Configurable metrics for channel state compression and feedback

Abstract

Methods, systems, and devices for wireless communication are described. In general, the described techniques in user equipment (UE) provide for efficiently reporting channel state information (CSI) to a base station at an appropriate accuracy level. In particular, the base station may indicate an accuracy level for the UE to report the CSI. The UE may encode the CSI using a first neural network, and the base station may decode the CSI using a second neural network. The first and second neural networks may form a neural network pair, and the UE may train the neural network pair based on the accuracy level indicated by the base station. For example, the base station may indicate a loss function corresponding to the accuracy level for which the CSI is to be reported by the UE, and the UE may train the neural network pair using the loss function.

Description

Technology Field

[0001] The following concerns wireless communications in general, and more specifically, configurable metrics for channel state compression and feedback. Background Technology

[0002] Wireless communication systems are widely deployed to provide various types of communication content, such as voice, video, packet data, messaging, and broadcast. These systems may be able to support communication with multiple users by sharing available system resources (e.g., time, frequency, and power). Examples of such multiple access systems include fourth-generation (4G) systems, such as Long Term Evolution (LTE) systems, LTE-Advanced (LTE-A) systems, or LTE-A Pro systems, and fifth-generation (5G) systems, which may be referred to as New Radio (NR) systems. These systems may employ technologies such as Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Frequency Division Multiple Access (FDMA), Orthogonal Frequency Division Multiple Access (OFDMA), or Discrete Fourier Transform Spread Orthogonal Frequency Division Multiplexing (DFT-S-OFDM).

[0003] A wireless multiple access communication system may include one or more base stations or one or more network access nodes, each of which simultaneously supports communication to multiple communication devices, which may otherwise be known as User Equipment (UE). In some wireless communication systems, the UE may be configured to report Channel State Information (CSI) to the base station to indicate downlink channel conditions, and the base station may use the CSI to improve the quality of downlink transmissions to the UE. For example, the CSI may include Channel Quality Indicator (CQI), and the base station may use the CQI to identify appropriate parameters (e.g., Modulation and Coding Scheme (MCS)) for transmitting downlink data to the UE. means of solving the problem

[0004] The described techniques relate to improved methods, systems, devices, and apparatus that support configurable metrics for channel state compression and feedback. Generally, the described techniques in a user equipment (UE) provide for efficiently reporting channel state information (CSI) to a base station at an appropriate level of accuracy. In particular, the base station may indicate the level of accuracy for which the UE reports the CSI. The UE may encode the CSI using a first neural network, and the base station may decode the CSI using a second neural network. The first and second neural networks may form a neural network pair, and the UE may train the neural network pair based on the level of accuracy indicated by the base station. For example, the base station may indicate a loss metric or function corresponding to the level of accuracy for which the CSI is to be reported by the UE, and the UE may train the neural network pair using the loss metric or function. Using these techniques, the base station may be able to configure the UE to report the CSI at an appropriate level of accuracy.

[0005] A method of wireless communication in a UE is described. The method may include receiving from a base station an indication of an accuracy level for reporting channel state feedback to the base station, receiving downlink data or reference signals from the base station, and reporting channel state feedback to the base station corresponding to the indicated accuracy level based on the downlink data or reference signals.

[0006] An apparatus for wireless communication in a UE is described. The apparatus may include a processor, a memory coupled to the processor, and instructions stored in the memory. The instructions may be executable by the processor to cause the apparatus to receive from a base station an indication of an accuracy level for reporting channel state feedback to the base station, to receive downlink data or reference signals from the base station, and to report channel state feedback to the base station corresponding to the indicated accuracy level based on the downlink data or reference signals.

[0007] Another device for wireless communication in a UE is described. The device may include means for receiving, from a base station, an indication of an accuracy level for reporting channel state feedback to the base station, means for receiving downlink data or reference signals from the base station, and means for reporting channel state feedback to the base station corresponding to the indicated accuracy level based on the downlink data or reference signals.

[0008] A non-transient computer-readable medium for storing code for wireless communication in a UE is described. The code may include instructions executable by a processor to receive from a base station an indication of an accuracy level for reporting channel state feedback to the base station, receive downlink data or reference signals from the base station, and report channel state feedback to the base station corresponding to the indicated accuracy level based on the downlink data or reference signals.

[0009] In some examples of the methods, apparatuses, and non-transient computer-readable media described herein, receiving an indication of an accuracy level for reporting channel state feedback may include operations, features, means, or instructions for receiving an indication of a loss function corresponding to an accuracy level for training a pair of neural networks, and the pair of neural networks includes a first neural network in an encoder for encoding channel state feedback and a second neural network in a decoder for decoding channel state feedback, and the method further includes the step of training a pair of neural networks using a loss function. Since the UE may receive an indication of a loss function from a base station, the UE may report CSI feedback to the base station at an appropriate accuracy level.

[0010] In some examples of the methods, apparatuses, and non-transient computer-readable media described herein, training a pair of neural networks using a loss function may include: repeatedly inputting a channel state feedback input to the pair of neural networks and identifying a channel state feedback output from the pair of neural networks; determining the difference between the channel state feedback input and the channel state feedback output for each iteration using a loss function, wherein the difference includes a loss; determining the difference between the channel state feedback input and the channel state feedback output; and, based on the determination, adjusting the coefficients of the pair of neural networks for each iteration to minimize the difference between the channel state feedback input and the channel state feedback output. Since the UE may train the pair of neural networks using a loss function corresponding to an indicated accuracy level, the UE may report CSI feedback at an appropriate accuracy level to minimize unnecessary overhead.

[0011] In some examples of the methods, apparatuses, and non-transient computer-readable media described herein, reporting channel state feedback corresponding to an indicated accuracy level may include operations, features, means, or instructions for encoding channel state feedback using a first neural network in an encoder based on training and reporting the encoded channel state feedback. Some examples of the methods, apparatuses, and non-transient computer-readable media described herein may further include operations, features, means, or instructions for transmitting coefficients of a second neural network for decoding channel state feedback based on training to a base station. Since the UE may transmit coefficients of the second neural network for decoding channel state feedback to a base station, the UE may train the neural network pair without exchanging signaling with the base station, resulting in reduced overhead.

[0012] Some examples of the methods, apparatuses, and non-transient computer-readable media described herein may further include actions, features, means, or instructions for receiving an indication from a base station to train a set of neural network pairs based on a set of accuracy levels, wherein the set of neural network pairs comprises neural network pairs, said indication, and for training each of the set of neural network pairs based on individual accuracy levels of the set of accuracy levels. In some examples of the methods, apparatuses, and non-transient computer-readable media described herein, receiving an indication of accuracy levels may include actions, features, means, or instructions for receiving an indication to use a neural network pair among the set of neural network pairs to report channel state feedback. Since the UE may train a set of neural network pairs, the base station may dynamically indicate an accuracy level to the UE, and the UE may select one of the sets of neural network pairs corresponding to the accuracy level to use for encoding channel state feedback. That is, the UE may avoid training neural network pairs based on the indicated accuracy level after receiving an indication of the accuracy level, thereby resulting in reduced latency.

[0013] Some examples of the methods, apparatuses, and non-transient computer-readable media described herein may further include operations, features, means, or instructions for autonomously selecting a pair of neural networks from a set of pairs of neural networks to report channel state feedback. Some examples of the methods, apparatuses, and non-transient computer-readable media described herein may further include operations, features, means, or instructions for receiving a indication of a subset of a set of pairs of neural networks for training a UE. Some examples of the methods, apparatuses, and non-transient computer-readable media described herein may further include operations, features, means, or instructions for receiving data from a base station on a subband or spatial layer or according to a channel tap based on reporting channel state feedback corresponding to an indicated accuracy level. Using these techniques, the base station may adapt the accuracy level for which the UE reports CSI feedback based on one or more configurations in which the base station transmits data to the UE.

[0014] Some examples of the methods, apparatuses, and non-transient computer-readable media described herein may further include operations, features, means, or instructions for receiving a retransmission of the same data that the UE failed to decode based on reporting channel state feedback corresponding to an indicated accuracy level. Some examples of the methods, apparatuses, and non-transient computer-readable media described herein may further include operations, features, means, or instructions for reporting channel state feedback corresponding to an indicated accuracy level by identifying a number of bits for reporting channel state feedback based on an accuracy level, said number of bits which may be directly related to the accuracy level, and by the identified number of bits.

[0015] Some examples of the methods, apparatuses, and non-transient computer-readable media described herein may further include operations, features, means, or instructions for receiving an indication of the number of bits for reporting channel state feedback based on an accuracy level. In some examples of the methods, apparatuses, and non-transient computer-readable media described herein, receiving an indication of an accuracy level may include operations, features, means, or instructions for receiving an indication of an accuracy level in radio resource control (RRC) signaling or in a MAC control element (MAC-CE). Since the UE may identify or select the number of bits for reporting channel state feedback based on an accuracy level, the overhead of reporting channel state feedback may be minimized where appropriate.

[0016] A method of wireless communication at a base station is described. The method may include the steps of transmitting to a UE an indication of an accuracy level for reporting channel state feedback to the base station, transmitting downlink data or reference signals to the UE, and receiving from the UE channel state feedback corresponding to the indication of the accuracy level based on transmitting downlink data or reference signals to the UE.

[0017] An apparatus for wireless communication at a base station is described. The apparatus may include a processor, a memory coupled to the processor, and instructions stored in the memory. The instructions may be executable by the processor to cause the apparatus to transmit an indication of an accuracy level to the UE for reporting channel state feedback to the base station, to transmit downlink data or reference signals to the UE, and to receive channel state feedback from the UE corresponding to the indicated accuracy level based on transmitting downlink data or reference signals to the UE.

[0018] Another device for wireless communication at a base station is described. The device may include means for transmitting to a UE an indication of an accuracy level for reporting channel state feedback to the base station, means for transmitting downlink data or reference signals to the UE, and means for receiving from the UE channel state feedback corresponding to the indication of the accuracy level based on transmitting downlink data or reference signals to the UE.

[0019] A non-transient computer-readable medium for storing code for wireless communication at a base station is described. The code may include instructions executable by a processor to transmit an indication of an accuracy level for reporting channel state feedback to the base station to the UE, transmit downlink data or reference signals to the UE, and receive channel state feedback corresponding to the indicated accuracy level from the UE based on transmitting downlink data or reference signals to the UE.

[0020] In some examples of the methods, apparatuses, and non-transient computer-readable media described herein, transmitting an indication of an accuracy level for reporting channel state feedback may include operations, features, means, or instructions for transmitting an indication of a loss function to be used by the UE to train a pair of neural networks for reporting channel state feedback. Some examples of the methods, apparatuses, and non-transient computer-readable media described herein may further include operations, features, means, or instructions for receiving coefficients of a neural network in a decoder from the UE to decode channel state feedback from the UE, and using the neural network in the decoder to decode channel state feedback from the UE.

[0021] Some examples of the methods, apparatuses, and non-transient computer-readable media described herein may further include actions, features, means, or instructions for transmitting an indication for a UE to train a set of neural network pairs based on a set of accuracy levels, and each neural network pair includes a first neural network in an encoder for encoding channel state feedback and a second neural network in a decoder for decoding channel state feedback. In some examples of the methods, apparatuses, and non-transient computer-readable media described herein, transmitting an indication of an accuracy level may include actions, features, means, or instructions for transmitting an indication for a UE to use a neural network pair among a set of neural network pairs to report channel state feedback.

[0022] Some examples of the methods, apparatuses, and non-transient computer-readable media described herein may further include operations, features, means, or instructions for transmitting a representation of a subset of a set of neural network pairs for training a UE. In some examples of the methods, apparatuses, and non-transient computer-readable media described herein, transmitting a representation of an accuracy level for reporting channel state feedback may include operations, features, means, or instructions for transmitting representations of different accuracy levels for reporting channel state feedback for different subbands, spatial layers, channel taps, or in response to a failure to decode a different number of downlink transmissions containing the same data. In some examples of the methods, apparatuses, and non-transient computer-readable media described herein, the indicated accuracy level may include operations, features, means, or commands for transmitting an indication of a second accuracy level for reporting channel status feedback to be used to schedule a second downlink transmission, and the first accuracy level is different from the second accuracy level.

[0023] Some examples of the methods, apparatuses, and non-transient computer-readable media described herein may further include operations, features, means, or instructions for transmitting an indication of the number of bits to be used by a UE to report channel state feedback based on an accuracy level, said number of bits being directly related to the accuracy level, and receiving channel state feedback corresponding to the indicated accuracy level with the identified number of bits. In some examples of the methods, apparatuses, and non-transient computer-readable media described herein, transmitting an indication of the accuracy level may include operations, features, means, or instructions for transmitting an indication of the accuracy level in RRC signaling or in MAC-CE. Brief explanation of the drawing

[0024] FIG. 1 illustrates an example of a wireless communication system that supports configurable metrics for channel state compression and feedback according to embodiments of the present disclosure. FIG. 2 illustrates an example of CSI feedback encoded by an encoder using a first neural network and decoded by a decoder using a second neural network, according to embodiments of the present disclosure. FIG. 3 illustrates an example of a wireless communication system that supports configurable metrics for channel state compression and feedback according to embodiments of the present disclosure. FIG. 4 illustrates an example of a process flow that supports configurable metrics for channel state compression and feedback according to embodiments of the present disclosure. FIGS. 5 and 6 illustrate block diagrams of devices supporting configurable metrics for channel state compression and feedback according to embodiments of the present disclosure. FIG. 7 illustrates a block diagram of a communication manager supporting configurable metrics for channel state compression and feedback according to embodiments of the present disclosure. FIG. 8 illustrates a diagram of a system including a device that supports configurable metrics for channel state compression and feedback according to embodiments of the present disclosure. FIGS. 9 and FIGS. 10 illustrate block diagrams of devices supporting configurable metrics for channel state compression and feedback according to embodiments of the present disclosure. FIG. 11 illustrates a block diagram of a communication manager supporting configurable metrics for channel state compression and feedback according to embodiments of the present disclosure. FIG. 12 illustrates a diagram of a system including a device that supports configurable metrics for channel state compression and feedback according to embodiments of the present disclosure. FIGS. 13 and 14 illustrate flowcharts illustrating methods supporting configurable metrics for channel state compression and feedback according to embodiments of the present disclosure. Specific details for implementing the invention

[0025] In some wireless communication systems, user equipment (UE) may be configured to perform channel measurements on downlink signals received from a base station and report the channel measurements to the base station. The UE may report the channel measurements as channel state information (CSI) feedback. Using CSI feedback, the base station may identify suitable parameters for downlink transmissions to the UE to improve the likelihood that the downlink transmissions will be received by the UE. The UE may encode the CSI using a first neural network, and the base station may decode the CSI using a second neural network. The first and second neural networks may form a neural network pair, and the UE may train the neural network pair based on an accuracy level (e.g., using a specific loss metric or function). However, in some cases, the UE (115) may be configured to train the neural network pair based on the same accuracy level for all CSI feedback, and it may be inefficient to transmit all CSI feedback at the same accuracy level. For example, if the accuracy level of the reported CSI feedback is unnecessarily high, the overhead of reporting the CSI may also be unnecessarily high. Alternatively, if the accuracy level of the reported CSI feedback is too low, downlink transmissions scheduled using the CSI feedback may be unreliable.

[0026] As described herein, a wireless communication system may support efficient techniques that enable a UE to report CSI to a base station at an appropriate accuracy level. In particular, the UE may be configured to train a pair of neural networks (e.g., including a first neural network in an encoder and a second neural network in a decoder) based on an accuracy level indicated by the base station. For example, the base station may indicate a loss metric or function corresponding to the accuracy level at which the CSI is to be reported by the UE, and the UE may train a pair of neural networks using the loss metric or function. In this way, the UE may report CSI at an accuracy level indicated by the base station. The indicated accuracy level may depend on how the base station intends to use the CSI. For example, a base station may indicate different accuracy levels for CSIs associated with different subbands, channel taps, spatial streams, feedback instances, etc. (e.g., so that latency-sensitive and reliability-sensitive transmits are scheduled based on very accurate CSI feedback). Accordingly, if a higher accuracy level is appropriate, the UE may report the CSI at a higher accuracy level. Otherwise, the UE may report the CSI at a lower accuracy level.

[0027] The embodiments of the present disclosure introduced above are described below in the context of wireless communication systems. Then, examples of processes and signaling exchanges supporting configurable metrics for channel state compression and feedback are described. The embodiments of the present disclosure are further illustrated and described with reference to device diagrams, system diagrams, and flowcharts related to configurable metrics for channel state compression and feedback.

[0028] FIG. 1 illustrates an example of a wireless communication system (100) that supports configurable metrics for channel state compression and feedback according to embodiments of the present disclosure. The wireless communication system (100) may include one or more base stations (105), one or more UEs (115), and a core network (130). In some examples, the wireless communication system (100) may be a Long Term Evolution (LTE) network, an LTE-Advanced (LTE-A) network, an LTE-A Pro network, or a New Radio (NR) network. In some examples, the wireless communication system (100) may support Mobile Broadband (MBB) communication, Enhanced MBB (eMBB) communication, Ultra-Reliable (e.g., Mission-Critical) communication, Low Latency communication, communication with low-cost and low-complexity devices, or any combination thereof.

[0029] Base stations (105) may be scattered across a geographical area to form a wireless communication system (100) and may be devices having different forms or different capabilities. Base stations (105) and UEs (115) may communicate wirelessly through one or more communication links (125). Each base station (105) may provide a coverage area (110) in which UEs (115) and base station (105) may establish one or more communication links (125). The coverage area (110) may be an example of a geographical area in which base stations (105) and UEs (115) may support the communication of signals according to one or more wireless access technologies.

[0030] UEs (115) may be scattered throughout the coverage area (110) of the wireless communication system (100), and each UE (115) may be stationary, mobile, or both at different times. UEs (115) may be devices having different forms or different capabilities. Some exemplary UEs (115) are illustrated in FIG. 1. The UEs (115) described herein may be able to communicate with various types of devices, such as other UEs (115), base stations (105), or network equipment (e.g., core network nodes, repeater devices, integrated access and backhaul (IAB) nodes, or other network equipment), as illustrated in FIG. 1.

[0031] Base stations (105) may communicate with the core network (130), with each other, or with both. For example, base stations (105) may interface with the core network (130) through one or more backhaul links (120) (e.g., through S1, N2, N3, or other interfaces). Base stations (105) may communicate with each other over the backhaul links (120) (e.g., through X2, Xn, or other interfaces) directly (e.g., directly between base stations (105)), indirectly (e.g., through the core network (130)), or both. In some examples, the backhaul links (120) may be one or more wireless links or may include them.

[0032] One or more of the base stations (105) described in this specification may include a base transceiver station, a wireless base station, an access point, a wireless transceiver, a node B, an e-node B (eNB), a next-generation node B or a gigabit node B (any of these may be referred to as a gNB), a home node B, a home e-node B, or other suitable terms, or may be referred to by those skilled in the art as such.

[0033] The UE (115) may include or be referred to as a mobile device, wireless device, remote device, handheld device, or subscriber device, or some other suitable term, wherein “device” may also be referred to as a unit, station, terminal, or client, among other examples. The UE (115) may also include or be referred to as a personal electronic device such as a cellular phone, personal digital assistant (PDA), tablet computer, laptop computer, or personal computer. In some examples, the UE (115) may include or be referred to as a wireless local loop (WLL) station, Internet of Things (IoT) device, Internet of Everything (IoE) device, or machine type communication (MTC) device, among other examples, which may be implemented in various objects such as appliances, or vehicles, instruments, etc., among other examples.

[0034] The UEs (115) described in this specification may be able to communicate with various types of devices, such as network equipment and base stations (105) including, among other examples as illustrated in FIG. 1, macro eNBs or gNBs, small cell eNBs or gNBs, or repeater base stations, as well as other UEs (115) that may sometimes operate as repeaters.

[0035] UEs (115) and base stations (105) may communicate wirelessly with each other over one or more communication links (125) on one or more carriers. The term “carrier” may refer to a set of radio frequency spectrum resources having a defined physical layer structure to support the communication links (125). For example, a carrier used for a communication link (125) may include a portion of a radio frequency spectrum band (e.g., Bandwidth Part (BWP)) that operates according to one or more physical layer channels for a given radio access technology (e.g., LTE, LTE-A, LTE-A Pro, NR). Each physical layer channel may carry capture signaling (e.g., synchronization signals, system information), control signaling that coordinates the operation for the carrier, user data, or other signaling. The wireless communication system (100) may support communication with the UE (115) using carrier aggregation or multi-carrier operation. The UE (115) may be composed of multiple downlink component carriers and one or more uplink component carriers depending on the carrier aggregation configuration. Carrier aggregation may be used with both frequency division duplexing (FDD) and time division duplexing (TDD) component carriers.

[0036] Signal waveforms transmitted over a carrier may consist of multiple subcarriers (e.g., using multi-carrier modulation (MCM) techniques such as orthogonal frequency division multiplexing (OFDM) or discrete Fourier transform spread OFDM (DFT-S-OFDM). In a system employing MCM techniques, a resource element may consist of one symbol period (e.g., the duration of one modulation symbol) and one subcarrier, wherein the symbol period and subcarrier spacing are inversely related. The number of bits carried by each resource element may depend on the modulation scheme (e.g., the order of the modulation scheme, the coding rate of the modulation scheme, or both). Thus, the more resource elements the UE (115) receives and the higher the order of the modulation scheme, the higher the data rate for the UE (115). Wireless communication resources may refer to a combination of wireless frequency spectrum resources, time resources, and spatial resources (e.g., spatial layers or beams), and the use of multiple spatial layers may further increase the data rate or data integrity for communication with the UE (115).

[0037] The time intervals for base stations (105) or UEs (115) are, for example, T s = 1 / (△f max ·N f It may also be expressed as multiples of a basic time unit that may refer to a sampling period of ) seconds, where △f max may represent the maximum supported subcarrier spacing, and N fmay also represent the maximum supported Discrete Fourier Transform (DFT) size. Time intervals of communication resources may be organized according to radio frames, each having a specified duration (e.g., 10 milliseconds (ms)). Each radio frame may be identified by a system frame number (SFN) (e.g., ranging from 0 to 1023).

[0038] Each frame may contain multiple consecutively numbered subframes or slots, and each subframe or slot may have the same duration. In some examples, a frame may be divided into subframes (e.g., in the time domain), and each subframe may be further divided into multiple slots. Alternatively, each frame may contain a variable number of slots, and the number of slots may depend on subcarrier spacing. Each slot may contain multiple symbol periods (e.g., depending on the length of the cyclic prefix pre-defended for each symbol period). In some wireless communication systems (100), a slot may be further divided into multiple mini-slots containing one or more symbols. Excluding the cyclic prefix, each symbol period may contain one or more (e.g., N f It may include sampling periods. The duration of the symbol period may depend on the frequency band or subcarrier spacing of the operation.

[0039] A subframe, slot, mini-slot, or symbol may be the minimum scheduling unit of the wireless communication system (100) (e.g., in the time domain) and may be referred to as a transmit time interval (TTI). In some examples, the TTI duration (e.g., the number of symbol cycles in the TTI) may be variable. Additionally or alternatively, the minimum scheduling unit of the wireless communication system (100) may be dynamically selected (e.g., in bursts of shortened TTIs (sTTIs).

[0040] Physical channels may be multiplexed on a carrier according to various techniques. Physical control channels and physical data channels may be multiplexed on a downlink carrier using, for example, one or more of time division multiplexing (TDM) techniques, frequency division multiplexing (FDM) techniques, or hybrid TDM-FDM techniques. A control domain for a physical control channel (e.g., a control resource set (CORESET)) may be defined by a number of symbol periods and may extend across the system bandwidth or a subset of the carrier's system bandwidth. One or more control domains (e.g., CORESETs) may be configured for a set of UEs (115). For example, one or more of the UEs (115) may monitor or search for control domains for control information according to one or more search space sets, and each search space set may include one or more control channel candidates in one or more aggregation levels arranged in a cascaded manner. The aggregation level for control channel candidates may refer to the number of control channel resources (e.g., control channel elements (CCEs)) associated with encoded information for a control information format having a given payload size. The search space sets may include common search space sets configured to transmit control information to multiple UEs (115) and UE-specific search space sets for transmitting control information to a specific UE (115).

[0041] In some examples, the base station (105) is mobile and thus may provide communication coverage for a moving geographical coverage area (110). In some examples, different geographical coverage areas (110) associated with different technologies may overlap, but different geographical coverage areas (110) may be supported by the same base station (105). In other examples, overlapping geographical coverage areas (110) associated with different technologies may be supported by different base stations (105). The wireless communication system (100) may include a heterogeneous network in which base stations (105) of different types provide coverage for various geographical coverage areas (110) using the same or different wireless access technologies.

[0042] The wireless communication system (100) may be configured to support ultra-reliable communication or low-latency communication, or various combinations thereof. For example, the wireless communication system (100) may be configured to support ultra-reliable low-latency communication (URLLC) or mission-critical communication. UEs (115) may be designed to support ultra-reliable, low-latency, or critical functions (e.g., mission-critical functions). Ultra-reliable communication may include private communication or group communication, and may be supported by one or more mission-critical services such as mission-critical push-to-talk (MCPTT), mission-critical video (MCVideo), or mission-critical data (MCData). Support for mission-critical functions may include the prioritization of services, and mission-critical services may be used for public safety or general commercial applications. The terms 'ultra-reliable', 'low latency', 'mission critical', and 'ultra-reliable low latency' may be used interchangeably in this specification.

[0043] In some examples, the UE (115) may also be able to communicate directly with other UEs (115) over a D2D communication link (135) (e.g., using a peer-to-peer (P2P) or device-to-device (D2D) protocol). One or more UEs (115) utilizing D2D communication may be within the geographical coverage area (110) of the base station (105). Other UEs (115) in such a group may be outside the geographical coverage area (110) of the base station (105) or otherwise may not be able to receive transmissions from the base station (105). In some examples, groups of UEs (115) communicating via D2D communication may utilize a one-to-many (1:M) system, where each UE (115) transmits to all other UEs (115) in the group. In some examples, the base station (105) facilitates the scheduling of resources for D2D communication. In other cases, D2D communication is performed between UEs (115) without the involvement of the base station (105).

[0044] The core network (130) may provide user authentication, access authorization, tracking, Internet Protocol (IP) connectivity, and other access, routing, or mobility functions. The core network (130) may be an advanced packet core (EPC) or a 5G core (5GC), and may include at least one control plane entity that manages access and mobility (e.g., a mobility management entity (MME), an access and mobility management function unit (AMF)) and at least one user plane entity that routes packets or interconnects to external networks (e.g., a serving gateway (S-GW), a packet data network (PDN) gateway (P-GW), or a user plane function unit (UPF)). The control plane entity may manage non-access stratum (NAS) functions, such as mobility, authentication, and bearer management for UEs (115) served by base stations (105) associated with the core network (130). User IP packets may be transmitted through a user plane entity that may provide other functions as well as IP address allocation. The user plane entity may be connected to network operator IP services (150). The operator IP services (150) may include access to the Internet, intranet(s), IP multimedia subsystem (IMS), or packet switching streaming services.

[0045] Some of the network devices, such as base stations (105), may include subcomponents, such as access network entities (140), which may be examples of access node controllers (ANCs). Each access network entity (140) may communicate with UEs (115) through one or more other access network transmitting entities (145), and those other access network transmitting entities may be referred to as radio heads, smart radio heads, or transmit / receive points (TRPs). Each access network transmitting entity (145) may include one or more antenna panels. In some configurations, the various functions of each access network entity (140) or base station (105) may be distributed across various network devices (e.g., radio heads and ANCs) or integrated into a single network device (e.g., base station (105)).

[0046] A wireless communication system (100) may typically operate using one or more frequency bands in the range of 300 megahertz (MHz) to 300 gigahertz (GHz). Generally, the range from 300 MHz to 3 GHz is known as the ultra-high frequency (UHF) region or decimeter band because the wavelengths range in length from approximately 1 decimeter to 1 meter. Although UHF waves may be blocked or redirected by buildings and environmental features, the waves may penetrate structures sufficiently to provide service to UEs (115) located indoors by macro cells. Transmission of UHF waves may be associated with smaller antennas and shorter ranges (e.g., less than 100 kilometers) compared to transmission using smaller frequencies and longer waves in the high frequency (HF) or very high frequency (VHF) portions of the spectrum below 300 MHz.

[0047] A wireless communication system (100) may utilize both licensed and unlicensed radio frequency spectrum bands. For example, a wireless communication system (100) may employ Licensed Assisted Access (LAA), LTE Unlicensed (LTE-U) radio access technology, or NR technology in an unlicensed band, such as the 5 GHz Industrial Scientific and Medical (ISM) band. When operating in unlicensed radio frequency spectrum bands, devices such as base stations (105) and UEs (115) may employ carrier sensing for collision detection and avoidance. In some examples, operations in unlicensed bands may be based on carrier aggregation configurations with component carriers operating in licensed bands (e.g., LAA). Operations in the unlicensed spectrum may include, among other examples, downlink transmissions, uplink transmissions, P2P transmissions, or D2D transmissions.

[0048] A base station (105) or UE (115) may be equipped with multiple antennas, and these multiple antennas may be used to employ techniques such as transmit diversity, receive diversity, multiple input multiple output (MIMO) communication, or beamforming. The antennas of the base station (105) or UE (115) may be located within one or more antenna arrays or antenna panels that may support MIMO operations or transmit or receive beamforming. For example, one or more base station antennas or antenna arrays may be placed in an antenna assembly such as an antenna tower. In some examples, the antennas or antenna arrays associated with the base station (105) may be located at various geographical locations. The base station (105) may have an antenna array having multiple rows and columns of antenna ports that the base station (105) may use to support beamforming of communication with the UE (115). Likewise, the UE (115) may have one or more antenna arrays that may support various MIMO or beamforming operations. Additionally or alternatively, the antenna panel may support radio frequency beamforming for signals transmitted through the antenna ports.

[0049] Beamforming, which may also be referred to as spatial filtering, directional transmission, or directional reception, is a signal processing technique that may be used at a transmitting device or a receiving device (e.g., base station (105), UE (115)) to shape or steer an antenna beam (e.g., transmit beam, receive beam) along a spatial path between a transmitting device and a receiving device. Beamforming may be achieved by combining signals communicated through antenna elements of an antenna array such that some signals propagating toward specific orientations experience constructive interference while others experience destructive interference. The adjustment of signals communicated through antenna elements may include the transmitting device or the receiving device applying amplitude offsets, phase offsets, or both to the signals being returned through the antenna elements associated with the device. The adjustments associated with each of the antenna elements may be defined by a set of beamforming weights associated with a specific orientation (e.g., for the antenna array of the transmitting device or the receiving device or for some other orientation).

[0050] In a wireless communication system (100), a UE (115) may be configured to perform channel measurements on downlink signals received from a base station (105) and report the channel measurements to the base station (105). The UE (115) may report the channel measurements as CSI feedback. Using CSI feedback, the base station (105) may identify suitable parameters for downlink transmissions to the UE (115) to improve the likelihood that the downlink transmissions will be received by the UE (115). In some cases, the UE (115) may use a first neural network to encode the CSI, and the base station (105) may use a second neural network to decode the CSI, wherein the first and second neural networks form a neural network pair.

[0051] FIG. 2 illustrates an example of CSI feedback (200) encoded by an encoder (205) using a first neural network and decoded by a decoder (210) using a second neural network, according to embodiments of the present disclosure. In the example of FIG. 2, the UE (115) may input channel recognition information to the encoder (205), and the encoder (205) may generate CSI feedback by encoding the channel recognition information using the first neural network. The channel recognition information may refer to a raw channel and may correspond to measurements performed on CSI reference signals (CSI-RS) received on the channel. Thus, the encoder may take the raw channel as input, and the UE (115) may use the encoder neural network to generate and feedback the CSI. The UE (115) may transmit CSI feedback to the base station (105), and the base station (105) may input the CSI feedback to the decoder (210). The decoder (210) may decode the CSI feedback using a second neural network to obtain the channel state (for example, the base station (105) may use a decoder neural network to restore the raw channel state from the CSI feedback). Then, the base station (105) may use the channel state to identify suitable parameters for downlink transmissions to the UE (115).

[0052] In some wireless communication systems, the UE (115) may train a pair of neural networks including a first neural network in the encoder and a second neural network in the decoder based on a fixed accuracy level. Training the neural networks may involve unsupervised learning, supervised learning, or a combination of both. For example, the UE (115) may train the encoder (205) according to one or more machine learning algorithms in the neural network. The neural networks in the encoder (205), the decoder (210), or both may include any number of machine learning layers (e.g., convolutional layers, fully connected layers, or some combination thereof). The UE (115) may implement any machine learning techniques to train the neural networks in the encoder (205), the decoder (210), or both. For example, UE (115) may implement deep learning (e.g., using a deep recurrent network), backpropagation, linear regression, K-means model, random forest model, or any combination of these or other machine learning techniques to train one or both of the neural networks.

[0053] In some machine learning examples, a network may train a machine learning model on a set of training data. The training data may be a subset of a larger dataset. In some cases, training may involve determining one or more target features from the dataset. Subsequently, the model may learn one or more features from the training data and evaluation metrics such as Mean Squared Error (MSE), precision, accuracy, and recall (based on linear regression techniques, for example, linear regression algorithms). In some cases, evaluation metrics may be calculated based on a loss function.

[0054] Neural network training may be iterative, so the UE (115) trains the neural network based on the current version of the neural network and measurements obtained since the current version of the neural network was implemented (e.g., rather than starting training from scratch using the entire set of historical measurements). Such iterative training processes may reduce the processing overhead associated with training neural networks and may reduce the amount of historical measurement information that the UE (115) stores for neural network training. During the training process, the UE may apply layers to measurement inputs or channel recognition to compress data from one or more base stations, sensors, radio access technologies (RATs), etc. The UE may feed back the compressed data through the decoder (210) to determine the number of decoding coefficients or parameters for the decoder neural network.

[0055] During training, the UE (115) may update encoder weights, encoder layers, decoder weights, decoder layers, or some combination thereof based on feedback information. For example, the UE (115) may update encoder weights based on performance metrics for encoding. Such performance metrics may be metrics measuring the level of compression achieved by the encoder neural network (e.g., comparing the number of bits associated with encoded CSI feedback with the number of bits associated with input measurements or channel recognition), metrics measuring the confidence in extracting input measurements from the encoder output using the decoder neural network, metrics measuring the computational complexity associated with compression, metrics measuring system performance based on the encoder neural network, or some combination thereof. Similarly, the UE may update decoder weights based on performance metrics for decoding. Such performance metrics may be metrics that measure similarity between output measurements and input measurements, metrics that measure computational complexity associated with decompression, metrics that measure system performance based on encoder output, or some combination thereof.

[0056] In some cases, the UE (115) may be configured with a fixed loss metric or loss function for training the encoder and decoder neural network pairs. However, the desired loss metric or loss function for training the encoder and decoder neural network pairs to report CSI feedback to the base station (105) may depend on how the base station (105) intends to use the CSI feedback. As an example, for single-user MIMO (SU-MIMO), the base station (105) may mostly be concerned with learning precoding directions, whereas for multi-user MIMO (MU-MIMO), the base station (105) may be concerned with learning raw channel states. Additionally, the desired feedback accuracy may differ in different subbands or different feedback instances (e.g., for eMBB communications versus URLLC or for MU-MIMO for specific subbands versus other subbands). Therefore, it may be inefficient for the UE (115) to transmit all CSI feedback at the same accuracy level.

[0057] The wireless communication system (100) may support efficient techniques that allow the UE (115) to report CSI to the base station at an appropriate (e.g., dynamically changed or configurable) accuracy level. For example, the UE (115) may receive an indication of a loss function or loss metric corresponding to the accuracy level, and the UE (115) may use the loss function or loss metric to compute the loss when training the encoder and decoder neural network pairs. The loss function may be a function used to compute the loss (e.g., a measure of the difference between the input to the neural network pair and the output from the neural network pair), and the loss metric may correspond to a metric used in the loss function. Based on the loss computed in one iteration of training, the UE (115) may adjust the coefficients in the neural network pair to minimize the loss computed in future iterations of training. For example, a neural network pair may implement activation functions for each layer of the network (e.g., for hidden layers between the input and output layers). A neural network pair may also implement a loss function or cost function based on the difference between the actual value and the predicted value. For each layer of a neural network pair, the cost function may be used to adjust weights for the next input based on a loss metric. In some examples, the cost function or loss function may implement an MSE function that calculates the square of the difference between the actual value and the predicted value. Thus, the loss function and the loss metric may differ from the loss. Additionally, the loss function may be used to minimize the difference between the input and output of the neural network pair or the difference between the modes of the input and output of the neural network pair.

[0058] Since the loss function may correspond to an accuracy level, any pair of neural networks trained using the loss function may encode and decode CSI feedback to the corresponding accuracy level. Thus, reporting CSI feedback corresponding to the accuracy level may refer to reporting CSI feedback encoded and decoded using a pair of neural networks trained using a loss function corresponding to the accuracy level. In some cases, the UE (115) may also be configured to train multiple pairs of neural networks using multiple loss functions, each corresponding to an accuracy level. In such cases, the UE (115) may receive an indication of which of the multiple pairs of neural networks the UE (115) will use to report CSI. Since the pair of neural networks may be trained using a loss function corresponding to an accuracy level, the indication of the pair of neural networks may correspond to the indication of the accuracy level. Therefore, reporting CSI feedback corresponding to the accuracy level may refer to reporting CSI feedback encoded and decoded using a pair of neural networks among a plurality of pairs of neural networks trained using a loss function corresponding to the accuracy level.

[0059] Additionally, since the UE (115) may train both the neural network in the encoder and the neural network in the decoder (e.g., the encoder and decoder may be at the UE (115), and the decoder may also be at the base station (105)), the UE (115) may transmit the coefficients of the decoder neural network to the base station (105). Accordingly, signaling between the UE (115) and the base station (105) may be minimized because the UE (115) may not need to receive the output of the decoder from the base station (105) (e.g., for each iteration of training). Once the UE (115) has finished training the pair of neural networks, the UE (115) may transmit the coefficients of the decoder neural network to the base station (105). Next, the UE (115) may use an encoder to encode the CSI feedback and report the CSI feedback to the base station (105), and the base station (105) may use a decoder to decode the CSI feedback based on the coefficients received from the UE (115). Using the techniques described herein, the UE (115) may be able to report the CSI at an appropriate level of accuracy to minimize unnecessary overhead while allowing the base station (105) to identify appropriate parameters for communications with the UE (115).

[0060] FIG. 3 illustrates an example of a wireless communication system (300) that supports configurable metrics for channel state compression and feedback according to embodiments of the present disclosure. The wireless communication system (300) includes a UE (115-a) which may be an example of the UE (115) described with reference to FIG. 1 and FIG. 2. The wireless communication system (300) also includes a base station (105-a) which may be an example of the base station (105) described with reference to FIG. 1 and FIG. 2. The base station (105-a) may provide communication coverage for a coverage area (110-a). The wireless communication system (300) may implement embodiments of the wireless communication system (100). For example, the wireless communication system (300) may support efficient techniques that allow the UE (115-a) to report CSI to the base station (105-a) at an appropriate accuracy level.

[0061] In the example of FIG. 3, the base station (105-a) may transmit an indication of an accuracy level (305) for the UE (115-a) to report the CSI feedback (310) to the base station (105-a), and the UE (115-a) may report the CSI feedback (310) to the base station (105-a) based on the indicated accuracy level (305). For example, the base station (105-a) may transmit an indication of a loss function to the UE (115-a), and the UE (115-a) may train a pair of encoder and decoder neural networks to encode and decode the CSI feedback (310) using the loss function. That is, the base station (105-a) may configure the UE (115-a) to use a specific loss function for training the encoder and decoder neural networks to report CSI feedback (310). The UE (115-a) may train the encoder and decoder neural network pairs using actual data (e.g., CSI based on actual channel measurements) or other data (e.g., provided by the base station (105-a)). To train the neural network pairs, the UE (115-a) may input CSI into the neural network pairs, and the UE (115-a) may compare the CSI output from the neural network pairs with the CSI input into the neural network pairs. Next, UE (115-a) may adjust the coefficients (e.g., weights) in the neural network pair to minimize the loss (e.g., to minimize the difference between the CSI input to the neural network pair and the CSI output from the neural network pair).

[0062] Because the accuracy level (305) or loss function may be configured by the base station (105-a), the accuracy of the CSI feedback (310) reported by the UE (115-a) may be aligned with the intentions of the base station (105-a). For example, the base station (105-a) may configure the UE (115-a) to report more accurate (e.g., very accurate) CSI feedback (310) for latency-sensitive or reliability-sensitive communications, and the base station (105-a) may configure the UE (115-a) to report less accurate CSI feedback (310) for other communications. In some cases, different accuracy levels or loss functions may be configured for different subbands, channel taps, spatial streams, or feedback instances. In particular, the base station (105-a) may configure different accuracy levels (e.g., different (relative) weightings for the loss function or different (relative) accuracy targets) across different subbands, channel taps, spatial streams, and feedback instances. Different subbands may be used for different types of communications (e.g., eMBB communications and URLLC), and if the UE (115-a) is operating on a time domain channel, different channel taps may be compressed to different degrees of accuracy, different spatial streams may correspond to transmits on different beams (e.g., higher accuracy or higher weighting for the strongest beam direction), and different feedback instances may correspond to feedback of different rounds (e.g., feedback of the first round for URLLC and feedback of the second round for URLLC).

[0063] The base station (105-a) may transmit an indication of the accuracy level (305) via an upper layer message (e.g., RRC signaling) or dynamic signaling (e.g., in a MAC control element (MAC-CE)). Additionally, in some cases, if quantization is performed on the CSI feedback (310) prior to transmission to the base station (105-a), some paths, subbands, channel taps, spatial streams, feedback instances, etc., may be compressed using a larger number of bits. For example, the UE (115-a) may transmit the CSI feedback (310) using a different number of bits depending on the accuracy level (305) indicated by the base station (105-a). In such cases, the base station (105-a) may transmit an indication of the number of bits that the UE (115-a) will use to transmit CSI feedback (310), and the UE (115-a) may transmit CSI feedback (310) with the indicated number of bits.

[0064] The base station (105-a) may also request that the UE (115-a) train multiple encoder and decoder neural network pairs (e.g., N neural network pairs), and the base station (105-a) may configure an accuracy level for each of the neural network pairs (e.g., for a trade-off between feedback accuracy and CSI feedback overhead). Then, the base station (105-a) may transmit a designation of one of the multiple neural network pairs to be used by the UE (115-a) to report CSI feedback (310). Alternatively, the UE (115-a) may autonomously select one of the multiple neural network pairs to be used to report CSI feedback (310) (e.g., without signaling from the base station (105-a). In some cases, instead of transmitting a representation of a loss metric or loss function for training a pair of neural networks, the base station (105-a) may transmit a representation of an expression for training a pair of neural networks. In other cases, the base station (105-a) may transmit a representation of a set of loss metrics or loss functions for training a pair of neural networks. In such cases, the base station (105-a) may transmit to the UE (115-a) a representation of a loss metric or loss function from a set to be used by the UE to train a pair of neural networks. Alternatively, the UE (115-a) may autonomously select a loss metric or loss function from a set to be used to train a pair of neural networks (e.g., based on a represented accuracy level).

[0065] FIG. 4 illustrates an example of a process flow (400) that supports configurable metrics for channel state compression and feedback according to embodiments of the present disclosure. The process flow (400) illustrates embodiments of techniques performed by a UE (115-b), which may be an example of a UE (115) described with reference to FIG. 1 through 3. The process flow (400) also illustrates embodiments of techniques performed by a base station (105-b), which may be an example of a base station (105) described with reference to FIG. 1 through 3. The process flow (400) may implement embodiments of a wireless communication system (300). For example, the process flow (400) may support efficient techniques that allow the UE (115-b) to report CSI to the base station (105-b) at an appropriate accuracy level. UE (115-b) may also support the ability to perform channel state compression and feedback differently for different subbands, channel taps, spatial streams, feedback instances, etc.

[0066] The accuracy level may refer to the difference between the raw CSI (or measured actual channel conditions) input as an input to the encoder in the UE (115-b) and the CSI generated by the decoder in the base station (105-b). For example, a high accuracy level may indicate a small difference or no difference between the raw CSI (or measured actual channel conditions) input as an input to the encoder and the CSI generated by the decoder, and a low accuracy level may indicate a large difference between the raw CSI (or measured actual channel conditions) input as an input to the encoder and the CSI generated by the decoder. That is, the higher the accuracy level, the lower the amount of compression, and vice versa. Additionally, in some cases, when the base station (105-b) signals the accuracy level, the base station (105-b) may signal one or more aspects of the CSI that the UE (115-b) will focus on when measuring and reporting the CSI. Next, the loss function may prioritize or apply greater weights to one or more aspects of the CSI indicated by the base station (105-b) when training the neural network pair. In this way, the UE (115-b) may be able to generate and report CSI feedback that aligns with the way the base station (105-b) intends to use the CSI feedback. In such cases, the accuracy level may refer to the priorities or weights applied to those aspects of the CSI preferred by the base station (105).

[0067] In 405, the UE (115-b) may identify a pair of neural networks including a first neural network in an encoder for encoding channel state feedback and a second neural network in a decoder for decoding channel state feedback. For example, the base station (105-b) may indicate the number of pairs of neural networks for the UE (115-b) to train. In some cases, the UE (115-b) may select a pair of neural networks based on that indication. In 410, the UE (115-b) may receive an indication of the accuracy level for reporting CSI feedback to the base station (105-b) (e.g., in an RRC or MAC-CE message). That is, the base station (105-b) and the UE (115-b) may support differentiated accuracy for CSI feedback. As an example, the UE (115-b) may receive an indication of a loss metric or loss function from the base station (105-b), and in 415, the UE (115-b) may train a pair of neural networks using the loss metric or loss function (e.g., where the loss metric or loss function implicitly indicates an accuracy level). In 420, the UE (115-b) may transmit coefficients of a second neural network in a decoder to the base station (105-b) based on training the pair of neural networks.

[0068] In some cases, the indicated accuracy level (or loss metric) may be based on one or more of the subband, spatial layer, or channel tap corresponding to the CSI feedback. That is, the base station (105-b) may indicate loss metrics or loss functions and weights to the UE (115-b) to train pairs of neural networks for different subbands, channel taps, spatial streams, or feedback instances. In such cases, the UE (115-b) may receive data from the base station (105-b) on the subband or spatial layer or according to the channel tap based on reporting the CSI feedback corresponding to the indicated accuracy level. Additionally or alternatively, the indicated accuracy level may be based on the number of downlink transmissions containing the same data (e.g., feedback instances or rounds of feedback) that the UE (115-b) failed to decode. In such cases, the UE (115-b) may receive a retransmission of the same data that the UE failed to decode based on reporting CSI feedback corresponding to the indicated accuracy level. For example, if the UE (115-b) fails to decode multiple retransmissions of the same data, the UE (115-b) may report very accurate CSI feedback. Accordingly, the base station (105-b) may retransmit the same data based on very accurate CSI feedback, and the likelihood that the UE (115-b) can successfully decode the retransmission of the same data may be high.

[0069] In some cases, the UE (115-b) may receive an indication to train multiple pairs of neural networks based on multiple loss metrics or loss functions (e.g., multiple accuracy levels), and the UE (115-b) may train each of the multiple pairs of neural networks based on an individual accuracy level among the multiple accuracy levels. In such cases, the base station (105-b) may transmit and the UE (115-b) may receive an indication of which of the pairs of neural networks the UE (115-b) will use to report CSI feedback to the base station (105-b). Alternatively, the UE (115-b) may autonomously select one of the multiple pairs of neural networks to report CSI feedback to the base station (105-b). Additionally, the UE (115-b) may receive an indication of a subset of multiple neural network pairs that the UE (115-b) will train. That is, the base station (105-b) may transmit the number of neural network pairs to be trained in the UE (115-b) and which of the neural network pairs the UE (115-b) will use to report CSI feedback for a specific subband, channel tap, spatial stream, or feedback instance.

[0070] In 425, the base station (105-b) may transmit downlink data or reference signals (e.g., CSI-RS) to the UE (115-b), and the UE (115-b) may perform channel measurements to generate CSI feedback based on the downlink data or reference signals received from the base station (105-b). In 430, the UE (115-b) may encode channel state feedback using the first neural network in the neural network pair, and in 435, the UE (115-b) may report CSI feedback (e.g., corresponding to an indicated accuracy level) to the base station (105-b). In some cases, the UE (115-b) may identify the number of bits for reporting CSI feedback based on the indicated accuracy level, and the UE (115-b) may report CSI feedback corresponding to the indicated accuracy level with the identified number of bits. The UE (115-b) may receive an indication of the number of bits for reporting CSI feedback based on the accuracy level. That is, the base station (105-b) may indicate to the UE (115-b) the number of bits for quantizing CSI feedback for different subbands, channel taps, spatial streams, or feedback instances. In 440, the base station (105-b) may use a second neural network (e.g., based on receiving decoder coefficients in 420) to decode the channel state feedback received from the UE (115-b).

[0071] FIG. 5 illustrates a block diagram (500) of a device (505) that supports configurable metrics for channel state compression and feedback according to embodiments of the present disclosure. The device (505) may be an example of embodiments of a UE (115) as described herein. The device (505) may include a receiver (510), a communication manager (515), and a transmitter (520). The device (505) may also include a processor. Each of these components may communicate with one another (e.g., via one or more buses).

[0072] The receiver (510) may receive information such as packets associated with various information channels, user data, or control information (e.g., information related to control channels, data channels, and configurable metrics for channel state compression and feedback). The information may be transmitted to other components of the device (505). The receiver (510) may be an example of an embodiment of the transceiver (820) described with reference to FIG. 8. The receiver (510) may utilize a single antenna or a set of antennas.

[0073] The communication manager (515) receives from the base station an indication of an accuracy level for reporting channel status feedback to the base station, receives downlink data or reference signals from the base station, and may report channel status feedback corresponding to the indicated accuracy level to the base station based at least partially on the downlink data or reference signals. The communication manager (515) may be an example of an embodiment of the communication manager (810) described herein.

[0074] The communication manager (515) or its sub-components may be implemented in hardware, code executed by a processor (e.g., software or firmware), or any combination thereof. When implemented in code executed by a processor, the functions of the communication manager (515) or its sub-components may be implemented by a general-purpose processor, a DSP, an application-specific integrated circuit (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 in this disclosure.

[0075] The communication manager (515) or its sub-components may be physically located in various positions, including distributed so that parts of the functions are implemented at different physical locations by one or more physical components. In some examples, the communication manager (515) or its sub-components may be separate and distinct components according to various aspects of the present disclosure. In some examples, the communication manager (515) or its sub-components may be combined with one or more other hardware components, including but not limited to input / output (I / O) components, transceivers, network servers, other computing devices, one or more other components described in the present disclosure, or combinations thereof according to various aspects of the present disclosure.

[0076] The transmitter (520) may transmit signals generated by other components of the device (505). In some examples, the transmitter (520) may be juxtaposed with the receiver (510) in the transceiver module. For example, the transmitter (520) may be an example of an embodiment of the transceiver (820) described with reference to FIG. 8. The transmitter (520) may utilize a single antenna or a set of antennas.

[0077] FIG. 6 illustrates a block diagram (600) of a device (605) that supports configurable metrics for channel state compression and feedback according to embodiments of the present disclosure. The device (605) may be an example of embodiments of the device (505) or UE (115) as described herein. The device (605) may include a receiver (610), a communication manager (615), and a transmitter (640). The device (605) may also include a processor. Each of these components may communicate with one another (e.g., via one or more buses).

[0078] The receiver (610) may receive information such as packets associated with various information channels, user data, or control information (e.g., information related to control channels, data channels, and configurable metrics for channel state compression and feedback). The information may be transmitted to other components of the device (605). The receiver (610) may be an example of an embodiment of the transceiver (820) described with reference to FIG. 8. The receiver (610) may utilize a single antenna or a set of antennas.

[0079] The communication manager (615) may be an example of an embodiment of the communication manager (515) as described in this specification. The communication manager (615) may include a CSI accuracy manager (620), a downlink manager (625), and a CSI reporter (630). The communication manager (615) may be an example of an embodiment of the communication manager (810) described in this specification.

[0080] The CSI accuracy manager (620) may receive an indication of the accuracy level from the base station for reporting channel status feedback to the base station. The downlink manager (625) may receive downlink data or reference signals from the base station. The CSI reporter (630) may report channel status feedback corresponding to the indicated accuracy level to the base station based on the downlink data or reference signals.

[0081] The transmitter (640) may transmit signals generated by other components of the device (605). In some examples, the transmitter (640) may be juxtaposed with the receiver (610) in the transceiver module. For example, the transmitter (640) may be an example of an embodiment of the transceiver (820) described with reference to FIG. 8. The transmitter (640) may utilize a single antenna or a set of antennas.

[0082] FIG. 7 illustrates a block diagram (700) of a communication manager (705) that supports configurable metrics for channel state compression and feedback according to embodiments of the present disclosure. The communication manager (705) may be an example of embodiments of the communication manager (515), the communication manager (615), or the communication manager (810) described herein. The communication manager (705) may include a CSI accuracy manager (710), a downlink manager (715), a CSI manager (720), a CSI reporter (725), a neural network manager (730), an encoder (735), and a loss function manager (740). Each of these modules may communicate directly or indirectly with one another (e.g., through one or more buses).

[0083] The CSI accuracy manager (710) may receive an indication of the accuracy level from the base station to report channel status feedback to the base station. The downlink manager (715) may receive downlink data or reference signals from the base station. The CSI reporter (725) may report channel status feedback corresponding to the indicated accuracy level to the base station based on the downlink data or reference signals.

[0084] The loss function manager (740) may receive an indication of a loss function corresponding to an accuracy level for training a pair of neural networks, and the pair of neural networks includes a first neural network in an encoder for encoding channel state feedback and a second neural network in a decoder for decoding channel state feedback. The neural network manager (730) may train the pair of neural networks using a loss function. The neural network manager (730) may repeatedly input a channel state feedback input to the pair of neural networks and identify a channel state feedback output from the pair of neural networks. Then, the neural network manager (730) may use a loss function to determine the difference between the channel state feedback input and the channel state feedback output for each iteration, wherein the difference includes a loss, and the neural network manager (730) may adjust the coefficients of the pair of neural networks for each iteration to minimize the difference between the channel state feedback input and the channel state feedback output based on the determination.

[0085] The encoder (735) may encode channel state feedback using the first neural network in the encoder based on training. In some examples, the CSI reporter (725) may report the encoded channel state feedback. In some examples, the neural network manager (730) may transmit coefficients of the second neural network to the base station for decoding the channel state feedback based on training. In some examples, the neural network manager (730) may receive a notification from the base station to train a set of neural network pairs based on a set of accuracy levels, and the set of neural network pairs includes neural network pairs. In some examples, the neural network manager (730) may train each of the set of neural network pairs based on individual accuracy levels of the set of accuracy levels. In some examples, the neural network manager (730) may receive a notification to use a neural network pair among the set of neural network pairs to report channel state feedback. In some examples, the neural network manager (730) may autonomously select a pair of neural networks from a set of pairs of neural networks to report channel state feedback. In some examples, the neural network manager (730) may receive an indication of a subset of a set of pairs of neural networks for the UE to train.

[0086] In some examples, the indicated accuracy level is based on one or more of the corresponding subband, spatial layer, or channel tap. In some examples, the downlink manager (715) may receive data from the base station on the subband or spatial layer or according to the channel tap based on reporting the channel status feedback corresponding to the indicated accuracy level. In some examples, the indicated accuracy level is based on the number of downlink transmissions containing the same data that the UE failed to decode. In some examples, the downlink manager (715) may receive retransmissions of the same data that the UE failed to decode based on reporting the channel status feedback corresponding to the indicated accuracy level.

[0087] In some examples, the CSI manager (720) may identify the number of bits to report channel status feedback based on the accuracy level, where the number of bits is directly related to the accuracy level. In some examples, the CSI reporter (725) may report channel status feedback corresponding to the indicated accuracy level with the identified number of bits. In some examples, the CSI manager (720) may receive an indication of the number of bits to report channel status feedback based on the accuracy level. In some examples, the CSI accuracy manager (710) may receive an indication of the accuracy level in RRC signaling or in MAC-CE.

[0088] FIG. 8 illustrates a diagram of a system (800) comprising a device (805) that supports configurable metrics for channel state compression and feedback according to embodiments of the present disclosure. The device (805) may be an example of a device (505), a device (605), or a UE (115) as described herein, or may include components thereof. The device (805) may include components for bidirectional voice and data communication, including a communication manager (810), an I / O controller (815), a transceiver (820), an antenna (825), a memory (830), and a processor (840), for transmitting and receiving communications. These components may communicate electronically through one or more buses (e.g., bus (845)).

[0089] The communication manager (810) receives from the base station an indication of an accuracy level for reporting channel status feedback to the base station, receives downlink data or reference signals from the base station, and may report channel status feedback corresponding to the indicated accuracy level to the base station based on the downlink data or reference signals.

[0090] The I / O controller (815) may manage input and output signals for the device (805). The I / O controller (815) may also manage peripherals that are not integrated into the device (805). In some cases, the I / O controller (815) may represent a physical connection or port to an external peripheral. In some cases, the I / O controller (815) may utilize an operating system such as iOS®, ANDROID®, MS-DOS®, MS-WINDOWS®, OS / 2®, UNIX®, LINUX®, or other known operating systems. In other cases, the I / O controller (815) may represent or interact with a modem, keyboard, mouse, touchscreen, or similar device. In some cases, the I / O controller (815) may be implemented as part of a processor. In some cases, the user may interact with the device (805) through the I / O controller (815) or through hardware components controlled by the I / O controller (815).

[0091] The transceiver (820) may communicate bidirectionally through one or more antennas, wired or wireless links, as described above. For example, the transceiver (820) may represent a wireless transceiver and may communicate bidirectionally with another wireless transceiver. The transceiver (820) may also include a modem for modulating packets, providing the modulated packets to the antennas for transmission, and demodulating packets received from the antennas.

[0092] In some cases, the wireless device may include a single antenna (825). However, in some cases, the device may have more than one antenna (825) capable of simultaneously transmitting or receiving multiple wireless transmissions.

[0093] Memory (830) may include RAM and ROM. Memory (830) may store computer-readable, computer-executable code (835) containing instructions, which, when executed, cause the processor to perform the various functions described herein. In some cases, memory (830) may include a BIOS that, among others, may control basic hardware or software operations, such as interactions with peripheral components or devices.

[0094] The processor (840) may include intelligent hardware devices (e.g., general-purpose processor, DSP, CPU, microcontroller, ASIC, FPGA, programmable logic device, discrete gate or transistor logic component, discrete hardware component, or any combination thereof). In some cases, the processor (840) may be configured to operate a memory array using a memory controller. In other cases, the memory controller may be integrated into the processor (840). The processor (840) may be configured to execute computer-readable instructions stored in memory (e.g., memory (830)) to enable the device (805) to perform various functions (e.g., functions or tasks supporting configurable metrics for channel state compression and feedback).

[0095] Code (835) may include instructions for implementing aspects of the present disclosure, including instructions for supporting wireless communication. Code (835) may be stored in a non-transient computer-readable medium, such as system memory or other types of memory. In some cases, Code (835) may not be directly executable by the processor (840), but may enable the computer to perform the functions described in this specification (e.g., when compiled and executed).

[0096] FIG. 9 illustrates a block diagram (900) of a device (905) that supports configurable metrics for channel state compression and feedback according to embodiments of the present disclosure. The device (905) may be an example of embodiments of a base station (105) as described herein. The device (905) may include a receiver (910), a communication manager (915), and a transmitter (920). The device (905) may also include a processor. Each of these components may communicate with one another (e.g., via one or more buses).

[0097] The receiver (910) may receive information such as packets associated with various information channels, user data, or control information (e.g., information related to control channels, data channels, and configurable metrics for channel state compression and feedback). The information may be transmitted to other components of the device (905). The receiver (910) may be an example of an embodiment of the transceiver (1220) described with reference to FIG. 12. The receiver (910) may utilize a single antenna or a set of antennas.

[0098] The communication manager (915) transmits an indication of an accuracy level to the UE for reporting channel status feedback to the base station, transmits downlink data or reference signals to the UE, and may receive channel status feedback from the UE corresponding to the indicated accuracy level based on transmitting downlink data or reference signals to the UE. The communication manager (915) may be an example of an embodiment of the communication manager (1210) described herein.

[0099] The communication manager (915) or its sub-components may be implemented in hardware, code executed by a processor (e.g., software or firmware), or any combination thereof. When implemented in code executed by a processor, the functions of the communication manager (915) or its sub-components may be implemented by a general-purpose processor, a DSP, an application-specific integrated circuit (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 in this disclosure.

[0100] The communication manager (915) or its sub-components may be physically located in various positions, including distributed so that parts of the functions are implemented at different physical locations by one or more physical components. In some examples, the communication manager (915) or its sub-components may be separate and distinct components according to various aspects of the present disclosure. In some examples, the communication manager (915) or its sub-components may be combined with one or more other hardware components, including but not limited to input / output (I / O) components, transceivers, network servers, other computing devices, one or more other components described in the present disclosure, or combinations thereof according to various aspects of the present disclosure.

[0101] The transmitter (920) may transmit signals generated by other components of the device (905). In some examples, the transmitter (920) may be juxtaposed with the receiver (910) in the transceiver module. For example, the transmitter (920) may be an example of an embodiment of the transceiver (1220) described with reference to FIG. 12. The transmitter (920) may utilize a single antenna or a set of antennas.

[0102] FIG. 10 illustrates a block diagram (1000) of a device (1005) that supports configurable metrics for channel state compression and feedback according to embodiments of the present disclosure. The device (1005) may be an example of embodiments of the device (905) or base station (105) as described herein. The device (1005) may include a receiver (1010), a communication manager (1015), and a transmitter (1035). The device (1005) may also include a processor. Each of these components may communicate with one another (e.g., via one or more buses).

[0103] The receiver (1010) may receive information such as packets associated with various information channels, user data, or control information (e.g., information related to control channels, data channels, and configurable metrics for channel state compression and feedback). The information may be transmitted to other components of the device (1005). The receiver (1010) may be an example of an embodiment of the transceiver (1220) described with reference to FIG. 12. The receiver (1010) may utilize a single antenna or a set of antennas.

[0104] The communication manager (1015) may be an example of an embodiment of the communication manager (915) as described herein. The communication manager (1015) may include a CSI accuracy manager (1020), a downlink manager (1025), and a CSI manager (1030). The communication manager (1015) may be an example of an embodiment of the communication manager (1210) described herein.

[0105] The CSI accuracy manager (1020) may transmit an indication of the accuracy level to the UE for reporting channel status feedback to the base station. The downlink manager (1025) may transmit downlink data or reference signals to the UE. The CSI manager (1030) may receive channel status feedback from the UE corresponding to the indicated accuracy level based on transmitting downlink data or reference signals to the UE.

[0106] The transmitter (1035) may transmit signals generated by other components of the device (1005). In some examples, the transmitter (1035) may be juxtaposed with the receiver (1010) in the transceiver module. For example, the transmitter (1035) may be an example of an embodiment of the transceiver (1220) described with reference to FIG. 12. The transmitter (1035) may utilize a single antenna or a set of antennas.

[0107] FIG. 11 illustrates a block diagram (1100) of a communication manager (1105) that supports configurable metrics for channel state compression and feedback according to embodiments of the present disclosure. The communication manager (1105) may be an example of embodiments of the communication manager (915), the communication manager (1015), or the communication manager (1210) described herein. The communication manager (1105) may include a CSI accuracy manager (1110), a downlink manager (1115), a CSI manager (1120), a neural network manager (1125), a decoder (1130), and a loss metric manager (1135). Each of these modules may communicate directly or indirectly with one another (e.g., through one or more buses).

[0108] The CSI accuracy manager (1110) may transmit an indication of the accuracy level to the UE for reporting channel state feedback to the base station. The downlink manager (1115) may transmit downlink data or reference signals to the UE. The CSI manager (1120) may receive channel state feedback from the UE corresponding to the indicated accuracy level based on transmitting downlink data or reference signals to the UE. The loss metric manager (1135) may transmit an indication of the loss metric or loss function to be used by the UE to train a pair of neural networks for reporting channel state feedback.

[0109] The neural network manager (1125) may receive coefficients of a neural network in a decoder from the UE to decode channel state feedback from the UE. The decoder (1130) may decode channel state feedback from the UE using a neural network in a decoder. In some examples, the neural network manager (1125) may transmit an indication for the UE to train a set of neural network pairs based on a set of accuracy levels, and each neural network pair includes a first neural network in an encoder to encode channel state feedback and a second neural network in a decoder to decode channel state feedback. In some examples, the neural network manager (1125) may transmit an indication for the UE to use a neural network pair among the set of neural network pairs to report channel state feedback. In some examples, the neural network manager (1125) may transmit a subset of the set of neural network pairs for the UE to train.

[0110] In some examples, the CSI accuracy manager (1110) may transmit indications of different accuracy levels for reporting channel status feedback for different subbands, spatial layers, channel taps, or in response to a failure to decode a different number of downlink transmissions containing the same data. In some examples, the indicated accuracy level includes a first accuracy level for reporting channel status feedback to be used to schedule a first downlink data transmission, and the CSI accuracy manager (1110) may transmit an indication of a second accuracy level for reporting channel status feedback to be used to schedule a second downlink transmission, and the first accuracy level is different from the second accuracy level. In some examples, the CSI manager (1120) may transmit an indication of the number of bits the UE will use to report channel status feedback based on the accuracy level, and the number of bits is directly related to the accuracy level. In some examples, the CSI manager (1120) may receive channel status feedback corresponding to the indicated accuracy level as the number of identified bits. In some examples, the CSI accuracy manager (1110) may transmit an indication of the accuracy level in RRC signaling or in MAC-CE.

[0111] FIG. 12 illustrates a diagram of a system (1200) comprising a device (1205) that supports configurable metrics for channel state compression and feedback according to embodiments of the present disclosure. The device (1205) may be an example of a device (905), a device (1005), or a base station (105) as described herein, or may include components thereof. The device (1205) may include components for bidirectional voice and data communication, including a communication manager (1210), a network communication manager (1215), a transceiver (1220), an antenna (1225), a memory (1230), a processor (1240), and an inter-station communication manager (1245), for transmitting and receiving communications. These components may communicate electronically through one or more buses (e.g., a bus (1250)).

[0112] The communication manager (1210) transmits an indication of an accuracy level to the UE for reporting channel status feedback to the base station, transmits downlink data or reference signals to the UE, and may receive channel status feedback corresponding to the indicated accuracy level from the UE based on the transmission of downlink data or reference signals to the UE.

[0113] The network communication manager (1215) may manage communication with the core network (e.g., through one or more wired backhaul links). For example, the network communication manager (1215) may manage the transmission of data communications to client devices such as one or more UEs (115).

[0114] The transceiver (1220) may communicate bidirectionally through one or more antennas, wired or wireless links, as described above. For example, the transceiver (1220) may represent a wireless transceiver and may communicate bidirectionally with another wireless transceiver. The transceiver (1220) may also include a modem for modulating packets, providing the modulated packets to the antennas for transmission, and demodulating packets received from the antennas.

[0115] In some cases, the wireless device may include a single antenna (1225). However, in some cases, the device may have more than one antenna (1225) capable of simultaneously transmitting or receiving multiple wireless transmissions.

[0116] Memory (1230) may include RAM, ROM, or a combination thereof. Memory (1230) may store computer-readable code (1235) containing instructions, which, when executed by a processor (e.g., processor (1240)), enable the device to perform the various functions described herein. In some cases, memory (1230) may include a BIOS, which, among others, may control basic hardware or software operations, such as interactions with peripheral components or devices.

[0117] The processor (1240) may include an intelligent hardware device (e.g., a general-purpose processor, a DSP, a CPU, a microcontroller, an ASIC, an FPGA, a programmable logic device, a discrete gate or transistor logic component, a discrete hardware component, or any combination thereof). In some cases, the processor (1240) may be configured to operate a memory array using a memory controller. In some cases, the memory controller may be integrated into the processor (1240). The processor (1240) may be configured to execute computer-readable instructions stored in memory (e.g., memory (1230)) to enable the device (1205) to perform various functions (e.g., functions or tasks supporting channel state compression and configurable metrics for feedback).

[0118] The inter-station communication manager (1245) may manage communication with other base stations (105) and may include a controller or scheduler to control communication with UEs (115) in cooperation with other base stations (105). For example, the inter-station communication manager (1245) may coordinate the scheduling of transmissions to UEs (115) for various interference mitigation techniques such as beamforming or joint transmission. In some examples, the inter-station communication manager (1245) may provide an X2 interface within LTE / LTE-A wireless communication network technology to provide communication between base stations (105).

[0119] Code (1235) may include instructions for implementing aspects of the present disclosure, including instructions for supporting wireless communication. Code (1235) may be stored in a non-transient computer-readable medium, such as system memory or other types of memory. In some cases, Code (1235) may not be directly executable by the processor (1240), but may enable the computer to perform the functions described herein (e.g., when compiled and executed).

[0120] FIG. 13 illustrates a flowchart illustrating a method (1300) that supports configurable metrics for channel state compression and feedback according to embodiments of the present disclosure. Operations of method (1300) may be implemented by a UE (115) or its components as described herein. For example, operations of method (1300) may be performed by a communication manager as described with reference to FIGS. 5 through 8. In some examples, the UE may execute a set of commands to control functional elements of the UE to perform the functions described below. Additionally or alternatively, the UE may use special-purpose hardware to perform embodiments of the functions described below.

[0121] In 1305, the UE may receive from the base station an indication of the accuracy level for reporting channel state feedback to the base station. The operations of 1305 may be performed according to the methods described herein. In some examples, aspects of the operations of 1305 may be performed by a CSI accuracy manager as described with reference to FIGS. 6 and 7.

[0122] In 1310, the UE may receive downlink data or reference signals from a base station. The operations of 1310 may be performed according to the methods described herein. In some examples, aspects of the operations of 1310 may be performed by a downlink manager as described with reference to FIGS. 6 and 7.

[0123] In 1315, the UE may report channel state feedback to the base station corresponding to an indicated accuracy level based on downlink data or reference signals. The operations of 1315 may be performed according to the methods described herein. In some examples, aspects of the operations of 1315 may be performed by a CSI reporter as described with reference to FIGS. 6 and 7.

[0124] FIG. 14 illustrates a flowchart illustrating a method (1400) that supports configurable metrics for channel state compression and feedback according to embodiments of the present disclosure. Operations of method (1400) may be implemented by a base station (105) or its components as described herein. For example, operations of method (1400) may be performed by a communication manager as described with reference to FIGS. 9 through 12. In some examples, the base station may perform the functions described below by executing a set of commands to control the functional elements of the base station. Additionally or alternatively, the base station may perform embodiments of the functions described below using special-purpose hardware.

[0125] In 1405, the base station may transmit an indication of the accuracy level to the UE for reporting channel state feedback to the base station. The operations of 1405 may be performed according to the methods described herein. In some examples, aspects of the operations of 1405 may be performed by a CSI accuracy manager as described with reference to FIGS. 10 and FIGS. 11.

[0126] In 1410, the base station may transmit downlink data or reference signals to the UE. The operations of 1410 may be performed according to the methods described herein. In some examples, aspects of the operations of 1410 may be performed by a downlink manager as described with reference to FIGS. 10 and FIGS. 11.

[0127] In 1415, the base station may receive channel state feedback from the UE corresponding to an indicated accuracy level based on transmitting downlink data or reference signals to the UE. The operations of 1415 may be performed according to the methods described herein. In some examples, aspects of the operations of 1415 may be performed by a CSI manager as described with reference to FIGS. 10 and FIGS. 11.

[0128] The following provides an overview of examples of the present disclosure:

[0129] Example 1: A method for wireless communication in a UE comprises the steps of: receiving from a base station an indication of an accuracy level for reporting channel state feedback to the base station; receiving downlink data or reference signals from the base station; and reporting channel state feedback to the base station corresponding to the indicated accuracy level based at least partially on the downlink data or reference signals.

[0130] Example 2: In the method of Example 1, the step of receiving an indication of an accuracy level for reporting channel state feedback includes the step of receiving an indication of a loss function corresponding to an accuracy level for training a pair of neural networks, wherein the pair of neural networks includes a first neural network in an encoder for encoding channel state feedback and a second neural network in a decoder for decoding channel state feedback, and the method further includes the step of training a pair of neural networks using a loss function.

[0131] Example 3: In either Example 1 or Example 2, the step of training a neural network pair using a loss function comprises: repeatedly inputting a channel state feedback input to the neural network pair and identifying a channel state feedback output from the neural network pair; determining the difference between the channel state feedback input and the channel state feedback output for each iteration using a loss function, wherein the difference includes a loss; and adjusting the coefficients of the neural network pair for each iteration to minimize the difference between the channel state feedback input and the channel state feedback output based at least partially on the determination.

[0132] Example 4: In any one of the methods of Examples 1 to 3, the step of reporting channel state feedback corresponding to an indicated accuracy level includes: encoding channel state feedback using a first neural network in an encoder based at least partially on training; and reporting the encoded channel state feedback.

[0133] Example 5: Any one of Examples 1 to 4 further includes the step of transmitting coefficients of a second neural network for decoding channel state feedback based at least partially on training to a base station.

[0134] Example 6: Any one of Examples 1 to 5 further comprises the step of receiving a signal from a base station for training a plurality of neural network pairs based at least partially on a plurality of accuracy levels, wherein the plurality of neural network pairs include a neural network pair; and the step of training each of the plurality of neural network pairs based at least partially on an individual accuracy level of the plurality of accuracy levels.

[0135] Example 7: In any one of the methods of Examples 1 to 6, the step of receiving an indication of an accuracy level includes the step of receiving an indication to use a pair of neural networks among a plurality of pairs of neural networks to report channel state feedback.

[0136] Example 8: Any one of Examples 1 to 7 further includes the step of autonomously selecting a pair of neural networks among a plurality of pairs of neural networks to report channel state feedback.

[0137] Example 9: Any one of Examples 1 through 8 further includes the step of receiving a representation of a subset of a plurality of neural network pairs for training the UE.

[0138] Example 10: In any one of the methods of Examples 1 to 9, the indicated accuracy level is at least partially based on one or more of the corresponding subband, spatial layer, or channel tap, and the method further comprises the step of receiving data from a base station on the subband or spatial layer or according to the channel tap, at least partially based on reporting the channel status feedback corresponding to the indicated accuracy level.

[0139] Example 11: In any one of the methods of Examples 1 through 10, the indicated accuracy level is at least partially based on the number of downlink transmissions containing the same data that the UE failed to decode, and the method further includes the step of receiving a retransmission of the same data that the UE failed to decode based at least partially on reporting channel status feedback corresponding to the indicated accuracy level.

[0140] Example 12: Any one of Examples 1 to 11 further comprises the step of identifying a number of bits for reporting channel state feedback based at least partially on an accuracy level, said number of bits being directly related to the accuracy level; and the step of reporting channel state feedback corresponding to an indicated accuracy level using the identified number of bits.

[0141] Example 13: Any one of Examples 1 to 12 further includes the step of receiving an indication of the number of bits for reporting channel status feedback based at least partially on the accuracy level.

[0142] Example 14: In any one of the methods of Examples 1 to 13, the step of receiving an indication of the accuracy level includes receiving an indication of the accuracy level in RRC signaling or in MAC-CE.

[0143] Example 15: A method for wireless communication at a base station comprises the steps of: transmitting an indication of an accuracy level for reporting channel state feedback to a base station to a user device (UE); transmitting downlink data or reference signals to the UE; and receiving channel state feedback corresponding to the indicated accuracy level from the UE based at least partially on transmitting downlink data or reference signals to the UE.

[0144] Example 16: In the method of Example 15, the step of transmitting an indication of an accuracy level for reporting channel state feedback includes the step of transmitting an indication of a loss function to be used by the UE to train a pair of neural networks for reporting channel state feedback.

[0145] Example 17: Any one of Example 15 or Example 16 includes the step of receiving coefficients of a neural network in a decoder from a UE to decode channel state feedback from the UE; and the step of decoding channel state feedback from the UE using a neural network in a decoder.

[0146] Example 18: Any one of Examples 15 to 17 further comprises the step of the UE transmitting a signal for training a plurality of neural network pairs based at least partially on a plurality of accuracy levels, wherein each neural network pair includes a first neural network in an encoder for encoding channel state feedback and a second neural network in a decoder for decoding channel state feedback.

[0147] Example 19: Any one of Examples 15 to 18 further includes the step of transmitting a signal for the UE to use a pair of neural networks among a plurality of neural network pairs to report channel state feedback.

[0148] Example 20: Any one of Examples 15 to 19 further includes the step of transmitting a representation of a subset of a plurality of neural network pairs for training the UE.

[0149] Example 21: In any one of the methods of Examples 15 to 20, the step of transmitting an indication of an accuracy level for reporting channel state feedback includes the step of transmitting indications of different accuracy levels for reporting channel state feedback for different subbands, spatial layers, channel taps, or in response to a failure to decode a different number of downlink transmissions containing the same data.

[0150] Example 22: In any one of the methods of Examples 15 to 21, the indicated accuracy level includes a first accuracy level for reporting channel status feedback to be used for scheduling a first downlink data transmission, and the method further includes the step of transmitting an indication of a second accuracy level for reporting channel status feedback to be used for scheduling a second downlink transmission, wherein the first accuracy level is different from the second accuracy level.

[0151] Example 23: Any one of Examples 15 to 22 further comprises the step of transmitting an indication of the number of bits to be used by a UE to report channel state feedback based at least partially on an accuracy level, said indication of the number of bits being directly related to the accuracy level; and means for receiving channel state feedback corresponding to the indicated accuracy level with the identified number of bits.

[0152] Example 24: In any one of the methods of Examples 15 to 23, the step of transmitting an indication of the accuracy level includes the step of transmitting an indication of the accuracy level in RRC signaling or in MAC-CE.

[0153] Example 25: A device for wireless communication includes at least one means for performing any one of the methods of Example 1 to Example 14.

[0154] Example 26: A device for wireless communication includes a processor and a memory coupled to the processor. The processor and the memory may be configured to enable the device to perform any one of the methods of Examples 1 through 14.

[0155] Example 27: A non-transient computer-readable medium for storing code for wireless communication comprises a processor, memory coupled to the processor, and instructions stored in the memory and executable by the processor to cause a device to perform any one of the methods of Examples 1 through 14.

[0156] Example 28: A device for wireless communication includes at least one means for performing any one of the methods of Examples 15 to 24.

[0157] Example 29: A device for wireless communication includes a processor and a memory coupled to the processor. The processor and the memory may be configured to enable the device to perform any one of the methods of Examples 15 through 24.

[0158] Example 30: A non-transient computer-readable medium for storing code for wireless communication comprises a processor, a memory coupled to the processor, and instructions stored in the memory and executable by the processor to cause a device to perform any one of the methods of Examples 15 to 24.

[0159] It should be noted that the methods described herein describe possible implementations, and that the operations and steps may be rearranged or otherwise modified, and that other implementations are possible. Additionally, aspects from two or more of the methods may be combined.

[0160] Although embodiments of LTE, LTE-A, LTE-A Pro, or NR systems may be described for illustrative purposes and the terms LTE, LTE-A, LTE-A Pro, or NR may be used throughout most of the description, the techniques described herein are applicable beyond LTE, LTE-A, LTE-A Pro, or NR networks. For example, the described techniques may be applicable to various other wireless communication systems, such as Ultra Mobile Broadband (UMB), IEEE (Institute of Electrical and Electronics Engineers) 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE 802.20, Flash-OFDM, as well as other systems and wireless technologies not explicitly mentioned in this specification.

[0161] The information and signals described herein may be represented using any various different technologies and techniques. For example, data, instructions, commands, information, signals, bits, symbols, and chips that may be referenced throughout this description may be represented by voltage, current, electromagnetic waves, magnetic fields or magnetic particles, optical fields or optical particles, or any combination thereof.

[0162] The various exemplary blocks and components described in connection with the disclosure herein may be implemented or performed as general-purpose processors, DSPs, ASICs, CPUs, FPGAs or other programmable logic devices, 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 alternatively, the processor may be any processor, controller, microcontroller, or state machine. The processor may also be implemented as a combination of computing devices (e.g., a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors combined with a DSP core, or any other such configuration).

[0163] The functions described herein may be implemented in hardware, software executed by a processor, firmware, or any combination thereof. If implemented in software executed by a processor, the functions may be stored or transmitted over a computer-readable medium as one or more instructions or code. Other examples and implementations are within the scope of this disclosure and the appended claims. For example, due to the nature of the software, the functions described herein may be implemented using software executed by a processor, hardware, firmware, hardwiring, or any combination thereof. Features implementing the functions may also be physically located at various positions, including distributed so that parts of the functions are implemented at different physical locations.

[0164] Computer-readable media include both communication media and non-transient computer storage media, which include any medium that facilitates the transfer of a computer program from one place to another. A non-transient storage medium may be any available medium that may be accessed by a general-purpose or special-purpose computer. By example, but not by limitation, non-transient computer-readable media may include random access memory (RAM), read-only memory (ROM), electrically erasable programmable ROM (EEPROM), flash memory, compact disc (CD) ROM or other optical disc storage, magnetic disc storage or other magnetic storage devices, or any other non-transient medium that may be used to record or store desired program code means in the form of instructions or data structures and may be accessed by a general-purpose or special-purpose computer or a general-purpose or special-purpose processor. Additionally, any connection is appropriately named as a computer-readable medium. For example, if software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of computer-readable media. As used herein, disks and discs include CDs, laser discs, optical discs, digital multifunction discs (DVDs), floppy discs, and Blu-ray discs, wherein disks typically reproduce data magnetically, while discs reproduce data optically using lasers. The above combinations are also included within the scope of computer-readable media.

[0165] As used herein, including in the claims, the word “or” as used in a list of items (e.g., a list of items beginning with phrases such as “at least one of” or “one or more of”) indicates a comprehensive list such 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, B, and C). Also, as used herein, the phrase “based on” should not be interpreted as a reference to a closed set of conditions. For example, an exemplary step described as “based on condition A” may be based on both condition A and condition B without departing from the scope of this disclosure. That is, as used herein, the phrase “based on” should be interpreted in the same way as the phrase “at least partially based on”.

[0166] In the attached drawings, similar components or features may have the same reference label. Additionally, various components of the same type may be distinguished by placing a dash and a second label following the reference label to distinguish between similar components. If only the first reference label is used in the specification, the description is applicable to any component among similar components having the same first reference label, regardless of the second reference label or other subsequent reference labels.

[0167] The descriptions provided herein in connection with the accompanying drawings describe exemplary configurations and do not represent all examples that may be implemented or are within the scope of the claims. As used herein, the term “example” means “functioning as an example, case, or illustration” and does not mean “preferred over other examples” or “advantageous.” The detailed description includes specific details for the purpose of providing an understanding of the described techniques. However, these techniques may be practiced without these specific details. In some cases, known structures and devices are illustrated in block diagram form to avoid obscuring the concepts of the described examples.

[0168] The description in this specification is provided to enable those skilled in the art to manufacture or use the present disclosure. Various modifications to the present disclosure will be obvious to those skilled in the art, and general principles defined in this specification may be applied to other variations without departing from the scope of the present disclosure. Accordingly, the present disclosure is not limited to the examples and designs described in this specification, but should be given the broadest scope consistent with the principles and novel features disclosed in this specification.

Claims

Claim 1 A method for wireless communication in a user device (UE), comprising: receiving from a base station an indication of an accuracy level for reporting channel state feedback to the base station; receiving downlink data or reference signals from the base station; and reporting the channel state feedback corresponding to the accuracy level to the base station at least partially based on the downlink data or the reference signals, wherein the accuracy level is at least partially based on the number of downlink transmissions containing the same data that the UE failed to decode, and the method further comprises receiving a retransmission of the same data that the UE failed to decode at least partially based on reporting the channel state feedback corresponding to the accuracy level. Claim 2 A method for wireless communication in a user device (UE), wherein the step of receiving an indication of an accuracy level for reporting the channel state feedback comprises receiving an indication of a loss function corresponding to the accuracy level for training a pair of neural networks, wherein the pair of neural networks comprises a first neural network in an encoder for encoding the channel state feedback and a second neural network in a decoder for decoding the channel state feedback, and the method further comprises the step of training the pair of neural networks using the loss function. Claim 3 A method for wireless communication in a user device (UE), wherein the step of training the neural network pair using the loss function comprises: repeatedly inputting a channel state feedback input to the neural network pair and identifying a channel state feedback output from the neural network pair; determining the difference between the channel state feedback input and the channel state feedback output for each iteration using the loss function, wherein the difference includes a loss; and adjusting the coefficients of the neural network pair for each iteration to minimize the difference between the channel state feedback input and the channel state feedback output based at least partially on the determination. Claim 4 A method for wireless communication in a user device (UE), wherein the step of reporting the channel state feedback corresponding to the accuracy level comprises: encoding the channel state feedback using the first neural network in the encoder based at least partially on the training; and reporting the encoded channel state feedback. Claim 5 A method for wireless communication in a user device (UE), further comprising the step of transmitting coefficients of the second neural network for decoding the channel state feedback based at least partially on the training to the base station. Claim 6 A method for wireless communication in a user device (UE), further comprising: receiving a signal from the base station for training a plurality of neural network pairs based at least partially on a plurality of accuracy levels, wherein the plurality of neural network pairs include said neural network pairs; and training each of the plurality of neural network pairs based at least partially on an individual accuracy level of said accuracy levels. Claim 7 A method for wireless communication in a user device (UE), wherein, in claim 1, the accuracy level is at least partially based on one or more of the subband, spatial layer, or channel tap corresponding to the channel state feedback, and the method further comprises the step of receiving data from the base station on the subband or spatial layer or according to the channel tap based at least partially on reporting the channel state feedback corresponding to the accuracy level. Claim 8 delete Claim 9 A method for wireless communication in a user device (UE), further comprising: a step of identifying a number of bits for reporting channel state feedback based at least partially on the accuracy level, wherein the number of bits is directly related to the accuracy level; and a step of reporting channel state feedback corresponding to the accuracy level with the identified number of bits. Claim 10 A method for wireless communication in a user device (UE), wherein the step of receiving an indication of the accuracy level comprises receiving an indication of the accuracy level in a wireless resource control (RRC) signaling or in a media access control (MAC) control element (MAC-CE). Claim 11 A method for wireless communication at a base station, comprising: transmitting an indication of an accuracy level for reporting channel state feedback to a user equipment (UE); transmitting downlink data or reference signals to the UE; and receiving channel state feedback corresponding to the accuracy level from the UE based at least partially on transmitting the downlink data or reference signals to the UE, wherein the step of transmitting an indication of an accuracy level for reporting channel state feedback comprises transmitting indications of different accuracy levels for reporting channel state feedback for different subbands, spatial layers, channel taps, or in response to a failure to decode a different number of downlink transmissions containing the same data. Claim 12 A method for wireless communication at a base station, wherein the step of transmitting an indication of an accuracy level for reporting the channel state feedback comprises the step of transmitting an indication of a loss function to be used by the UE to train a pair of neural networks for reporting the channel state feedback. Claim 13 A method for wireless communication at a base station, further comprising: receiving coefficients for a neural network in a decoder from the UE to decode the channel state feedback from the UE; and decoding the channel state feedback from the UE using the neural network in the decoder. Claim 14 A device for wireless communication in a user device (UE), comprising: means for receiving from a base station an indication of an accuracy level for reporting channel state feedback to the base station; means for receiving downlink data or reference signals from the base station; and means for reporting the channel state feedback corresponding to the accuracy level to the base station at least partially based on the downlink data or the reference signals, wherein the accuracy level is at least partially based on the number of downlink transmissions containing the same data that the UE failed to decode, and the device further comprises means for receiving a retransmission of the same data that the UE failed to decode at least partially based on reporting the channel state feedback corresponding to the accuracy level. Claim 15 A device for wireless communication in a user device (UE), comprising: a processor; a memory coupled to the processor; and instructions stored in the memory, wherein the instructions are executable by the processor to cause the device to receive, from a base station, an indication of an accuracy level for reporting channel state feedback to the base station; to receive downlink data or reference signals from the base station; and to report the channel state feedback corresponding to the accuracy level to the base station at least partially based on the downlink data or the reference signals, wherein the accuracy level is at least partially based on the number of downlink transmissions containing the same data that the UE failed to decode, and the instructions are further executable by the processor to cause the device to receive a retransmission of the same data that the UE failed to decode at least partially based on the reporting of the channel state feedback corresponding to the accuracy level. Claim 16 A device for wireless communication in a user device (UE), wherein, in claim 15, commands for receiving an indication of an accuracy level for reporting the channel state feedback are executable by the processor to cause the device to receive an indication of a loss function corresponding to the accuracy level for training a pair of neural networks, the pair of neural networks comprises a first neural network in an encoder for encoding the channel state feedback and a second neural network in a decoder for decoding the channel state feedback, and the commands are further executable by the processor to cause the device to train the pair of neural networks using the loss function. Claim 17 A device for wireless communication in a user device (UE), wherein, in the 16th, instructions for training the neural network pair using the loss function cause the device to repeatedly input a channel state feedback input to the neural network pair and identify a channel state feedback output from the neural network pair; cause the device to determine, using the loss function, the difference between the channel state feedback input and the channel state feedback output for each iteration, wherein the difference includes a loss; and cause the coefficients of the neural network pair for each iteration to be adjusted to minimize the difference between the channel state feedback input and the channel state feedback output, at least in part, based on the determination of the difference. Claim 18 A device for wireless communication in a user device (UE), wherein, in the 16th, commands for reporting the channel state feedback corresponding to the accuracy level are executable by the processor to cause the device to encode the channel state feedback using the first neural network in the encoder based at least partially on the training of the neural network pair; and to report the encoded channel state feedback. Claim 19 In claim 16, the above commands are further executable by the processor to cause the device to transmit coefficients of the second neural network to the base station for decoding the channel state feedback based at least partially on the training of the neural network pair, for a device for wireless communication in a user device (UE). Claim 20 In claim 16, the commands further enable the device to receive a signal from the base station for training a plurality of neural network pairs based at least partially on a plurality of accuracy levels, wherein the plurality of neural network pairs include the neural network pairs; and enable the device to train each of the plurality of neural network pairs based at least partially on an individual accuracy level of the plurality of accuracy levels, the device for wireless communication in a user device (UE). Claim 21 A device for wireless communication in a user device (UE), wherein, in claim 20, commands for receiving an indication of the accuracy level are executable by the processor to cause the device to receive an indication to use the neural network pair among the plurality of neural network pairs to report the channel state feedback. Claim 22 In claim 20, the above commands are further executable by the processor to enable the device to autonomously select the neural network pair among the plurality of neural network pairs to report the channel state feedback, a device for wireless communication in a user device (UE). Claim 23 In claim 20, the above commands are additionally executable by the processor to cause the device to receive a representation of a subset of the plurality of neural network pairs for the UE to train, a device for wireless communication in a user device (UE). Claim 24 A device for wireless communication in a user device (UE), wherein, in claim 15, the accuracy level is at least partially based on one or more of the corresponding subband, spatial layer, or channel tap, and the commands are further executable by the processor to cause the device to receive data from the base station on the subband or spatial layer or according to the channel tap, at least partially based on the reporting of the channel state feedback corresponding to the accuracy level. Claim 25 delete Claim 26 In claim 15, the commands further enable the device to identify a number of bits for reporting the channel state feedback based at least partially on the accuracy level, wherein the number of bits is directly related to the accuracy level; and the device for wireless communication in a user device (UE), executable by the processor to report the channel state feedback corresponding to the accuracy level with the identified number of bits. Claim 27 In claim 26, the above commands are further executable by the processor to cause the device to receive an indication of the number of bits for reporting the channel state feedback based at least partially on the accuracy level, for a device for wireless communication in a user device (UE). Claim 28 A device for wireless communication in a user device (UE), wherein, in claim 15, commands for receiving an indication of the accuracy level are executable by the processor to cause the device to receive an indication of the accuracy level in a wireless resource control (RRC) signaling or in a media access control (MAC) control element (MAC-CE). Claim 29 A device for wireless communication at a base station, comprising: a processor; a memory coupled to the processor; and instructions stored in the memory, wherein the instructions are executable by the processor to cause the device to transmit an indication of an accuracy level for reporting channel state feedback to the base station to a user equipment (UE); to transmit downlink data or reference signals to the UE; and to receive channel state feedback corresponding to the accuracy level from the UE based at least partially on the transmission of the downlink data or reference signals to the UE, and the instructions for transmitting an indication of an accuracy level for reporting channel state feedback are executable by the processor to cause the device to transmit indications of different accuracy levels for reporting channel state feedback in response to failure to decode different subbands, spatial layers, channel taps, or different numbers of downlink transmissions containing the same data. Claim 30 A device for wireless communication at a base station, wherein, in claim 29, commands for transmitting an indication of an accuracy level for reporting the channel state feedback are executable by the processor to cause the device to transmit an indication of a loss function to be used by the UE to train a pair of neural networks for reporting the channel state feedback. Claim 31 In claim 29, the above commands further enable the device to receive coefficients for a neural network in a decoder from the UE to decode the channel state feedback from the UE; and enable the device to decode the channel state feedback from the UE using the neural network in the decoder, the device for wireless communication at a base station, executable by the processor. Claim 32 In claim 29, the commands are further executable by the processor to cause the device to transmit a signal for training a plurality of neural network pairs based at least partially on a plurality of accuracy levels, and each neural network pair comprises a first neural network in an encoder for encoding the channel state feedback and a second neural network in a decoder for decoding the channel state feedback, for a device for wireless communication at a base station. Claim 33 delete Claim 34 In claim 29, the accuracy level includes a first accuracy level for reporting channel state feedback to be used for scheduling a first downlink data transmission, and the commands are further executable by the processor to cause the device to transmit an indication of a second accuracy level for reporting channel state feedback to be used for scheduling a second downlink transmission, wherein the first accuracy level is different from the second accuracy level, a device for wireless communication at a base station. Claim 35 In claim 29, the above commands further cause the device to transmit an indication of the number of bits to be used by the UE to report the channel state feedback based at least partially on the accuracy level, wherein the number of bits is directly related to the accuracy level; and the device for wireless communication at a base station, executable by the processor to receive the channel state feedback corresponding to the accuracy level with the number of bits.

Images