communications using a neural network based at least in part on an indicated input size and / or input structure

Abstract

Various aspects of the disclosure generally relate to wireless communication. In some aspects, a user equipment (UE) can receive an indication of one or more of an input size or an input structure for a transmitting side neural network, where the input structure indicates one or more of a non-zero value or a location of the non-zero value within an input having the input size to the transmitting side neural network. The UE can communicate with a base station based at least in part on the indication of one or more of the input size or the input structure of the transmitting side neural network. Numerous other aspects are provided.

Description

TECHNICAL FIELD

[0001] Aspects of the disclosure relate generally to wireless communication, and more particularly to techniques and apparatuses for communication using a neural network based at least in part on an indicated input size and / or input structure. BACKGROUND

[0002] Wireless communication systems are widely deployed to provide various telecommunication services such as telephony, video, data, messaging, and broadcasts. Typical wireless communication systems can employ multiple-access technologies capable of supporting communication with multiple users by sharing available system resources (e.g., bandwidth, transmit power, etc.). Examples of such multiple-access technologies include code division multiple access (CDMA) systems, time division multiple access (TDMA) systems, frequency division multiple access (FDMA) systems, orthogonal frequency division multiple access (OFDMA) systems, single-carrier frequency division multiple access (SC-FDMA) systems, time division synchronous code division multiple access (TD-SCDMA) systems, and long term evolution (LTE). LTE / LTE-Advanced is a set of enhancements to the Universal Mobile Telecommunications System (UMTS) mobile standard promulgated by the Third Generation Partnership Project (3 GPP).

[0003] A wireless network can include a number of base stations (BSs) that can support communication for a number of user equipment (UEs). A user equipment (UE) can communicate with a base station (BS) via the downlink and uplink. The downlink (or forward link) refers to the communication from the BS to the UE, and the uplink (or reverse link) refers to the communication from the UE to the BS. As will be described in more detail herein, a BS can be referred to as a Node B, a gNB, an access point (AP), a radio head, a transmit receive point (TRP), a new radio (NR) BS, a 5G Node B, and / or the like.

[0004] The above multiple access technologies have been adopted in various telecommunication standards to provide common protocols facilitating communication between wireless devices from different technologies. New Radio (NR), which can also be referred to as 5G, is a set of enhancements to the LTE mobile standard promulgated by the Third Generation Partnership Project (3GPP). NR is designed to better support mobile broadband Internet access by improving spectral efficiency, lowering costs, improving services, making use of new spectrum, and better integrating with other open standards using orthogonal frequency division multiplexing (OFDM) with a cyclic prefix (CP) (CP-OFDM) on the downlink (DL), using CP- OFDM and / or SC-FDM (e.g., also known as discrete Fourier transform spread orthogonal frequency division multiplexing (DFT-s-OFDM)) for the uplink (UL), as well as promoting SUMMARY

[0005] In some aspects, a method of wireless communication performed by a user equipment (UE) includes receiving an indication of one or more of an input size or an input structure for a transmit-side neural network, wherein the input structure indicates one or more of non-zero values or locations of non-zero values within an input having the input size to the transmit-side neural network; and communicating with a base station based at least in part on the indication of the one or more of the input size or the input structure of the transmit-side neural network.

[0006] In some aspects, a UE for wireless communication includes a memory and one or more processors operatively coupled to the memory, the memory and the one or more processors configured to: receive an indication of one or more of an input size or an input structure for a transmit-side neural network, wherein the input structure indicates one or more of non-zero values or locations of non-zero values within an input having the input size to the transmit-side neural network; and communicate with a base station based at least in part on the indication of the one or more of the input size or the input structure of the transmit-side neural network.

[0007] In some aspects, a method of wireless communication performed by a base station includes transmitting an indication of one or more of an input size or an input structure for a transmit-side neural network, wherein the input structure indicates one or more of non-zero values or locations of non-zero values within an input having the input size to the transmit-side neural network; and communicating with a UE based at least in part on the indication of the one or more of the input size or the input structure of the transmit-side neural network.

[0008] In some aspects, a base station for wireless communication includes: a memory; and one or more processors operatively coupled to the memory, the memory and the one or more processors being configured to: transmit indications of one or more of an input size or input structure for a transmitting-side neural network, wherein the input structure indicates one or more of the positions of non-zero values ​​or non-zero values ​​within an input having an input size leading to the transmitting-side neural network; and communicate with a UE at least in part based on the indications of one or more of the input size or input structure of the transmitting-side neural network.

[0009] In some aspects, a non-transitory computer-readable medium storing a set of instructions for wireless communication includes one or more instructions that, when executed by one or more processors of a UE, cause the UE to: receive an indication of one or more of the input size or input structure for a transmitting-side neural network, wherein the input structure indicates one or more of the positions of non-zero values ​​or non-zero values ​​within an input having the input size leading to the transmitting-side neural network; and communicate with a base station at least in part based on the indication of one or more of the input size or input structure of the transmitting-side neural network.

[0010] In some aspects, a non-transitory computer-readable medium storing a set of instructions for wireless communication includes one or more instructions that, when executed by one or more processors of a base station, cause the base station to: send an indication of one or more of the input size or input structure for a transmitting-side neural network, wherein the input structure indicates one or more of the positions of non-zero values ​​or non-zero values ​​within the input having the input size to the transmitting-side neural network; and communicate with a UE at least in part based on the indication of one or more of the input size or input structure of the transmitting-side neural network.

[0011] In some aspects, an apparatus for wireless communication includes: a unit for receiving an indication of one or more of the input size or input structure for a transmitting-side neural network, wherein the input structure indicates one or more of the positions of non-zero values ​​or non-zero values ​​within an input having an input size leading to the transmitting-side neural network; and a unit for communicating with a base station based at least in part on the indication of one or more of the input size or input structure of the transmitting-side neural network.

[0012] In some aspects, an apparatus for wireless communication includes means for transmitting an indication of one or more of an input size or an input structure for a transmitting side neural network, wherein the input structure indicates one or more of a non-zero value or a location of a non-zero value within an input of the input size to the transmitting side neural network; and means for communicating with a UE based at least in part on the indication of one or more of the input size or the input structure of the transmitting side neural network.

[0013] Aspects generally include methods, apparatus, systems, computer program products, non-transitory computer-readable media, user equipment, base stations, wireless communication devices, and / or processing systems, as substantially described herein with reference to and as illustrated by the accompanying drawings and specification.

[0014] The foregoing has outlined rather broadly the features and technical advantages of examples according to the disclosure in order that the detailed description that follows can be better understood. Additional features and advantages will be described hereinafter. The disclosed concepts and specific examples can be readily utilized as bases for modifying or designing other structures for carrying out the same purposes of the present disclosure. Such equivalent constructions are not to be considered as departing from the scope of the appended claims. The illustrative examples described herein will be better understood with reference to the drawings outlined herein. The drawings are provided as is and are not limiting of the scope of the disclosure. The drawings provided are for purposes of illustration only and merely show typical or example aspects in accordance with the disclosure. One skilled in the art will readily recognize from the following description that alternative examples of the aspects disclosed can be used as well. The detailed description especially, along with the drawings also provided below, will best explain the aspects of the disclosure and will best show how it can be carried out in practice. BRIEF DESCRIPTION OF DRAWINGS

[0015] In order that the above-recited features of the present disclosure can be understood in detail, a more particular description of the application, briefly summarized above, can be had by reference to various aspects, some of which are illustrated in the appended drawings. It is to be noted, however, that the appended drawings illustrate only typical aspects of this disclosure and are therefore not to be considered limiting of its scope, for the description can admit to other equally effective aspects. Like reference numerals in the various drawings can designate the same or similar elements.

[0016] Figure 1 FIG. 1 is a diagram illustrating an example of a wireless network according to various aspects of the disclosure.

[0017] Figure 2 FIG. 1 is a diagram illustrating an example of a wireless network according to various aspects of the disclosure.

[0018] Figure 3 FIG. 1 is a diagram illustrating an example of a wireless network according to various aspects of the disclosure.

[0019] Figures 4-6FIG. 1 is a diagram illustrating an example of a communication associated with using a neural network based at least in part on an indicated input size and / or input structure, in accordance with various aspects of the present disclosure.

[0020] Figure 7 and Figure 8 FIG. 2 is a diagram illustrating an example process of a communication associated with using a neural network based at least in part on an indicated input size and / or input structure, in accordance with various aspects of the present disclosure.

[0021] Figure 9 and Figure 10 FIG. 3 is a block diagram of an example apparatus for wireless communication, in accordance with various aspects of the present disclosure. DETAILED DESCRIPTION

[0022] Various aspects of the disclosure are now described with reference to the drawings. However, the disclosure can be embodied in many different forms and should not be construed as limited to the particular aspects set forth throughout this disclosure. Rather, these aspects are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art. Based on the teachings herein one skilled in the art should appreciate that the scope of the disclosure is intended to cover any aspect of the disclosure disclosed herein, whether implemented independently of, or combined with, any other aspect of the disclosure. For example, an apparatus can be implemented or a method can be practiced using any number of the aspects set forth herein. In addition, the scope of the disclosure is intended to cover such an apparatus or method which is practiced using, in addition to or in place of the aspects set forth herein, other structures, functionalities, or structures and

[0023] Several aspects of telecommunication systems will now be presented with reference to various apparatus and methods. These apparatus and methods will be described in the following detailed description and illustrated in the accompanying drawings by various blocks, modules, components, circuits, steps, processes, algorithms, etc. (collectively referred to as "elements"). These elements can be implemented using hardware, software, or combinations thereof. Whether such elements are implemented as hardware or software depends on the particular application and design constraints imposed on the overall system.

[0024] It should be noted that while aspects can be described herein using terminology commonly associated with a 5G or NR radio access technology (RAT), aspects of the present disclosure can be applied to other RATs, such as a 3G RAT, a 4G RAT, and / or a RAT subsequent to 5G (e.g., 6G).

[0025] Figure 1 FIG. 1 is a diagram illustrating an example of a wireless network 100, in accordance with various aspects of the present disclosure. The wireless network 100 can be or include elements of a 5G (NR) network and / or an LTE network, among other examples. The wireless network 100 can include a number of base stations 110 (shown as BS 110a, BS 110b, BS 110c, and BS 1 lOd) and other network entities. A base station (BS) is an entity that communicates with user equipment (UEs) and can also be referred to as an NR BS, a Node B, a gNB, a 5G node B (NB), an access point, a transmit receive point (TRP), or the like. Each BS can provide communication coverage for a particular geographic area. In 3GPP, the term “cell” can refer to a coverage area of a BS and / or a BS subsystem serving the coverage area, depending on the context in which the term is used.

[0026] BSs can provide communication coverage for a macro cell, a pico cell, a femto cell, and / or another type of cell. A macro cell can cover a relatively large geographic area (e.g., several kilometers in radius) and can allow unrestricted access by UEs with service subscriptions appropriate for the Figure 1 In an example shown in FIG. 1, the BS 110a can be a macro BS for a macro cell 102a, the BS 110b can be a pico BS for a pico cell 102b, and the BS 110c can be a femto BS for a femto cell 102c. A BS can be referred to as a gNB, Node B, or some other similar terminology in other examples. A BS can support one or multiple (e.g., three) cells. The term “BS,” “gNB,” “cell,” “NR BS,” “TRP,” “AP,” “Node B,” “5G NB,” and “cell” can be used interchangeably herein.

[0027] In some aspects, a cell can not necessarily be stationary, and the geographic area of the cell can move based on the location of a mobile BS. In some aspects, BSs can be interconnected to one another and / or to one or more other BSs or network nodes in the wireless network 100 through various types of backhaul interfaces such as a direct physical connection or a virtual network, using any appropriate transfer network.

[0028] Wireless network 100 can also include relay stations. A relay station is an entity that can receive a transmission of data from an upstream station (e.g., a BS or a UE) and send a transmission of the data to a downstream station (e.g., a UE or a BS). A relay station can also be a UE that can relay transmissions for other UEs. Figure 1 In the example shown in Figure 1, a relay BS 1 lOd can communicate with macro BS 110a and a UE 120d in order to facilitate communications between BS 110a and UE 120d. A relay BS can also be referred to as a relay station, a relay base station, a relay, or the like.

[0029] Wireless network 100 can be a heterogeneous network that includes BSs of different types, such as macro BSs, pico BSs, femto BSs, relay BSs, or the like. These different types of BSs can have different transmit power levels, different coverage areas, and different impacts on interference. For example, macro BSs can have a high transmit power level (e.g., 5 to 40 Watts) whereas pico BSs, femto BSs, and relay BSs can have relatively lower transmit power levels (e.g., 0.1 to 2 Watts).

[0030] A network controller 130 can couple to a set of BSs and can provide coordination and control for these BSs. Network controller 130 can be

[0031] The UEs 120 (e.g., 120a, 120b, 120c) can be dispersed throughout the wireless network 100, and each UE can be stationary or mobile. A UE can also be referred to as an access terminal, a terminal, a mobile station, a subscriber unit, a station, etc. A UE can be a cellular phone (e.g., a smart phone), a personal digital assistant (PDA), a wireless modem, a wireless communication device, a handheld device, a laptop computer, a cordless phone, a wireless local loop (WLL) station, a tablet, a camera, a gaming device, a netbook, a smartbook, an ultrabook, a medical device or equipment, a biometric sensor / device, a wearable device such as a smart watch, smart clothing, smart glasses, a smart wrist band, a smart jewel (e.g., a smart ring, a smart bracelet), an entertainment device (e.g., a music or video device, or a satellite radio), a vehicular component or sensor, a smart meter / sensor, industrial manufacturing equipment, a global positioning system device, or any other suitable device that is configured to communicate via a wireless or wired medium.

[0032] Some UEs can be considered machine-type communication (MTC) or evolved or enhanced machine-type communication (eMTC) UEs. MTC and eMTC UEs include, e.g., robots, drones, remote devices, sensors, meters, monitors, and / or location tags, that can communicate with a base station, another device (e.g., remote device), or some other entity. A wireless node can provide, for example, connectivity for or to a network (e.g., a wide area network such as Internet or a cellular network) via a wired or wireless communication link. Some UEs can be considered Intemet-of-Things (IoT) devices, and / or can be implemented as NB-IoT (narrowband

[0033] In general, any number of wireless networks can be deployed in a given geographic area. Each wireless network can support a particular RAT and can operate on one or more frequencies. A RAT can also be referred to as a radio technology, an air interface, etc. A frequency can also be referred to as a carrier, a frequency channel, etc. Each frequency can support a single RAT in a given geographic area in order to avoid interference between wireless networks of different RATs. In some cases, NR or 5G RAT networks can be deployed.

[0034] In some aspects, two or more UEs 120 (e.g., shown as UE 120a and UE 120e) can communicate directly using one or more sidelink channels (e.g., without using base station 110 as an intermediary). For example, UEs 120 can communicate using peer-to-peer (P2P) communications, device-to-device (D2D) communications, vehicle-to-everything (V2X) protocols (which can include vehicle-to- vehicle (V2V) protocols or vehicle-to-infrastructure (V2I) protocols), and / or netw orked communications. In this case, UE 120 can perform scheduling operations, resource selection operations, and / or other operations described elsewhere herein as being performed by base station 110.

[0035] Devices of wireless network 100 can communicate using the electromagnetic spectrum, which can be subdivided, based on frequency or wavelength, into various classes, bands, channels, and / or the like. For example, devices of wireless network 100 can communicate using an operating band having a first frequency range (FR1), which can span, for example, from 410 MHz to 7.125 GHz, and / or can communicate using an operating band having a second frequency range (FR2), which can span, for example, from 24.25 GHz to 52.6 GHz. Frequencies between FR1 and FR2 are sometimes referred to as mid-band frequencies. Although a portion of FR1 is greater than 6 GHz, FR1 is often referred to as a “sub-6 GHz” band. Similarly, FR2 is often referred to as a “millimeter wave” band despite being different from the extremely high frequency (EHF) band (30 GHz - 300 GHz) which is identified by the International Telecommunications Union (ITU) as a “millimeter wave” band. Thus, unless specifically stated otherwise, the term “sub-6 GHz” or the like, if used herein, can broadly represent frequencies less than 6 GHz, frequencies within FR1, and / or mid-band frequencies (e.g., greater than 7.125 GHz). Similarly, unless specifically stated otherwise, the term “millimeter wave” or the like, if used herein, can broadly represent frequencies within the EHF band, frequencies within FR2, and / or mid-band frequencies (e.g., less than 24.25 GHz). It is contemplated that the frequencies included in FR1 and FR2 can be modified, and techniques described herein are applicable to those modified frequencies ranges.

[0036] As indicated above, Figure 1 are provided as examples. Other examples can differ from what is described with respect to at least one of the following. Figure 1

[0037] Figure 2 ​is a diagram illustrating an example 200 of a base station 110 in communication with a UE 120 in a wireless network 100, in accordance with various aspects of the present disclosure. Base stations 110 can be equipped with T antennas 234a through 234t, and UEs 120 can be equipped with R antennas 252a through 252r, where in general T > 1 and R > 1.

[0038] At base station 110, a transmit processor 220 can receive data from a data source 212 for one or more UEs, select one or more modulation and coding schemes (MCSs) for each UE based at least in part on channel quality indicators (CQIs) received from the UE, process (e.g., encode and modulate) the data for each UE based at least in part on the MCS(s) selected for the UE, and provide data symbols for all UEs. Transmit processor 220 can also process system information (e.g., for semi-static resource partitioning information (SRPI)) and control information (e.g., CQI requests, grants, and / or upper layer signaling) and provide overhead symbols and control symbols. Transmit processor 220 can also generate reference symbols for reference signals (e.g., a cell-specific reference signal (CRS) or a demodulation reference signal (DMRS)) and synchronization signals (e.g., a primary synchronization signal (PSS) or a secondary synchronization signal (SSS)). A transmit (TX) multiple-input multiple-output (MIMO) processor 230 can perform spatial processing (e.g., precoding) on the data symbols, the control symbols, the overhead symbols, and / or the reference symbols, if applicable, and can provide T output symbol streams to T modulators (MODs) 232a through 232t. Each modulator 232 can process a respective output symbol stream (e.g., for OFDM) to obtain an output sample stream. Each modulator 232 can further process (e.g., convert to analog, amplify, filter, and upconvert) the output sample stream to obtain a downlink signal. T downlink signals from modulators 232a through 232t can be transmitted via T antennas 234a through 234t, respectively.

[0039] At the UE 120, the antennas 252a-252r can receive the downlink signals from the base station 110 and / or other base stations and can provide received signals to the demodulators (DEMODs) 254a-254r, respectively. Each demodulator 254 can condition (e.g., filter, amplify, downconvert, and digitize) a received signal to obtain input samples. Each demodulator 254 can further process the input samples (e.g., for OFDM) to obtain received symbols. A MIMO detector 256 can obtain received symbols from all R demodulators 254a-254r, perform MIMO detection on the received symbols if applicable, and provide detected symbols. A receive processor 258 can process (e.g., demodulate and decode) the detected symbols, provide decoded data for the UE 120 to a data sink 260, and provide decoded control information and system information to a controller / processor 280. The term “controller / processor” can refer to one or more controllers, one or more processors, or a combination thereof. A channel processor can determine reference signal received power (RSRP) parameters, received signal strength indicator (RSSI) parameters, reference signal received quality (RSRQ) parameters, and / or channel quality indicator (CQI) parameters, among other examples. In some aspects, one or more components of UE 120 can be included in a housing 284.

[0040] The network controller 130 can include a communication unit 294, a controller / processor 290, and a memory 292. The network controller 130 can include, for example, one or more devices in a core network. The network controller 130 can communicate with the base station 110 via the communication unit 294.

[0041] Antennas (e.g., antennas 234a-234t and / or antennas 252a-252r) can include or be included within one or more antenna panels, antenna groups, antenna element sets, and / or antenna arrays, among other examples. An antenna panel, antenna group, antenna element set, and / or antenna array can include one or more antenna elements. An antenna panel, antenna group, antenna element set, and / or antenna array can include a set of co-planar antenna elements and / or a set of non-co-planar antenna elements. An antenna panel, antenna group, antenna element set, and / or antenna array can include antenna elements within a single housing and / or antenna elements within multiple housings. An antenna panel, antenna group, antenna element set, and / or antenna array can include one or more antenna elements coupled to one or more transmit and / or receive components (such as one or more components of a transceiver 264). Figure 2 An antenna panel, antenna group, antenna element set, and / or antenna array can include one or more antenna elements coupled to one or more transmit and / or receive components (such as one or more components of a transceiver 264).

[0042] On the uplink, at UE 120, a transmit processor 264 can receive and process data from a data source 262 and control information (e.g., for reports including RSRP, RSSI, RSRQ, and / or CQI) from controller / processor 280. Transmit processor 264 can also generate reference symbols for one or more reference signals. The symbols from transmit processor 264 can be precoded by a TX MIMO processor 266 if applicable, further processed by modulators 254a through 254r (e.g., for DFT-s-OFDM or CP-OFDM), and transmitted to base station 110. In some aspects, a modulator and a demodulator (e.g., MOD / DEMOD 254) of the UE 120 can be included in a modem of the UE 120. In some aspects, the UE 120 includes a transceiver. The transceiver can include any combination of antennas 252, modulators and / or demodulators 254, MIMO detector 256, receive processor 258, transmit processor 264, and / or TX MIMO processor 266. The transceiver can be used by a processor (e.g., controller / processor 280) and memory 282 to perform any of the methods described herein (for example, as described with reference to Figures 4-8 FIGS. 15 and 16).

[0043] At base station 110, the uplink signals from UE 120 and other UEs can be received by antennas 234, processed by demodulators 232, detected by a MIMO detector 236 if applicable, and further processed by a receive processor 238 to obtain decoded data and control information sent by UE 120. Receive processor 238 can provide the decoded data to a data sink 239 and the decoded control information to controller / processor 240. Base station 110 can include communication unit 244 and communicate to network controller 130 via communication unit 244. Base station 110 can include a scheduler 246 to schedule UEs 120 for downlink and / or uplink communications. In some aspects, a modulator and a demodulator (e.g., MOD / DEMOD 232) of the base station 110 can be included in a modem of the base station 110. In some aspects, the base station 110 includes a transceiver. The transceiver can include any combination of antennas 234, modulators and / or demodulators 232, MIMO detector 236, receive processor 238, transmit processor 220, and / or TX MIMO processor 230. The transceiver can be used by a processor (e.g., controller / processor 240) and memory 242 to perform any of the methods described herein (for example, as described with reference to Figures 4-8 FIGS. 15 and 16).

[0044] The controller / processor 240 of base station 110, the controller / processor 280 of UE 120 and / or Figure 2 Any other component may perform one or more techniques associated with communication using a neural network at least in part based on an indicated input size and / or input structure, as described in more detail elsewhere herein. For example, the controller / processor 240 of base station 110, the controller / processor 280 of UE 120, and / or Figure 2 Any other component can perform or direct, for example Figure 7 Process 700 Figure 8 The operation of process 800 and / or other processes as described herein. Memory 242 and 282 may store data and program code for base station 110 and UE 120, respectively. In some aspects, memory 242 and / or memory 282 may include a non-transitory computer-readable medium storing one or more instructions (e.g., code and / or program code) for wireless communication. For example, one or more instructions, when executed by one or more processors of base station 110 and / or UE 120 (e.g., directly, or after compilation, translation, and / or interpretation), may cause one or more processors, UE 120, and / or base station 110 to perform or direct, for example... Figure 7 Process 700 Figure 8 The operation of process 800 and / or other processes as described herein. In some aspects, execution instructions may include run instructions, translation instructions, compilation instructions and / or interpretation instructions, and other examples.

[0045] In some aspects, the UE includes: a unit for receiving an indication of one or more of the input size or input structure for a transmitting-side neural network, wherein the input structure indicates one or more of the positions of non-zero values ​​or non-zero values ​​within an input having the input size leading to the transmitting-side neural network; or a unit for communicating with a base station at least in part based on the indication of one or more of the input size or input structure for the transmitting-side neural network. The unit for the UE to perform the operations described herein may include, for example, one or more of antenna 252, demodulator 254, MIMO detector 256, receiver processor 258, transmitter processor 264, TX MIMO processor 266, modulator 254, controller / processor 280, or memory 282.

[0046] In some aspects, the UE includes: a unit for receiving instructions within scheduling permissions.

[0047] In some aspects, the UE includes: a unit for receiving one or more parameters of input size or input structure.

[0048] In some aspects, the UE includes means for determining an uplink control channel format configuration based at least in part on an indication of one or more of an input size or an input structure.

[0049] In some aspects, the UE includes means for receiving, within a scheduling grant for a downlink shared channel transmission, an indication associated with uplink control channel resources.

[0050] In some aspects, the UE includes means for receiving, from a base station via a downlink control channel, a transmission generated by the base station based at least in part on a transmit-side neural network.

[0051] In some aspects, the base station includes means for transmitting an indication of one or more of an input size or an input structure for a transmit-side neural network, where the input structure indicates one or more of a non-zero value or a location of a non-zero value within an input of the input size to the transmit-side neural network; or means for communicating with a UE based at least in part on the indication of one or more of the input size or the input structure of the transmit-side neural network. Means for the base station to perform operations described herein can include, for example, one or more of transmit processor 220, TX MIMO processor 230, modulator 232, antenna 234, demodulator 232, MIMO detector 236, receive processor 238, controller / processor 240, memory 242, or scheduler 246.

[0052] In some aspects, the base station includes means for transmitting the indication within a scheduling grant; or means for transmitting the indication within an uplink control channel format configuration.

[0053] In some aspects, the base station includes means for transmitting the indication within a scheduling grant; or means for transmitting the indication within an uplink control channel format configuration.

[0054] Although Figure 2 The blocks in FIG. 13 are illustrated as distinct components, but the functionality described above in relation to these blocks can be implemented in a single hardware, software, or combined component or in various combinations of components. For example, the functionality described in relation to transmit processor 264, receive processor 258, and / or TX MIMO processor 266 can be performed by controller / processor 280 or under the control of controller / processor 280.

[0055] As indicated above, Figure 2 are provided by way of example. Other examples can differ from those described in relation to Figure 2 the examples described in relation to

[0056] Figure 3is a diagram illustrating an example 300 of neural-network-based communication, in accordance with various aspects of the present disclosure. A transmit-side neural network can be stored, executed, operated, and / or maintained on a transmitting device (e.g., a UE or a base station). A receive-side neural network can be stored, executed, operated, and / or maintained on a receiving device (e.g., a UE or a base station). The receive-side neural network can be associated with the transmit-side neural network. For example, the receive-side neural network can be configured to decode bits (e.g., bits carried on signaling) encoded by the transmit-side neural network (e.g., using a neural network with mirrored operations from the transmit-side neural network).

[0057] As shown in Figure 3 As shown by reference number 305, the transmit-side neural network can receive source bits for encoding. The source bits can be associated with control information and / or data for transmission to a receiving device.

[0058] As shown by reference number 310, the transmitting device can encode the source bits using the transmit-side neural network. For example, the transmitting device can provide the source bits as input to the transmit-side neural network, and can obtain outputted encoded bits as output from the transmit-side neural network. The encoded bits can have a reduced payload when compared to a payload of the source bits.

[0059] The transmit-side neural network can perform (e.g., the transmitting device can use the transmit-side neural network to perform) one or more operations on the source bits, such as one or more convolution operations (e.g., by a convolutional layer), flattening operations (e.g., by a flattening layer), and / or fully connected operations (e.g., by a fully connected layer), among other examples.

[0060] The transmitting device can map the encoded bits to radio resources (e.g., resource elements (REs), RE segments, and / or physical resource blocks (PRBs), among other examples).

[0061] As shown by reference number 315, the transmitting device can transmit the encoded bits to the receiving device via a signal mapped to radio resources of a wireless channel. For example, the transmitting device can map the encoded bits to carrier signaling for transmission during time-frequency resources (e.g., allocated by the transmitting device or the receiving device, among other examples) for transmitting the encoded bits. The transmitting device can provide the mapped encoded bits to one or more antennas for transmission to the receiving device.

[0062] As shown by reference number 320, the receiving device can decode the received signal using a receive-side neural network. For example, one or more antennas of the receiving device can receive a signal carrying encoded bits. The receiving device can perform one or more operations to extract the encoded bits from the signaling. The one or more operations can mirror one or more operations performed by the transmit-side neural network to encode the encoded bits.

[0063] The receiving device can provide the encoded bits to the receive-side neural network for decoding. The receive-side neural network can perform one or more operations on the encoded bits, such as one or more convolution operations (e.g., by a convolutional layer), reshaping operations (e.g., by a reshaping layer), and / or fully connected operations (e.g., by a fully connected layer), among other examples.

[0064] As shown by reference number 325, the receiving device can output decoded bits from the receive-side neural network. The decoded bits can include some or all of the bits in the source bits.

[0065] As indicated above, Figure 3 are provided by way of example. Other examples can differ from what is described Figure 3 without departing from the spirit of the disclosure.

[0066] In some wireless networks, using a transmit-side neural network and an associated receive-side neural network for communication between a transmitting device and a receiving device can improve the efficiency of wireless communications. For example, the transmitting device can utilize operations of the transmit-side neural network in place of encoding, modulation, and / or precoding. Additionally or alternatively, the receiving device can utilize operations of the transmit-side neural network in place of synchronization, channel estimation, detection, decoding, and / or demodulation. In some wireless networks, the neural networks can replace one or more transmit and / or receive modules for performing encoding, modulation, precoding, synchronization, channel estimation, detection, decoding, and / or demodulation.

[0067] The transmitting device and / or the receiving device can train the transmit-side neural network and / or the receive-side neural network using offline training. The transmitting device and / or the receiving device can refine (e.g., update) the transmit-side neural network and / or the receive-side neural network using online refinement (e.g., receive updates to the neural network, such as weights of layers of the neural network, among other examples).

[0068] The transmit-side neural network and / or the transmitting device using the transmit-side neural network can be defined as an autoencoder. Similarly, the receive-side neural network and / or the receiving device using the receive-side neural network can be defined as an autoencoder.

[0069] In some wireless networks, a neural network input for autoencoder-based transmission includes one or more one-hot vectors. A one-hot vector includes a vector of entries, where only one entry has a value of 1 and all other entries have a value of 0. The length of a one-hot vector can be M (which can be a power of 2), which can carry M candidate messages (e.g., each candidate message is associated with a different position of the entry having a value of 1) and / or log2M bits. In these wireless networks, the input size grows linearly with the value of M, and the increased input size increases the size and / or complexity of the transmitting-side neural network and / or an associated receiving-side neural network. In other words, to increase the number of candidate messages, a transmitting device can consume additional computational resources to use the transmitting-side neural network, and / or a receiving device can consume additional computational resources to use the receiving-side neural network.

[0070] In some wireless networks, a neural network input for autoencoder-based transmission includes an M-select-m vector. An M-select-m vector can include a vector having a size M (e.g., having M entries) with m entries having a non-zero value. The non-zero value can be the same for all m entries. The m entries can have values that are normalized to represent probabilities (e.g., the length of the vector M can be 1). For an M-select-m vector, the number of candidate messages can be and / or bits. An M-select-m vector can provide an output with reduced complexity when compared to a one-hot vector that provides the same output. For example, for an M-select-m vector with M = 8 and m = 4, the throughput is the same as a one-hot vector input with a length of 64. In other words, the example M-select-m vector has an input size that is 1 / 8 of an equivalent one-hot vector to provide 64 candidate messages.

[0071] While neural network-based communications with relatively high complexity (e.g., with a large size (M) and structure (size m)) provide relatively high throughput, the complexity of the transmitting-side neural network can impact the complexity of detection at the receiving-side neural network. The ability of a receiver device to decode bits encoded with a transmitting-side neural network can depend on channel conditions, such as a signal-to-interference-plus-noise ratio (SINR), RSRP, RSSI, RSRQ, and / or CQI, among others. Based at least in part on using a transmitting-side neural network with a complexity not supported by the channel conditions, a receiving device can fail to decode bits encoded with the transmitting-side neural network. Failing to decode bits encoded with the transmitting-side neural network can consume computational, network, communication, and / or power resources to detect and correct (e.g., using signaling between the transmitting device and the receiving device and / or retransmission).

[0072] In some aspects described herein, a UE (e.g., the UE 120) can receive an indication of an input size and / or an input structure for a transmit-side neural network. The input size can indicate a number of values M in an input vector of an M-select-m vector, and the input structure can indicate a number of non-zero values m and / or locations of non-zero values within the input vector. The UE can communicate with a base station based at least in part on the indication of the input size and / or the input structure of the transmit-side neural network. In some aspects, the UE can be a transmitting device that uses the transmit-side neural network. In some aspects, the base station can be a transmitting device that uses the transmit-side neural network.

[0073] Based at least in part on the UE communicating with the base station based at least in part on the indication of the input size and / or the input structure, the base station can indicate one or more parameters that affect a complexity of a communication using the transmit-side neural network. In this way, the base station can dynamically and / or semi-statically adjust a complexity of the transmit-side neural network and / or a receive-side neural network. This can allow the base station to change the complexity based at least in part on one or more parameters, such as channel conditions, that can affect a likelihood of reception by a receiving device. In this way, the base station and / or the UE can conserve computational, network, communication, and / or power resources to detect and correct for failures to receive and / or decode bits transmitted using the transmit-side neural network.

[0074] In some aspects, the base station can transmit an uplink grant for the UE to transmit a physical uplink shared channel (PUSCH) communication and / or a physical uplink control channel (PUCCH) communication. In some aspects, the base station can transmit a downlink grant for the UE to receive a physical downlink shared channel (PDSCH) communication and / or a physical downlink control channel (PDCCH) communication. The uplink grant or the downlink grant can configure and / or indicate a configuration of an input size and / or an input structure for a transmission of a communication to be used for scheduling (e.g., by the UE or by the base station).

[0075] In some aspects, the input size and / or the input structure can be partially defined. In these aspects, the uplink grant or the downlink grant can configure and / or indicate values of the input size and / or the input structure that are not defined. For example, a value of M can be defined in a communication standard, and the uplink grant or the downlink grant can configure and / or indicate a value of m and / or locations of non-zero values within the input vector.

[0076] In some aspects, the indication can identify (e.g., can be used by the UE to identify) an associated receiving-side neural network for receiving (e.g., detecting) the encoded bits transmitted using the transmitting-side neural network.

[0077] In some aspects, a first transmitting-side neural network can be used to encode a first RE segment of a PRB to be transmitted, a second transmitting-side neural network can be used to encode a second RE segment of the PRB, and / or a third transmitting-side neural network can be used to encode a third RE segment of the PRB, among other examples. Each RE segment can include a number of REs of the PRB. A source bit sequence can be segmented into a plurality of sub-sequences, where each of the sub-sequences is associated with a different RE segment.

[0078] In some aspects, each RE segment can include a common number of REs, a common number of OFDM symbols, and / or a common input size and input structure for the transmitting-side neural network. In some aspects, one or more RE segments can have a different number of REs, a different number of OFDM symbols, and / or a different input size and / or input structure for the transmitting-side neural network used to encode bits of the one or more RE segments. For example, a first segment can be control signaling piggybacked on one or more segments used to carry data signaling.

[0079] In some aspects, an uplink grant or a downlink grant can implicitly or explicitly configure and / or indicate the input size and / or input structure associated with different RE segments. For example, the uplink grant or the downlink grant can implicitly configure and / or indicate the input size and / or input structure based at least in part on an indicated RE segment associated with the input size and / or input structure (e.g., based at least in part on an indicated definition, a communication standard, and / or radio resource control (RRC) configuration signaling, among other examples).

[0080] In some aspects, one or more of the RE segments can be associated with different cyclic redundancy checks and / or different hybrid automatic repeat request (HARQ) feedback. In this way, a receiving device can individually indicate HARQ feedback for segments having different input sizes and / or input structures. This can provide information for updating the input size and / or input structure for subsequent communications.

[0081] In some aspects, a UE can be configured to transmit uplink control information (e.g., a PUCCH communication) using a transmit-side neural network having an indicated input size and / or input structure. In some aspects, a PUCCH format can be defined based at least in part on the indicated input size and / or input structure. In some aspects, different PUCCH resources within a PUCCH resource set can be associated with different input sizes and / or input structures. In some aspects, all PUCCH resources within a PUCCH resource set can be associated with a same set of input sizes and / or input structures. In some aspects, a base station can indicate, within downlink control information (DCI) and / or RRC signaling that schedules a PDSCH communication, an input size and / or input structure of one or more PUCCH resources for HARQ feedback (e.g., HARQ-ACK) feedback associated with the PDSCH communication.

[0082] In some aspects, a UE can receive an indication that a base station will transmit a PDCCH communication using a transmit-side neural network having an indicated input size and / or input structure. In some aspects, a first input size and / or input structure can apply to a first REG bundling and / or a first control channel element. In some aspects, a second input size and / or input structure can apply to a second REG bundling and / or a second control channel element. In some aspects, all REG bundlings and control channel elements can have a same input size and / or input structure.

[0083] In some aspects, different aggregation levels can be configured and / or predefined with different input sizes and / or input structures. In some aspects, different search spaces and / or different control resource sets can be configured and / or predefined with different input sizes and / or input structures. Different search spaces and / or different control resource sets having different input sizes and / or input structures can have different priorities for monitoring by a UE when a PDCCH is overbooked (e.g., the UE is not configured to or is unable to monitor all search spaces and / or control resource sets), and / or when the UE is configured in a power saving mode (e.g., a discontinuous reception (DRX) mode or a sleep mode, etc.). For example, when the UE is configured in a DRX mode and / or if a number of blind decoding attempts or a channel estimation threshold is less than a determined number of PDCCH decodes (e.g., associated with a number of search spaces and / or control resource sets), a search space and / or control resource set associated with an input size and / or input structure that includes a relatively large number of neurons (e.g., having a relatively high complexity) can have a relatively low priority.

[0084] Figure 4is a diagram illustrating example 400 associated with communications using a neural network based at least in part on an indicated input size and / or input structure, in accordance with various aspects of the present disclosure. As shown, a UE (e.g., UE 120) can communicate with a base station (e.g., base station 110). The UE and base station can be part of a wireless network (e.g., wireless network 100). In some aspects, the UE and base station can be configured with one or more of a transmit-side neural network or a receive-side neural network. In some aspects, the UE is configured with a transmit-side neural network for uplink communications and a receive-side neural network for downlink communications. In some aspects, the base station is configured with a transmit-side neural network for downlink communications and a receive-side neural network for uplink communications. Figure 4 As shown, a UE (e.g., UE 120) can communicate with a base station (e.g., base station 110). The UE and base station can be part of a wireless network (e.g., wireless network 100). In some aspects, the UE and base station can be configured with one or more of a transmit-side neural network or a receive-side neural network. In some aspects, the UE is configured with a transmit-side neural network for uplink communications and a receive-side neural network for downlink communications. In some aspects, the base station is configured with a transmit-side neural network for downlink communications and a receive-side neural network for uplink communications.

[0085] As shown, a UE (e.g., UE 120) can communicate with a base station (e.g., base station 110). The UE and base station can be part of a wireless network (e.g., wireless network 100). In some aspects, the UE and base station can be configured with one or more of a transmit-side neural network or a receive-side neural network. In some aspects, the UE is configured with a transmit-side neural network for uplink communications and a receive-side neural network for downlink communications. In some aspects, the base station is configured with a transmit-side neural network for downlink communications and a receive-side neural network for uplink communications.

[0086] In some aspects, the configuration information can indicate that the UE is to use a transmit-side neural network for uplink communications with the base station. In some aspects, the configuration information can indicate that the UE is to use a receive-side neural network for receiving downlink communications from the base station. In some aspects, the configuration information can indicate that the UE is to use a default input size and / or input structure for the transmit-side neural network and / or for the receive-side neural network until or unless the base station transmits an indication of an input size and / or input structure for one or more communications (e.g., uplink communications and / or downlink communications).

[0087] As shown, a UE (e.g., UE 120) can communicate with a base station (e.g., base station 110). The UE and base station can be part of a wireless network (e.g., wireless network 100). In some aspects, the UE and base station can be configured with one or more of a transmit-side neural network or a receive-side neural network. In some aspects, the UE is configured with a transmit-side neural network for uplink communications and a receive-side neural network for downlink communications. In some aspects, the base station is configured with a transmit-side neural network for downlink communications and a receive-side neural network for uplink communications.

[0088] As shown by reference number 415, the UE can receive and the base station can transmit an indication of an input size and / or an input structure for a transmit-side (Tx-side) neural network (NN). In some aspects, the UE can receive an indication of one or more of an input size M or an input structure for a transmit-side neural network, where the input structure indicates one or more of non-zero values m or locations of non-zero values within an input of the input size to the transmit-side neural network. In some aspects, the UE can receive the indication within a scheduling grant (e.g., a grant for a PUSCH).

[0089] In some aspects, the indication can include a definition of one or more parameters of the input size and / or the input structure. For example, the indication can define a number of one or more non-zero values and / or locations of non-zero values within the input.

[0090] In some implementations, the indication can include an indication of one or more receive-side neural networks associated with the transmit-side neural network. For example, if the indication includes an indication that the base station will use a transmit-side neural network having the input size and / or the input structure, the indication can include an implicit and / or explicit indication of one or more receive-side neural networks to use to decode a communication transmitted by the base station. In some implementations, the indicated receive-side neural networks can have a corresponding (e.g., mirrored) input size and / or input structure from the transmit-side neural network.

[0091] In some implementations, the input size and / or the input structure is associated with one or more RE segments of a PRB. In some aspects, the input size and / or the input structure is associated with all RE segments of a PRB. In some aspects, an additional input size and / or an additional input structure is associated with one or more additional RE segments of the PRB. In some aspects, the indication can include an indication of the additional input size and / or the additional input structure for the one or more additional RE segments of the PRB.

[0092] As shown by reference number 420, the UE can input the transmission bits into a transmit-side neural network having the input size and / or the input structure. In some aspects, the UE can input all of the transmission bits into a single transmit-side neural network having a common input size and / or input structure. In some aspects, the UE can input different RE segments of the PRB into different transmit-side neural networks and / or into a common transmit-side neural network having different input sizes and / or input structures.

[0093] As shown by reference number 425, the UE can transmit the PRB based at least in part on the output of the transmit-side neural network, and the base station can receive the PRB based at least in part on the output of the transmit-side neural network. In some aspects, the UE can transmit the PRB with normalized power over one or more RE segments of the PRB after applying the transmit-side neural network.

[0094] As shown by reference number 430, the base station can decode the PRB based at least in part on a receive-side (Rx-side) neural network associated with the transmit-side neural network used by the UE to transmit the PRB. For example, the base station can decode the PRB based at least in part on inputting encoded bits extracted from signaling carrying the PRB to the receive-side neural network. In some aspects, the base station can decode the PRB based at least in part on the receive-side neural network outputting decoded bits.

[0095] The operations shown by reference numbers 420-430 can be associated with uplink communications in which the UE uses a transmit-side neural network. For example, the operations can be associated with the UE transmitting data using a PUSCH.

[0096] As shown by reference number 435, the base station can input the transport bits into a transmit-side neural network having an input size and / or an input structure. In some aspects, the base station can input all of the transport bits into a single transmit-side neural network having a common input size and / or input structure. In some aspects, the base station can input different RE segments of the PRB into different transmit-side neural networks and / or into a common transmit-side neural network having different input sizes and / or input structures.

[0097] As shown by reference number 440, the base station can transmit the PRB based at least in part on the output of the transmit-side neural network, and the UE can receive the PRB based at least in part on the output of the transmit-side neural network. In some aspects, the base station can transmit the PRB with normalized power over one or more RE segments of the PRB after applying the transmit-side neural network.

[0098] As shown by reference number 445, the UE can decode the PRB based at least in part on a receive-side neural network associated with the transmit-side neural network used by the base station to transmit the PRB. For example, the UE can decode the PRB based at least in part on inputting encoded bits extracted from signaling carrying the PRB to the receive-side neural network. In some aspects, the UE can decode the PRB based at least in part on the receive-side neural network outputting decoded bits.

[0099] Operations shown by reference numbers 435-445 can be associated with downlink communications in which the UE uses a receive-side neural network. For example, these operations can be associated with the UE receiving data using a PDSCH.

[0100] Based at least in part on the UE communicating with the base station based at least in part on the indication of the input size and / or the input structure, the base station can indicate one or more parameters that affect a complexity of communications using a transmit-side neural network. In this way, the base station can dynamically and / or semi-statically adjust a complexity of a transmit-side neural network and / or a receive-side neural network. This can allow the base station to change the complexity based at least in part on one or more parameters, such as channel conditions, that can affect a likelihood of reception by a receiving device. In this way, the base station and / or the UE can conserve computational, network, communication, and / or power resources to detect and correct for failures to receive and / or decode bits transmitted using a transmit-side neural network.

[0101] As indicated above, Figure 4 are provided by way of example. Other examples can differ from those described Figure 4 without departing from the scope of the disclosure.

[0102] Figure 5 is a diagram illustrating an example 500 associated with communications using a neural network based at least in part on an indicated input size and / or input structure, in accordance with various aspects of the present disclosure. As shown by Figure 5 the UE (e.g., a UE 120) can communicate with a base station (e.g., a base station 110). The UE and the base station can be part of a wireless network (e.g., the wireless network 100). In some aspects, the UE and the base station can be configured with one or more of a transmit-side neural network or a receive-side neural network. In some aspects, the UE is configured with a transmit-side neural network for uplink communications and a receive-side neural network for downlink communications. In some aspects, the base station is configured with a transmit-side neural network for downlink communications and a receive-side neural network for uplink communications.

[0103] As shown by reference number 505, the base station can transmit configuration information, and the UE can receive the configuration information. In some aspects, the UE can receive the configuration information from another device (e.g., from another base station and / or another UE) and / or receive the configuration information according to a communication standard, among other examples. In some aspects, the UE can receive the configuration information via one or more of RRC signaling, MAC-CE signaling, or DCI signaling, and / or the UE can determine the configuration information according to a communication standard, among other examples. In some aspects, the configuration information can include an indication of one or more configuration parameters (e.g., already known by the UE) for selection by the UE, explicit configuration information for the UE to use to configure the UE, and / or the like.

[0104] In some aspects, the configuration information can indicate that the UE is to use a transmit-side neural network for uplink communications with the base station. In some aspects, the configuration information can indicate that the UE is to use a receive-side neural network for receiving downlink communications from the base station. In some aspects, the configuration information can indicate that the UE is to use a default input size and / or input structure for the transmit-side neural network and / or for the receive-side neural network until or unless the base station transmits an indication of an input size and / or input structure for one or more communications (e.g., uplink communications and / or downlink communications).

[0105] As shown by reference number 510, the UE can configure the UE for communications with the base station. In some aspects, the UE can configure the UE based at least in part on the configuration information. In some aspects, the UE can be configured to perform one or more operations described herein.

[0106] As shown by reference number 515, the UE can receive and the base station can transmit an indication of an input size and / or input structure for a transmit-side (Tx-side) neural network (NN). In some aspects, the UE can receive an indication of one or more of an input size M or an input structure for the transmit-side neural network, where the input structure indicates one or more of a non-zero value m or a location of a non-zero value within an input of the input size to the transmit-side neural network. In some aspects, the UE can receive the indication within a scheduling grant (e.g., a grant for a PUCCH, or a grant for a PDSCH with an associated PUCCH).

[0107] In some aspects, the indication can include a definition of one or more parameters of the input size and / or the input structure. For example, the indication can define a number of one or more non-zero values and / or a location of a non-zero value within the input.

[0108] In some implementations, the indication can include an indication of one or more receive-side neural networks associated with the transmit-side neural network. For example, if the indication includes an indication that the base station will use a transmit-side neural network having an input size and / or an input structure, the indication can include an implicit and / or explicit indication of one or more receive-side neural networks to be used to decode communications transmitted by the base station. In some implementations, the indicated receive-side neural networks can have a corresponding (e.g., mirrored) input size and / or input structure from the transmit-side neural network.

[0109] In some aspects, a first uplink control channel (e.g., PUCCH) resource within the first set of uplink control channel resources is associated with an input size and / or an input structure. In some aspects, a second uplink control channel resource within the first set of uplink control channel resources is associated with an additional input size and / or an additional input structure.

[0110] In some aspects, a UE can receive an indication associated with an uplink control channel resource within a scheduling grant for a downlink shared channel (e.g., PDSCH) transmission. In these aspects, the UE can transmit HARQ feedback using the uplink control channel resource via a transmit-side neural network using an input size and / or an input structure.

[0111] In some aspects, an input size and / or an input structure can be associated with one or more RE group bundles, one or more control channel elements, one or more aggregation levels, one or more search spaces, and / or one or more control resource sets, among other examples. In some aspects, different input sizes and / or input structures can be associated with different RE group bundles, control channel elements, aggregation levels, search spaces, and / or control resource sets, among other examples.

[0112] As shown by reference number 520, the UE can input uplink control information (UCI) bits and / or HARQ feedback bits into a transmit-side neural network having an input size and / or an input structure. In some aspects, the UE can use a common transmit-side neural network having a common input size and / or input structure for all UCI bits and / or HARQ feedback bits. In some aspects, the UE can use different transmit-side neural networks and / or different input sizes and / or input structures for UCI bits and / or HARQ feedback bits to be transmitted via different sets of uplink control channel resources.

[0113] As shown by reference number 525, the UE can determine an uplink control channel format configuration. In some aspects, the UE can determine the uplink control channel format configuration based at least in part on the indication of the input size and / or the input structure. For example, the UE can be configured with an association (e.g., a mapping) between the uplink control channel format configuration and the indication of the input size and / or the input structure.

[0114] As shown by reference number 530, the UE can transmit the UCI and / or HARQ feedback based at least in part on the output of the transmit-side neural network, and the base station can receive the UCI and / or HARQ feedback based at least in part on the output of the transmit-side neural network.

[0115] As shown by reference number 535, the base station can decode the UCI and / or HARQ feedback based at least in part on a receive-side (Rx-side) neural network associated with the transmit-side neural network used by the UE to transmit the PRB. For example, the base station can decode the PRB based at least in part on inputting encoded bits extracted from the signaling carrying the UCI and / or HARQ feedback to the receive-side neural network. In some aspects, the base station can decode the UCI and / or HARQ feedback based at least in part on the receive-side neural network outputting decoded bits.

[0116] The operations shown by reference numbers 520-535 can be associated with uplink communications in which the UE uses a transmit-side neural network. For example, these operations can be associated with the UE transmitting data using a PUCCH.

[0117] As shown by reference number 540, the base station can input the DCI bits into a transmit-side neural network having an input size and / or an input structure. In some aspects, the base station can use a common transmit-side neural network having a common input size and / or input structure for all DCI bits. In some aspects, the base station can use different transmit-side neural networks and / or different input sizes and / or input structures for DCI bits to be transmitted via different sets of uplink control channel resources.

[0118] As shown by reference number 545, the base station can transmit the DCI based at least in part on an output of the transmit-side neural network, and the UE can receive the DCI based at least in part on the output of the transmit-side neural network. For example, the UE can receive the DCI generated by the base station based at least in part on the transmit-side neural network. The input size and / or the input structure can be associated with one or more RE group bundles, one or more control channel elements, one or more aggregation levels, one or more search spaces, and / or one or more control resource sets, among other examples. In some aspects, different input sizes and / or different input structures can be associated with different RE group bundles, different control channel elements, different aggregation levels, different search spaces, and / or different control resource sets, among other examples.

[0119] As shown by reference number 550, the UE can decode the DCI based at least in part on a receive-side neural network associated with the transmit-side neural network used by the base station to transmit the DCI. For example, the UE can decode the DCI based at least in part on inputting encoded bits extracted from the signaling carrying the DCI to the receive-side neural network. In some aspects, the UE can decode the DCI based at least in part on the receive-side neural network outputting decoded bits.

[0120] In some aspects, the UE can be configured to prioritize monitoring of the first control resource set or the second control resource set based at least in part on the UE operating in a power saving mode, a number of blind decoding attempts satisfying a threshold, and / or a number of channel estimation attempts satisfying a threshold.

[0121] The operations shown by reference numbers 540-550 can be associated with downlink communications in which the UE uses a receive-side neural network. For example, these operations can be associated with the UE receiving data using a PDCCH.

[0122] As indicated above, Figure 5 are provided by way of example. Other examples can differ from those described Figure 5 without departing from the spirit of the disclosure.

[0123] Figure 6 is a diagram illustrating an example 600 associated with communications using a neural network based at least in part on an indicated input size and / or input structure, in accordance with various aspects of the present disclosure. As shown by Figure 6 the transmitting device and the receiving device can be part of a wireless network (e.g., wireless network 100). In some aspects, the transmitting device can be configured with one or more transmit-side neural networks.

[0124] As shown in Figure 6 the first bit 605 can be provided as input to a first neural network 625, the second bit 610 can be provided as input to a second neural network 630, the third bit 615 can be provided as input to a third neural network 635, and / or the fourth bit 620 can be provided as input to a fourth neural network 640. In some aspects, each of the neural networks can be a different neural network. In some aspects, each of the neural networks can be the same neural network. In some aspects, one or more of the neural networks can be the same neural network with different parameters, such as an input size and / or an input structure for the input.

[0125] As shown in Figure 6 further, the first neural network 625 can generate a first neural network output (e.g., encoded bits) mapped to a first segment 645 of REs, the second neural network 630 can generate a second neural network output mapped to a second segment 650 of REs, the third neural network 635 can generate a third neural network output mapped to a third segment 655 of REs, and / or the fourth neural network 640 can generate a fourth neural network output mapped to a fourth segment 660 of REs.

[0126] As shown in Figure 6 further, the first neural network output mapped to the first segment 645 of REs, the second neural network output mapped to the second segment 650 of REs, the third neural network output mapped to the third segment 655 of REs, and / or the fourth neural network output mapped to the fourth segment 660 of REs can be normalized (e.g., with a transmit power) to form a PRB with power normalized by segment 665. The transmitting device can transmit the PRB with power normalized by segment 665 to a receiving device for decoding using one or more receive-side neural networks associated with the first neural network 625, the second neural network 630, the third neural network 635, and / or the fourth neural network 640.

[0127] As indicated above, Figure 6 are provided by way of example. Other examples can differ from those described in this regard. Figure 6

[0128] Figure 7 ​is a diagram illustrating an example process 700 performed, for example, by a UE, in accordance with various aspects of the present disclosure. Example process 700 is an example where the UE (e.g., UE 120) performs operations associated with communications using a neural network based at least in part on an indicated input size and / or input structure.

[0129] As further shown, in some aspects, process 700 can include communicating with the base station based at least in part on the indication of one or more of the input size or the input structure of the transmitting-side neural network (block 720). For example, the UE (e.g., using reception component 902 and / or transmission component 904, depicted in FIG. 9) can communicate with the base station based at least in part on the indication of one or more of the input size or the input structure of the transmitting-side neural network, as described above. Figure 7 For example, the UE (e.g., using reception component 902, depicted in FIG. 9) can receive the indication of one or more of the input size or the input structure of the transmitting-side neural network, as described above. Figure 9 For example, the UE (e.g., using reception component 902, depicted in FIG. 9) can receive the indication of one or more of the input size or the input structure of the transmitting-side neural network, as described above.

[0130] As further shown, in some aspects, process 700 can include communicating with the base station based at least in part on the indication of one or more of the input size or the input structure of the transmitting-side neural network (block 720). For example, the UE (e.g., using reception component 902 and / or transmission component 904, depicted in FIG. 9) can communicate with the base station based at least in part on the indication of one or more of the input size or the input structure of the transmitting-side neural network, as described above. Figure 7 For example, the UE (e.g., using reception component 902, depicted in FIG. 9) can receive the indication of one or more of the input size or the input structure of the transmitting-side neural network, as described above. Figure 9 For example, the UE (e.g., using reception component 902, depicted in FIG. 9) can receive the indication of one or more of the input size or the input structure of the transmitting-side neural network, as described above.

[0131] Process 700 can include additional aspects, such as any single aspect or any combination of aspects described below and / or in connection with one or more other processes described elsewhere herein.

[0132] In a first aspect, the transmitting-side neural network is associated with uplink communications, or the transmitting-side neural network is associated with downlink communications.

[0133] In a second aspect, alone or in combination with the first aspect, receiving the indication of one or more of the input size or the input structure includes receiving the indication within a scheduling grant.

[0134] In a third aspect, alone or in combination with one or more of the first and second aspects, receiving the indication of one or more of the input size or the input structure includes receiving a definition of one or more parameters of the input size or the input structure.

[0135] In a fourth aspect, alone or in combination with one or more of the first through third aspects, the indication of one or more of the input size or the input structure includes an indication of one or more receive-side neural networks associated with the transmit-side neural network.

[0136] In a fifth aspect, alone or in combination with one or more of the first through fourth aspects, one or more of the input size or the input structure is associated with one or more RE segments of a PRB, wherein one or more of the additional input size or the additional input structure is associated with one or more additional RE segments of the PRB, or a combination thereof.

[0137] In a sixth aspect, alone or in combination with one or more of the first through fifth aspects, the indication of one or more of the input size or the input structure includes an indication of one or more of the additional input size or the additional input structure.

[0138] In a seventh aspect, alone or in combination with one or more of the first through sixth aspects, the transmit device of the UE or the base station transmits the PRB with power normalized over one or more RE segments of a PRB after applying the transmit-side neural network.

[0139] In an eighth aspect, alone or in combination with one or more of the first through fifth aspects, the process 700 includes determining an uplink control channel format configuration based at least in part on the indication of one or more of the input size or the input structure.

[0140] In a ninth aspect, alone or in combination with the eighth aspect, a first uplink control channel resource within the first set of uplink control channel resources is associated with one or more of the input size or the input structure, and wherein a second uplink control channel resource within the first set of uplink control channel resources is associated with one or more of the additional input size or the additional input structure.

[0141] In a tenth aspect, alone or in combination with one or more of the first through fifth aspects, the eighth aspect, or the ninth aspect, receiving the indication of one or more of the input size or the input structure includes receiving the indication associated with the uplink control channel resource within a scheduling grant for a downlink shared channel transmission, wherein the communicating with the base station includes transmitting hybrid automatic repeat request (HARQ) feedback using the uplink control channel resource via the transmit-side neural network.

[0142] In a eleventh aspect, alone or in combination with one or more of the first through fifth aspects, communicating with the base station comprises: receiving, from the base station via a downlink control channel, a transmission generated by the base station based at least in part on a transmit-side neural network, wherein one or more of an input size or an input structure is associated with one or more of the one or more RE groups, one or more control channel elements, one or more aggregation levels, one or more search spaces, one or more control resource sets, or a combination thereof.

[0143] In a twelfth aspect, alone or in combination with one or more of the first through fifth aspects or the eleventh aspect, a first control resource set associated with one or more of an input size or an input structure is associated with a first priority, wherein a second control resource set associated with one or more of an additional input size or an additional input structure is associated with a second priority, and wherein the UE is configured to prioritize monitoring of the first control resource set or the second control resource set based at least in part on one or more of: the UE operating in a power saving mode, a number of blind decoding attempts satisfying a threshold, or a number of channel estimation attempts satisfying a threshold.

[0144] Although Figure 7 Example blocks of the process 700 are illustrated, but in some aspects, the process 700 can include additional blocks, fewer blocks, different blocks, or differently arranged blocks than those depicted in FIG. 7. Additionally, or alternatively, two or more of the blocks of the process 700 can be performed in parallel. Figure 7 In some aspects, the process 800 can include more, fewer, or a different

[0145] Figure 8 FIG. 8 is a diagram illustrating an example process 800 performed, for example, by a UE, in accordance with various aspects of the present disclosure. Example process 800 is an example where a base station (e.g., base station 110) performs operations associated with communications using a neural network based at least in part on an indicated input size and / or input structure.

[0146] As Figure 8 In some aspects, the process 800 can include transmitting an indication of one or more of an input size or an input structure for a transmit-side neural network, where the input structure indicates one or more of a non-zero value or a location of a non-zero value within an input having the input size to the transmit-side neural network (block 810), as described above. For example, the base station (e.g., using transmission component 1004 described Figure 10 In some aspects, the process 800 can include transmitting an indication of one or more of an input size or an input structure for a transmit-side neural network, where the input structure indicates one or more of a non-zero value or a location of a non-zero value within an input having the input size to the transmit-side neural network (block 810), as described above. For example, the base station (e.g., using transmission component 1004 described In some aspects, the process 800 can include transmitting an indication of one or more of an input size or an input structure for a transmit-side neural network, where the input structure indicates one or more of a non-zero value or a location of a non-zero value within an input having the input size to the transmit-side neural network (block 810), as described above. For example, the base station (e.g., using transmission component 1004 described

[0147] As Figure 8 Further as Figure 10 the reception component 1002 and / or the transmission component 1004 depicted in

[0148] Process 800 can include additional aspects, such as any single aspect or any combination of aspects described below and / or in connection with one or more other processes described elsewhere herein.

[0149] In a first aspect, transmitting the indication of one or more of the input size or the input structure comprises transmitting the indication within a scheduling grant or transmitting the indication within an uplink control channel format configuration.

[0150] Although Figure 8 Example blocks of the process 800 are illustrated, but in some aspects, the process 800 can include additional blocks, fewer blocks, different blocks, or differently arranged blocks than those depicted in Figure 8 the blocks depicted in

[0151] Figure 9 FIG. 9 is a block diagram of an example apparatus 900 for wireless communication. The apparatus 900 can be a UE, or a UE can include the apparatus 900. In some aspects, the apparatus 900 includes a reception component 902 and a transmission component 904, which can be in communication with one another (for example, via one or more buses and / or one or more other components). As shown, the apparatus 900 can communicate with another apparatus 906 (such as a UE, a base station, or another wireless communication device) using the reception component 902 and the transmission component 904. As further shown, the apparatus 900 can include a determination component 908.

[0152] In some aspects, the apparatus 900 can be configured to perform one or more operations described herein with reference to one or more of the methods described Figures 4-6 Additionally or alternatively, the apparatus 900 can be configured to perform one or more processes described herein, such as process 700 of Figure 7 In some aspects, the apparatus 900 and / or one or more components of the apparatus 900 can include one or more components of the UE described Figure 9 above with reference to FIG. 2. Additionally or alternatively, the apparatus 900 and / or one or more components of the apparatus 900 can include one or more components of the UE described Figure 2 above with reference to FIG. 2. Additionally or alternatively, the apparatus 900 and / or one or more components of the apparatus 900 can include one or more components of the UE described Figure 9One or more components illustrated in FIG. 10 can be included in the apparatus 1000, in combination, or exclusively. Figure 2 One or more components described can be implemented within the apparatus 1000. Additionally or alternatively, one or more components of the set of components can be implemented at least in part as software stored in a memory. For example, a component (or a portion of a component) can be implemented as instructions or code stored in a non-transitory computer-readable medium and executable by a controller or a processor to perform the functions or operations of the component.

[0153] The reception component 902 can receive communications, such as reference signals, control information, data communications, or a combination thereof, from the apparatus 906. The reception component 902 can provide received communications to one or more other components of the apparatus 900. In some aspects, the reception component 902 can perform signal processing on the received communications (such as filtering, amplification, demodulation, analog-to-digital conversion, demultiplexing, deinterleaving, de-mapping, equalization, interference cancellation, or decoding, among other examples), and can provide the processed signals to the one or more other components of the apparatus 906. In some aspects, the reception component 902 can include one or more antennas, a demodulator, a MIMO detector, a receive processor, a controller / processor, a memory, or a combination thereof, of the UE described above in connection with Fig. 2. Figure 2 The reception component 902 can receive communications, such as reference signals, control information, data communications, or a combination thereof, from the apparatus 906. The reception component 902 can provide received communications to one or more other components of the apparatus 900. In some aspects, the reception component 902 can perform signal processing on the received communications (such as filtering, amplification, demodulation, analog-to-digital conversion, demultiplexing, deinterleaving, de-mapping, equalization, interference cancellation, or decoding, among other examples), and can provide the processed signals to the one or more other components of the apparatus 906. In some aspects, the reception component 902 can include one or more antennas, a demodulator, a MIMO detector, a receive processor, a controller / processor, a memory, or a combination thereof, of the UE described above in connection with Fig. 2.

[0154] The transmission component 904 can transmit communications, such as reference signals, control information, data communications, or a combination thereof, to the apparatus 906. In some aspects, one or more other components of the apparatus 906 can generate communications and can provide the generated communications to the transmission component 904 for transmission to the apparatus 906. In some aspects, the transmission component 904 can perform signal processing on the generated communications (such as filtering, amplification, modulation, digital-to-analog conversion, multiplexing, interleaving, mapping, or encoding, among other examples), and can transmit the processed signals to the apparatus 906. In some aspects, the transmission component 904 can include one or more antennas, a modulator, a transmit MIMO processor, a transmit processor, a controller / processor, a memory, or a combination thereof, of the UE described above in connection with Fig. 2. In some aspects, the transmission component 904 can be co-located with the reception component 902 in a transceiver. Figure 2 The transmission component 904 can transmit communications, such as reference signals, control information, data communications, or a combination thereof, to the apparatus 906. In some aspects, one or more other components of the apparatus 906 can generate communications and can provide the generated communications to the transmission component 904 for transmission to the apparatus 906. In some aspects, the transmission component 904 can perform signal processing on the generated communications (such as filtering, amplification, modulation, digital-to-analog conversion, multiplexing, interleaving, mapping, or encoding, among other examples), and can transmit the processed signals to the apparatus 906. In some aspects, the transmission component 904 can include one or more antennas, a modulator, a transmit MIMO processor, a transmit processor, a controller / processor, a memory, or a combination thereof, of the UE described above in connection with Fig. 2. In some aspects, the transmission component 904 can be co-located with the reception component 902 in a transceiver.

[0155] The reception component 902 can receive an indication of one or more of an input size or an input structure for a transmitting-side neural network, where the input structure indicates one or more of non-zero values or locations of non-zero values within an input of the input size to the transmitting-side neural network. The reception component 902 and / or the transmission component 904 can communicate with the base station based at least in part on the indication of one or more of the input size or the input structure of the transmitting-side neural network.

[0156] Determining component 908 can determine the uplink control channel format configuration based at least in part on the indication of one or more of the input size or the input structure.

[0157] Determining component 908 and / or receiving component 902 can decode the PRB based at least in part on a receive-side neural network associated with a transmit-side neural network used by the base station to transmit the PRB.

[0158] Determining component 908 and / or transmitting component 904 can encode the UCI based at least in part on a transmit-side neural network having an input size and / or an input structure as indicated by the base station.

[0159] Determining component 908 and / or receiving component 902 can decode the DCI based at least in part on a receive-side neural network associated with a transmit-side neural network used by the base station to transmit the DCI.

[0160] Determining component 908 and / or transmitting component 904 can encode the UCI based at least in part on a transmit-side neural network having an input size and / or an input structure as indicated by the base station.

[0161] Figure 9 The number and arrangement of components shown is provided as an example. In practice, there can be additional components, fewer components, different components, or differently arranged components than those shown. Figure 9 than those illustrated, or arranged in a different manner than those illustrated. Furthermore, Figure 9 Two or more components illustrated can be implemented within a single component, or Figure 9 A single component illustrated can be implemented as multiple, distributed components. Additionally or alternatively, Figure 9 A set of one or more components illustrated can be configured to perform one or more functions described as being performed by another set of one or more components illustrated. Figure 9 one or more functions described as being performed by another set of one or more components illustrated.

[0162] Figure 10 FIG. 10 is a block diagram of an example apparatus 1000 for wireless communication. The apparatus 1000 can be a base station, or a base station can include the apparatus 1000. In some aspects, the apparatus 1000 includes a reception component 1002 and a transmission component 1004, which can be in communication with one another (for example, via one or more buses and / or one or more other components). As shown, the apparatus 1000 can communicate with another apparatus 1006 (such as a UE, a base station, or another wireless communication device) using the reception component 1002 and the transmission component 1004. As further shown, the apparatus 1000 can include a determination component 1008.

[0163] In some aspects, the apparatus 1000 can be configured to perform one or more of the functions described herein as being performed by a base station. In some aspects, the apparatus 1000 can be configured to perform one or more of the functions described herein as being performed by a UE.Figures 4-6 One or more operations described herein. Alternatively or concurrently, the apparatus 1000 may be configured to perform one or more processes described herein, such as... Figure 8 The process is 800. In some aspects, Figure 10 The device 1000 and / or one or more components shown may include the above-described components. Figure 2 One or more components of the described base station. Alternatively, Figure 10 One or more components shown can be combined with the above. Figure 2 The description is implemented within one or more components. Alternatively, one or more components in the set of components may be implemented, at least partially, as software stored in memory. For example, a component (or a portion of a component) may be implemented as instructions or code stored in a non-transitory computer-readable medium and executable by a controller or processor to perform the function or operation of the component.

[0164] Receiver 1002 may receive communications from device 1006, such as reference signals, control information, data communications, or combinations thereof. Receiver 1002 may provide the received communications to one or more other components of device 1000. In some aspects, receiver 1002 may perform signal processing on the received communications (e.g., filtering, amplification, demodulation, analog-to-digital conversion, demultiplexing, deinterleaving, demapping, equalization, interference cancellation, or decoding, and other examples), and may provide the processed signal to one or more other components of device 1006. In some aspects, receiver 1002 may include the elements described above. Figure 2 The described base station includes one or more antennas, demodulators, MIMO detectors, receiver processors, controllers / processors, memory, or combinations thereof.

[0165] Transmitting component 1004 can transmit communications, such as reference signals, control information, data communications, or combinations thereof, to device 1006. In some aspects, one or more other components of device 1006 can generate communications and provide the generated communications to transmitting component 1004 for transmission to device 1006. In some aspects, transmitting component 1004 can perform signal processing (e.g., filtering, amplification, modulation, digital-to-analog conversion, multiplexing, interleaving, mapping, or encoding, and other examples) on the generated communications and can transmit the processed signal to device 1006. In some aspects, transmitting component 1004 can include the combinations described above. Figure 2 The described base station includes one or more antennas, a modulator, a transmit MIMO processor, a transmit processor, a controller / processor, a memory, or a combination thereof. In some aspects, the transmit component 1004 may be co-located with the receive component 1002 in a transceiver.

[0166] The transmission component 1004 can transmit an indication of one or more of an input size or an input structure for a transmitting-side neural network, where the input structure indicates one or more of a non-zero value or a location of a non-zero value within an input of the input size to the transmitting-side neural network. The reception component 1002 and / or the transmission component 1004 can communicate with the UE based at least in part on the indication of one or more of the input size or the input structure of the transmitting-side neural network.

[0167] The determination component 1008 and / or the reception component 1002 can decode the PRB based at least in part on a receiving-side neural network associated with a transmitting-side neural network used by the UE to transmit the PRB.

[0168] The determination component 1008 and / or the transmission component 1004 can encode the PRB based at least in part on a transmitting-side neural network having an input size and / or an input structure as indicated to the UE.

[0169] The determination component 1008 and / or the reception component 1002 can decode the uplink control information based at least in part on a receiving-side neural network associated with a transmitting-side neural network used by the UE to transmit the uplink control information.

[0170] The determination component 1008 and / or the transmission component 1004 can encode the downlink control information based at least in part on a transmitting-side neural network having an input size and / or an input structure as indicated to the UE.

[0171] Figure 10 The number and arrangement of components shown is provided as an example. In practice, there can be additional components, fewer components, different components, or differently arranged components than those shown. Figure 10 than those illustrated, or arranged in a different manner than those illustrated. Furthermore, Figure 10 Two or more components illustrated can be implemented within a single component, or Figure 10 A single component illustrated can be implemented as multiple, distributed components. Additionally or alternatively, Figure 10 A set of one or more components illustrated can perform one or more functions described as being performed by another set of one or more components illustrated. Figure 10 one or more functions described as being performed by another set of components illustrated.

[0172] The following provides a summary of aspects of the disclosure:

[0173] Aspect 1: A method of wireless communication performed by a user equipment (UE), comprising: receiving an indication of one or more of an input size or an input structure for a transmit-side neural network, wherein the input structure indicates one or more of a non-zero value or a location of the non-zero value within an input of the input size to the transmit-side neural network; and communicating with a base station based at least in part on the indication of one or more of the input size or the input structure of the transmit-side neural network.

[0174] Aspect 2: The method of aspect 1, wherein the transmit-side neural network is associated with uplink communications, or wherein the transmit-side neural network is associated with downlink communications.

[0175] Aspect 3: The method of any of aspects 1 through 2, wherein receiving the indication of one or more of the input size or the input structure comprises: receiving the indication within a scheduling grant.

[0176] Aspect 4: The method of any of aspects 1 through 3, wherein receiving the indication of one or more of the input size or the input structure comprises: receiving a definition of one or more parameters of the input size or the input structure.

[0177] Aspect 5: The method of any of aspects 1 through 4, wherein the indication of one or more of the input size or the input structure comprises an indication of one or more receive-side neural networks associated with the transmit-side neural network.

[0178] Aspect 6: The method of any of aspects 1 through 5, wherein one or more of the input size or the input structure is associated with one or more segments of resource elements of a physical resource block, wherein one or more of an additional input size or an additional input structure is associated with one or more additional segments of resource elements of the physical resource block, or a combination thereof.

[0179] Aspect 7: The method of any of aspects 1 through 6, wherein the indication of one or more of the input size or the input structure comprises an indication of one or more of the additional input size or the additional input structure.

[0180] Aspect 8: The method of any of aspects 1 through 7, wherein a transmitting device of the UE or the base station transmits the physical resource block with power normalized across one or more segments of resource elements of a physical resource block after applying the transmit-side neural network.

[0181] Aspect 9: The method of any of aspects 1 through 6, further comprising: determining an uplink control channel format configuration based at least in part on the indication of one or more of the input size or the input structure.

[0182] Aspect 10: The method of aspect 9, wherein a first uplink control channel resource within a first set of uplink control channel resources is associated with one or more of the input size or the input structure, and wherein a second uplink control channel resource within the first set of uplink control channel resources is associated with one or more of an additional input size or an additional input structure.

[0183] Aspect 11: The method of any of aspects 1 through 6, 8, or 9, wherein receiving the indication of one or more of the input size or the input structure comprises: receiving an indication associated with an uplink control channel resource within a scheduling grant for a downlink shared channel transmission, wherein the communicating with the base station comprises: transmitting hybrid automatic repeat request feedback via the uplink control channel resource using the transmit-side neural network.

[0184] Aspect 12: The method of any of aspects 1 through 6, wherein the communicating with the base station comprises: receiving a transmission from the base station via a downlink control channel generated by the base station based at least in part on the transmit-side neural network, wherein one or more of the input size or the input structure is associated with: one or more resource element group bundling, one or more control channel elements, one or more aggregation levels, one or more search spaces, one or more control resource sets, or a combination thereof.

[0185] Aspect 13: The method of any of aspects 1 through 6, or 12, wherein a first control resource set associated with one or more of the input size or the input structure is associated with a first priority, wherein a second control resource set associated with one or more of an additional input size or an additional input structure is associated with a second priority, and wherein the UE is configured to prioritize monitoring of the first control resource set or the second control resource set based at least in part on one or more of: the UE operating in a power saving mode, a number of blind decoding attempts satisfying a threshold, or a number of channel estimation attempts satisfying a threshold.

[0186] Aspect 14: A method of wireless communication performed by a base station, comprising: transmitting an indication of one or more of an input size or an input structure for a transmitting side neural network, wherein the input structure indicates one or more of non-zero values or locations of the non-zero values within an input of the input size to the transmitting side neural network; and communicating with a user equipment (UE) based at least in part on the indication of the one or more of the input size or the input structure of the transmitting side neural network.

[0187] Aspect 15: The method of aspect 14, wherein the transmitting the indication of the one or more of the input size or the input structure comprises: transmitting the indication within a scheduling grant; or transmitting the indication within an uplink control channel format configuration.

[0188] Aspect 16: The method of aspect 15, wherein the one or more processors, when transmitting the indication of the one or more of the input size or the input structure, are configured to: transmit the indication within a scheduling grant; or transmit the indication within an uplink control channel format configuration.

[0189] Aspect 17: An apparatus for wireless communication at a device, comprising a processor; memory coupled with the processor; and instructions stored in the memory and executable by the processor to cause the apparatus to perform the method of one or more of Aspects 1-16.

[0190] Aspect 18: A device for wireless communication comprising a memory and one or more processors coupled to the memory, the memory and the one or more processors configured to perform the method of one or more of Aspects 1-16.

[0191] Aspect 19: An apparatus for wireless communication comprising at least one means for performing the method of one or more of Aspects 1-16.

[0192] Aspect 20: A non-transitory computer-readable medium storing code for wireless communication, the code comprising instructions executable by a processor to perform the method of one or more of Aspects 1-16.

[0193] Aspect 33: A non-transitory computer-readable medium storing a set of instructions for wireless communication, the set of instructions comprising one or more instructions that, when executed by one or more processors of a device, cause the device to perform the method of one or more of Aspects 1-16.

[0194] The foregoing disclosure provides illustration and description, but is not intended to be exhaustive or to limit the aspects to the precise form disclosed. Modifications and variations can be made in light of the above disclosure or can be acquired from practice of the aspects.

[0195] As used herein, the term “component” is intended to be broadly interpreted to encompass hardware and / or a combination of hardware and software. “Software” shall be broadly construed to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executables, threads of execution, procedures, and / or functions, among other examples, whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise. As used herein, a processor is implemented in hardware and / or a combination of hardware and software. It will be apparent that systems and / or methods described herein can be implemented in different forms of hardware and / or a combination of hardware and software. The actual specialized control hardware or software code used to implement these systems and / or methods is not limiting of the aspects. Thus, the operation and behavior of the systems and / or methods were described herein without reference to specific software code — it is understood that software and hardware can be designed to implement the systems and / or methods based, at least in part, on the description herein.

[0196] As used herein, depending on the context, satisfying a threshold can refer to a value being greater than the threshold, greater than or equal to the threshold, less than the threshold, less than or equal to the threshold, equal to the threshold, not equal to the threshold, and / or the like.

[0197] Even if a particular feature is expressly identified in the claims and / or the specification as having a particular combination of features, other combinations of the features are not excluded from the disclosure. In fact, many of the features can be combined in ways that are not specifically noted in the claims and / or the specification. While each dependent claim listed below can only directly depend on one claim, the disclosure of the aspects includes combinations of each dependent claim with every other claim in the set of claims. As used herein, a phrase referring to “at least one of’ a list of items refers to any combination of those items, including single members. As an example, “at least one of: a, b, or c” is intended to cover a, b, c, a-b, a-c, b-c, and a-b-c, as well as any combination of multiples of the same element (e.g., a-a, a-a-a, a-a-b, a-a-c, a-b-b, a-c-c, b-b, b-b-b, b-b-c, c-c, and c-c-c or any other ordering of a, b, and c).

[0198] No element, act, or instruction used herein should be construed as critical or essential unless explicitly described as such. Also, as used herein, the articles “a” and “an” are intended to include one or more items, and can be used interchangeably with “one or more.” Furthermore, as used herein, the article “the” is intended to include one or more items, and can be used interchangeably with “the one or more.” Also, as used herein, the terms “set” and “group” are intended to include one or more items (for example, related items, unrelated items, or a combination of related and unrelated items), and can be used interchangeably with “one or more.” Where only one item is intended, the phrase “only one” or similar language is used. Also, as used herein, the terms “has,” “have,” “having,” or the like are intended to be open-ended terms. Further, the phrase “based on” is intended to mean “based, at least in part, on” unless explicitly stated otherwise. Also, as used herein, the term “or” is intended to be inclusive when used in a series and can be used interchangeably with “and / or,” unless explicitly stated otherwise (e.g., if used in combination with “either” or “only one of”).

Claims

1. A method of wireless communication performed by a user equipment (UE), the method comprising: receiving an indication of an input structure for an input to a transmit-side neural network, wherein the input structure indicates one or more of a non-zero value within the input or a location for the non-zero value within the input; determining an uplink channel format configuration based at least in part on the indication of the input structure; and communicating with a base station based at least in part on the indication of the input structure.

2. The method of claim 1, wherein, the transmit-side neural network is associated with uplink communications, or wherein the transmit-side neural network is associated with downlink communications.

3. The method of claim 1, wherein, receiving the indication of the input structure comprises: receiving the indication within a scheduling grant.

4. The method of claim 1, wherein, receiving the indication of the input structure comprises: receiving a definition of one or more parameters of the input structure.

5. The method of claim 1, wherein, the indication of the input structure comprises: an indication of one or more receive-side neural networks associated with the transmit-side neural network.

6. The method of claim 1, wherein, the input structure is associated with one or more resource element segments of a physical resource block, wherein an additional input structure is associated with one or more additional resource element segments of the physical resource block, or a combination thereof.

7. The method of claim 6, wherein, the indication of the input structure comprises: an indication of the additional input structure.

8. The method of claim 1, wherein, a transmitting device of the UE or the base station transmits the physical resource block with power normalized over one or more resource element segments of a physical resource block after applying the transmit-side neural network.

9. The method of claim 1, wherein, a first uplink control channel resource within a first set of uplink control channel resources is associated with the input structure, and wherein a second uplink control channel resource within the first set of uplink control channel resources is associated with an additional input structure.

10. The method of claim 1, wherein, receiving the indication of the input structure comprises: receiving an indication associated with an uplink control channel resource within a scheduling grant for a downlink shared channel transmission, wherein the communicating with the base station comprises transmitting hybrid automatic repeat request feedback using the uplink control channel resource via the transmit-side neural network.

11. The method of claim 1, wherein, the communicating with the base station comprises: receiving a transmission from the base station via a downlink control channel generated by the base station based at least in part on the transmit-side neural network, wherein the input structure is associated with: one or more resource element group bundles, one or more control channel elements, one or more aggregation levels, one or more search spaces, one or more control resource sets, or a combination thereof.

12. The method of claim 11, wherein, a first control resource set associated with the input structure is associated with a first priority, wherein a second control resource set associated with an additional input structure is associated with a second priority, and wherein the UE is configured to prioritize monitoring of the first control resource set or the second control resource set based at least in part on one or more of: the UE operating in a power saving mode, a number of blind decoding attempts satisfies a threshold, or a number of channel estimation attempts satisfies another threshold.

13. A user equipment (UE) for wireless communication, the UE comprising: one or more memories; and one or more processors coupled to the one or more memories, the one or more memories storing instructions configurable to be executed by the one or more processors to cause the UE to: receive an indication of an input structure for an input to a transmitting-side neural network, wherein the input structure indicates one or more of a non-zero value within the input or a location for the non-zero value within the input; determine an uplink channel format configuration based at least in part on the indication of the input structure; and communicate with a base station based at least in part on the indication of the input structure.

14. The UE of claim 13, wherein, the transmitting-side neural network is associated with uplink communications, or wherein the transmitting-side neural network is associated with downlink communications.

15. The UE of claim 13, wherein, when receiving the indication of the input structure, the instructions are further executable by the one or more processors to cause the UE to: receive the indication within a scheduling grant.

16. The UE of claim 13, wherein, when receiving the indication of the input structure, the instructions are further executable by the one or more processors to cause the UE to: receive a definition of one or more parameters of the input structure.

17. The UE of claim 13, wherein, the indication of the input structure includes: an indication of one or more receiving-side neural networks associated with the transmitting-side neural network.

18. The UE of claim 13, wherein, the input structure is associated with one or more segments of resource elements of a physical resource block, wherein an additional input structure is associated with one or more additional segments of resource elements of the physical resource block, or a combination thereof.

19. The UE of claim 18, wherein, the indication of the input structure includes: an indication of the additional input structure.

20. The UE of claim 13, wherein, a transmitting device of the UE or the base station transmits the physical resource block with normalized power over one or more segments of resource elements of a physical resource block after applying the transmitting-side neural network.

21. The UE of claim 13, wherein, a first uplink control channel resource within a first set of uplink control channel resources is associated with the input structure, and wherein a second uplink control channel resource within the first set of uplink control channel resources is associated with an additional input structure.

22. The UE of claim 13, wherein, when receiving the indication of the input structure, the instructions are further executable by the one or more processors to cause the UE to: receive, within a scheduling grant for a downlink shared channel transmission, an indication associated with an uplink control channel resource, wherein the communicating with the base station includes transmitting hybrid automatic repeat request feedback using the uplink control channel resource via the transmitting-side neural network.

23. The UE of claim 13, wherein, when communicating with the base station, the instructions are further executable by the one or more processors to cause the UE to: receive, from the base station via a downlink control channel, a transmission generated by the base station based at least in part on the transmitting-side neural network, wherein the input structure is associated with: one or more resource element group bundles, one or more control channel elements, one or more aggregation levels, one or more search spaces, one or more control resource sets, or a combination thereof.

24. The UE of claim 23, wherein, a first control resource set associated with the input structure is associated with a first priority, wherein a second control resource set associated with an additional input structure is associated with a second priority, and wherein the UE is configured to prioritize monitoring of the first control resource set or the second control resource set based at least in part on one or more of: the UE operating in a power saving mode, a number of blind decoding attempts satisfies a threshold, or a number of channel estimation attempts satisfies another threshold.

25. A method of wireless communication performed by a base station, the method comprising: transmitting an indication of an input structure for an input to a transmitting side neural network, wherein the input structure indicates one or more of a non-zero value within the input or a location for the non-zero value within the input; and communicating with a user equipment (UE) based at least in part on the indication of the input structure, wherein an uplink channel format configuration is based at least in part on the indication of the input structure.

26. The method of claim 25, wherein, transmitting the indication of the input structure comprises: transmitting the indication within a scheduling grant, or transmitting the indication within the uplink channel format configuration.

27. The method of claim 25, wherein, the transmitting side neural network is associated with uplink communications, or wherein the transmitting side neural network is associated with downlink communications.

28. A base station for wireless communication, the base station comprising: one or more memories; and one or more processors coupled to the one or more memories, the one or more memories storing instructions configurable to cause the one or more processors to cause the base station to: transmit an indication of an input structure for an input to a transmitting side neural network, wherein the input structure indicates one or more of a non-zero value within the input or a location for the non-zero value within the input; and communicate with a user equipment (UE) based at least in part on the indication of the input structure, wherein an uplink channel format configuration is based at least in part on the indication of the input structure.

29. The base station of claim 28, wherein, when transmitting the indication of the input structure, the instructions are further executable by the one or more processors to cause the base station to: transmit the indication within a scheduling grant, or transmit the indication within the uplink channel format configuration.

30. The base station of claim 28, wherein, the transmitting side neural network is associated with uplink communications, or wherein the transmitting side neural network is associated with downlink communications.

Images