to the coordination.

13. The method according to claim 12, further comprising: determining, by the first wireless communication device, that the antenna configuration of the third wireless communication device shares characteristics with the antenna configuration at the second wireless communication device; as well as The at least one neural network parameter and the at least one training weight are sent, by the first wireless communication device, to the third wireless communication device in response to the determination, for use in the second communication.

14. The method according to claim 12, further comprising: An allocation of resources on which the at least one training weight was received is provided by the first wireless communication device to the second wireless communication device.

15. The method according to claim 8, wherein The neural network includes one of a plurality of neural networks preconfigured at the first wireless communication device and the second wireless communication device.

16. The method according to claim 8, wherein The antenna configuration information includes a panel direction of an antenna array of the second wireless communication device, dimensions of the antenna array, and polarization information of the antenna array.

17. A first wireless communication device, comprising: a transceiver configured to: establish communication with a second wireless communication device; as well as a processor configured to determine a neural network to use for the communication with the second wireless communication device based on an antenna configuration of the first wireless communication device, The transceiver is further configured to: in response to the processor determining the neural network, send antenna configuration information and at least one neural network parameter to the second wireless communication device based on the neural network.

18. The first wireless communication device of claim 17, wherein: For the determination, the processor is further configured to: select the neural network from a plurality of neural networks, and The at least one neural network parameter includes an identifier of the selected neural network.

19. The first wireless communication device of claim 17, wherein: The first wireless communication device comprises a user equipment and the second wireless communication device comprises a base station, and The transmission is on a physical uplink control channel.

20. The first wireless communication device of claim 17, wherein: The first wireless communication device comprises a base station and the second wireless communication device comprises a user equipment, and The transmission is on a physical downlink control channel.

21. The first wireless communication device of claim 17, wherein: The first wireless communication device comprises a first user equipment (UE) and the second wireless communication device comprises a second UE, and The transmission is on a physical sidelink control channel.

22. The first wireless communication device of claim 17, wherein: The processor is further configured to train the neural network in coordination with the second wireless communication device, and The transceiver is further configured to transmit at least one training weight resulting from the training to the second wireless communication device.

23. The first wireless communication device according to claim 17, wherein: The processor includes a neural network unit integrated with the transceiver.

24. The first wireless communication device according to claim 17, wherein: The processor includes an application processor separate from the transceiver.

25. A first wireless communication device, comprising: A transceiver, the transceiver being configured to: establishing communication with a second wireless communication device; as well as receiving, from the second wireless communication device, antenna configuration information at the second wireless communication device in response to determining a neural network and at least one neural network parameter of the neural network at the second wireless communication device for use in the communication; as well as A processor is configured to store the antenna configuration information and the at least one neural network parameter at the first wireless communication device.

26. The first wireless communication device according to claim 25, wherein: The neural network is used for the communication between the first wireless communication device and the second wireless communication device.

27. The first wireless communication device according to claim 25, wherein: The transceiver is further configured to: A second communication is established with a third wireless communication device.

28. The first wireless communication device of claim 27, wherein: The processor is further configured to: determine an antenna configuration of the third wireless communication device sharing characteristics with the antenna configuration information from the second wireless communication device; and The transceiver is further configured to transmit the at least one neural network parameter to the third wireless communication device for use in the second communication in response to the determination.

29. The first wireless communication device of claim 27, wherein: The processor is further configured to coordinate with the second wireless communication device to train the neural network, and The transceiver is further configured to receive at least one training weight for the neural network from the second wireless communication device in response to the coordination.

30. The first wireless communication device of claim 29, wherein: The processor is further configured to: determine that the antenna configuration of the third wireless communication device shares characteristics with the antenna configuration at the second wireless communication device; and The transceiver is further configured to transmit the at least one neural network parameter and the at least one training weight to the third wireless communication device in response to the determination for use in the second communication.

31. The first wireless communication device according to claim 25, wherein: The processor includes a neural network unit integrated with the transceiver.

32. The first wireless communication device according to claim 25, wherein: The processor includes an application processor separate from the transceiver.

Images