Frequency domain compression of channel state information

Abstract

In general, various aspects of the present disclosure relate to wireless communications. In some aspects, a user equipment (UE) may identify channel information based at least in part on a set of channel state information reference signals (CSI-RSs). The UE may use the model to generate compressed channel state information (CSI), where the input of the neural network model is based at least in part on the channel information, and where the compressed CSI is compressed in the frequency domain. The UE may transmit the compressed CSI. Numerous other aspects are described.

Description

Technical Field

[0001]

[0006] Generally speaking, aspects of the disclosure relate to wireless communications and aspects of the disclosure relate to techniques and apparatus for frequency domain compression of channel state information (CSI). Background Art

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

[0003] A wireless network may include one or more base stations that support communications for a user equipment (UE) or multiple UEs. A UE may communicate with a base station via downlink communications and uplink communications. A "downlink" (or "DL") refers to a communication link from a base station to a UE, and an "uplink" (or "UL") refers to a communication link from a UE to a base station.

[0004] The above multiple access technologies have been adopted in various telecommunication standards to provide a common protocol that enables different UEs to communicate at a city, country, region, and / or even global level. New Radio (NR), which may be referred to as 5G, is a set of enhancements to the LTE mobile standard released by 3GPP. NR is designed to better integrate with other open standards by improving spectrum efficiency, reducing costs, improving services, utilizing new spectrum, and using orthogonal frequency division multiplexing (OFDM) (CP-OFDM) with a cyclic prefix (CP) on the downlink, using CP-OFDM and / or single carrier frequency division multiplexing (SC-FDM) (also referred to as discrete Fourier transform spread OFDM (DFT-s-OFDM)) on the uplink, so as to better support mobile broadband Internet access, as well as support beamforming, multiple input multiple output (MIMO) antenna technology, and carrier aggregation. As the demand for mobile broadband access continues to grow, further improvements to LTE, NR, and other wireless access technologies remain useful. Summary of the invention

[0005] Some aspects described herein relate to a method of wireless communication performed by a user equipment (UE). The method may include identifying channel information based at least in part on a CSI-RS set. The method may include using a model to generate compressed CSI, wherein an input to the neural network model is based at least in part on the channel information, and wherein the compressed CSI is compressed in the frequency domain. The method may include sending the compressed CSI.

[0006] Some aspects described herein relate to a method of wireless communication performed by a network node. The method may include receiving compressed CSI compressed in the frequency domain. The method may include using a model to decompress the compressed CSI to obtain channel information. The method may include configuring communications based at least in part on the channel information.

[0007] Some aspects described herein relate to a UE for wireless communication. The UE may include a memory and one or more processors coupled to the memory. The one or more processors may be configured to identify channel information based at least in part on a CSI-RS set. The one or more processors may be configured to generate compressed CSI using a model, wherein an input to the neural network model is based at least in part on the channel information, and wherein the compressed CSI is compressed in the frequency domain. The one or more processors may be configured to send the compressed CSI.

[0008] Some aspects described herein relate to a network node for wireless communication. The network node may include a memory and one or more processors coupled to the memory. The one or more processors may be configured to receive compressed CSI compressed in the frequency domain. The one or more processors may be configured to use a model to decompress the compressed CSI to obtain channel information. The one or more processors may be configured to configure communication based at least in part on the channel information.

[0009] Some aspects described herein relate to a non-transitory computer-readable medium storing an instruction set for wireless communication by a UE. When executed by one or more processors of the UE, the instruction set may cause the UE to identify channel information based at least in part on a CSI-RS set. When executed by one or more processors of the UE, the instruction set may cause the UE to use a model to generate compressed CSI, wherein an input to the neural network model is based at least in part on the channel information, and wherein the compressed CSI is compressed in the frequency domain. When executed by one or more processors of the UE, the instruction set may cause the UE to send the compressed CSI.

[0010] Some aspects described herein relate to a non-transitory computer-readable medium storing an instruction set for wireless communication by a network node. When executed by one or more processors of the network node, the instruction set may cause the network node to receive compressed CSI compressed in the frequency domain. When executed by one or more processors of the network node, the instruction set may cause the network node to use a model to decompress the compressed CSI to obtain channel information. When executed by one or more processors of the network node, the instruction set may cause the network node to configure communication based at least in part on the channel information.

[0011] Some aspects described herein relate to an apparatus for wireless communication. The apparatus may include means for identifying channel information based at least in part on a CSI-RS set. The apparatus may include means for generating compressed CSI using a model, wherein an input to the neural network model is based at least in part on the channel information, and wherein the compressed CSI is compressed in the frequency domain. The apparatus may include means for sending the compressed CSI.

[0012] Some aspects described herein relate to an apparatus for wireless communication. The apparatus may include means for receiving compressed CSI compressed in the frequency domain. The apparatus may include means for decompressing the compressed CSI using a model to obtain channel information. The apparatus may include means for configuring communication based at least in part on the channel information.

[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 fully described herein with reference to and as illustrated by the accompanying drawings.

[0014] The foregoing has been fairly broadly outlined according to the features and technical advantages of the examples of the present disclosure so that the following specific embodiments may be better understood. Additional features and advantages will be described hereinafter. The disclosed concepts and specific examples may be easily used as the basis for modifying or designing other structures for the same purpose of performing the present disclosure. Such equivalent constructions do not depart from the scope of the appended claims. When considered in conjunction with the accompanying drawings, the features of the concepts disclosed herein (both their organization and methods of operation) and the associated advantages will be better understood from the following description. Each of the figures in the accompanying drawings is provided for the purpose of illustration and description, and is not intended to be a limitation to the claims.

[0015] Although various aspects are described in the present disclosure by the description of some examples, it will be understood by those skilled in the art that such aspects can be implemented in many different arrangements and scenarios. Different platform types, devices, systems, shapes, sizes and / or packaging arrangements can be used to implement the technology described herein. For example, some aspects can be implemented via integrated chip embodiments and other devices based on non-module components (e.g., end-user devices, vehicles, communication devices, computing devices, industrial equipment, retail / purchase equipment, medical devices and / or artificial intelligence devices). Various aspects can be implemented in chip-level components, modular components, non-modular components, non-chip-level components, device-level components and / or system-level components. The equipment incorporating the described aspects and features may include additional components and features for implementing and implementing the claimed and described aspects. For example, the transmission and reception of wireless signals may include one or more components for analog and digital purposes (e.g., hardware components including antennas, radio frequency (RF) chains, power amplifiers, modulators, buffers, processors, interleavers, adders and / or summers). The aspects described herein are intended to be implemented in a variety of devices, components, systems, distributed arrangements and / or end-user devices of varying sizes, shapes and configurations. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to understand the above features of the present disclosure in detail, a more specific description of the content briefly summarized above can be obtained by referring to various aspects (some of which are shown in the accompanying drawings). However, it should be noted that the accompanying drawings show only some typical aspects of the present disclosure, and because the description may allow other equally effective aspects, it is not considered to be a limitation on its scope. The same reference numerals in different drawings may identify the same or similar elements.

[0017] Figure 1 is a diagram illustrating an example of a wireless network according to the present disclosure.

[0018] Figure 2 is a diagram showing an example of a base station communicating with a user equipment (UE) in a wireless network according to the present disclosure.

[0019] Figure 3 is a diagram illustrating an example decomposed base station architecture according to the present disclosure.

[0020] Figure 4 is a diagram illustrating an example of a channel state information (CSI) reference signal (CSI-RS) beam management process according to the present disclosure.

[0021] Figure 5 is a diagram illustrating an example of end-to-end CSI compression according to the present disclosure.

[0022] Figure 6 is a diagram illustrating an example of hybrid CSI compression according to the present disclosure.

[0023] Figure 7 is a diagram illustrating an example of signaling associated with compressed CSI feedback according to the present disclosure.

[0024] Figure 8 is a diagram illustrating an example associated with an autoencoder for CSI compression according to the present disclosure.

[0025] Fig. 9 is a diagram illustrating an example process performed, for example, by a UE according to the present disclosure.

[0026] Fig.10 is a diagram illustrating an example process, eg, performed by a network node, in accordance with various aspects of the present disclosure.

[0027] Fig.11 is a diagram of an example apparatus for wireless communications in accordance with the present disclosure.

[0028] Fig.12 is a diagram of an example apparatus for wireless communications in accordance with the present disclosure. DETAILED DESCRIPTION

[0029] The various aspects of the present disclosure are described more fully below with reference to the accompanying drawings. However, the present disclosure can be embodied in many different forms, and should not be interpreted as being limited to any specific structure or function given throughout the present disclosure. Instead, these aspects are provided so that the present disclosure will be thorough and complete, and the scope of the present disclosure will be fully conveyed to those skilled in the art. It should be understood by those skilled in the art that the scope of the present disclosure is intended to cover any aspect of the present disclosure disclosed herein, whether the aspect is implemented independently of any other aspect of the present disclosure or implemented in combination with any other aspect of the present disclosure. For example, using any number of aspects set forth herein, a device can be implemented or a method can be implemented. In addition, the scope of the present disclosure is intended to cover such devices or methods that are practiced using other structures, functionality, or structures and functionality of the various aspects of the present disclosure set forth herein or structures, functionality, or structures and functionality of the various aspects of the present disclosure set forth herein. It should be understood that any aspect of the present disclosure disclosed herein can be embodied by one or more elements of the claims.

[0030] Several aspects of telecommunication systems will now be presented with reference to various devices and techniques. These devices and techniques will be described in the detailed description below and illustrated in the accompanying drawings by various blocks, modules, components, circuits, steps, processes, algorithms, etc. (collectively referred to as "elements"). These elements may be implemented using hardware, software, or a combination thereof. Whether such elements are implemented as hardware or software depends on the specific application and the design constraints imposed on the overall system.

[0031] Although various aspects may be described herein using terminology generally associated with 5G or new radio (NR) radio access technology (RAT), various aspects of the present disclosure may be applied to other RATs, such as 3G RAT, 4G RAT, and / or post-5G RATs (e.g., 6G).

[0032] Figure 1 1 is a diagram showing an example of a wireless network 100. The wireless network 100 may be or may include elements of a 5G (e.g., NR) network and / or a 4G (e.g., Long Term Evolution (LTE)) network, as well as other examples. The wireless network 100 may include one or more network entities, such as one or more network nodes 110 (shown as network node 110a, network node 110b, network node 110c, and network node 110d), a user equipment (UE) 120 or multiple UEs 120 (shown as UE 120a, UE 120b, UE 120c, UE 120d, and UE 120e), or other entities. The network node 110 is an example of a network node that communicates with the UE 120. As shown in the figure, the network node 110 may include one or more network entities. For example, the network node 110 may be an aggregated network node, which means that the aggregated network node is configured to utilize a radio protocol stack that is physically or logically integrated within a single RAN node (e.g., within a single device or unit). As another example, the network node 110 may be a decomposed network node (sometimes referred to as a decomposed base station), which means that the network node 110 includes two or more non-co-located network nodes. A decomposed network node may be configured to utilize a protocol stack that is physically or logically distributed between two or more nodes, such as one or more central units (CUs), one or more distributed units (DUs), or one or more radio units (RUs).

[0033] In some examples, the network node 110 includes an entity, such as an RU, that communicates with the UE 120 via a radio access link. In some examples, the network node 110 includes an entity, such as a DU, that communicates with other network nodes 110 via a forward link or a mid-haul link. In some examples, the network node 110 includes an entity, such as a CU, that communicates with other network nodes 110 via a mid-haul link or communicates with the core network via a backhaul link. In some examples, the network node 110 (such as an aggregated network node 110 or a decomposed network node 110) may include multiple network entities, such as one or more RUs, one or more CUs, or one or more DUs. The network node 110 may include, for example, an NR base station, an LTE base station, a Node B, an eNB (e.g., in 4G), a gNB (e.g., in 5G), an access point, or a transmit receive point (TRP), a DU, a RU, a CU, a mobility element of a network, a core network node, a network element, a network device, a RAN node, or a combination thereof. In some examples, network nodes 110 may be interconnected to each other or to one or more other network nodes 110 in wireless network 100 through various types of forward, mid-haul, or backhaul interfaces (such as direct physical connections, air interfaces, or virtual networks) using any appropriate transport network.

[0034] In some examples, network node 110 may provide communication coverage for a particular geographic area.In the 3rd Generation Partnership Project (3GPP), the term "cell" may refer to a coverage area of ​​a network node 110 or a network node subsystem serving that coverage area, depending on the context in which the term is used.

[0035] The network node 110 may provide communication coverage for a macro cell, a pico cell, a femto cell, or another type of cell. A macro cell may cover a relatively large geographic area (e.g., several kilometers in radius) and may allow unrestricted access by UEs 120 with service subscriptions. A pico cell may cover a relatively small geographic area and may allow unrestricted access by UEs 120 with service subscriptions. A femto cell may cover a relatively small geographic area (e.g., a residence) and may allow restricted access by UEs 120 associated with the femto cell (e.g., UEs 120 in a closed subscriber group (CSG)). A network node 110 for a macro cell may be referred to as a macro network node. A network node 110 for a pico cell may be referred to as a pico network node. A network node 110 for a femto cell may be referred to as a femto network node or an in-home network node. In Figure 1In the example shown, network node 110a may be a macro network node for macro cell 102a, network node 110b may be a pico network node for pico cell 102b, and network node 110c may be a femto network node for femto cell 102c. A network node may support one or more (e.g., three) cells. In some examples, a cell may not necessarily be stationary, and the geographic area of ​​a cell may move depending on the location of a mobile network node 110 (e.g., a mobile network node).

[0036] In some aspects, the term "base station" or "network node" may refer to an aggregated base station, a decomposed base station, an integrated access and backhaul (IAB) node, a relay node, or one or more components thereof. For example, in some aspects, a "base station" or "network node" may refer to a CU, a DU, a RU, a near real-time (near RT) RAN intelligent controller (RIC) or a non-real-time (non-RT) RIC, or a combination thereof. In some aspects, the term "base station" or "network node" may refer to a device configured to perform one or more functions (e.g., the functions described herein in conjunction with the network node 110). In some aspects, the term "base station" or "network node" may refer to multiple devices configured to perform one or more functions. For example, in some distributed systems, each of a plurality of different devices (which may be located in the same geographic location or in different geographic locations) may be configured to perform at least a portion of a function, or to repeatedly perform at least a portion of the function, and the term "base station" or "network node" may refer to any one or more of these different devices. In some aspects, the term "base station" or "network node" may refer to one or more virtual base stations or one or more virtual base station functions. For example, in some aspects, two or more base station functions may be instantiated on a single device. In some aspects, the term "base station" or "network node" may refer to one of the base station functions but not the other. Thus, a single device may include more than one base station.

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

[0038] The wireless network 100 may be a heterogeneous network that includes different types of network nodes 110, such as macro network entities, pico network entities, femto network entities, or relay network entities. These different types of network nodes 110 may have different transmit power levels, different coverage areas, or different impacts on interference in the wireless network 100. For example, a macro network entity may have a high transmit power level (e.g., 5 to 40 watts), while a pico network entity, a femto network entity, and a relay network entity may have a lower transmit power level (e.g., 0.1 to 2 watts).

[0039] The network controller 130 may be coupled to or communicate with a set of network nodes 110 and may provide coordination and control for these network nodes 110. The network controller 130 may communicate with the network nodes 110 via a backhaul communication link or a mid-range communication link. The network nodes 110 may communicate with each other directly or indirectly via a wireless or wired backhaul communication link. In some aspects, the network controller 130 may be a CU, or may include a CU or a core network device.

[0040] UE 120 can be distributed throughout the wireless network 100, and each UE 120 can be stationary or mobile. UE 120 can include, for example, an access terminal, a terminal, a mobile station, or a user unit. UE 120 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 device, a camera, a gaming device, a netbook, a smartbook, an ultrabook, a medical device, a biometric device, a wearable device (e.g., a smart watch, smart clothing, smart glasses, a smart wristband, smart jewelry (e.g., a smart ring or a smart bracelet)), an entertainment device (e.g., a music device, a video device, or a satellite radio unit), a vehicle component or sensor, a smart meter / sensor, an industrial manufacturing device, a global positioning system device, a UE function of a network node, or any other suitable device configured to communicate via a wireless or wired medium.

[0041] Some UEs 120 may be considered as machine type communication (MTC) or evolved or enhanced machine type communication (eMTC) UEs. MTC UEs or eMTC UEs may include, for example, robots, drones, remote devices, sensors, meters, monitors, or location tags, which may communicate with a base station, another device (e.g., a remote device), or some other entity. Some UEs 120 may be considered as Internet of Things (IoT) devices, or may be implemented as NB-IoT (narrowband IoT) devices. Some UEs 120 may be considered as user premises equipment. UE 120 may be included inside a housing 284 housing components of UE 120 (such as processor components or memory components). In some examples, the processor component and the memory component may be coupled together. For example, a processor component (e.g., one or more processors) and a memory component (e.g., a memory) may be operationally coupled, communicatively coupled, electronically coupled, or electrically coupled.

[0042] In general, any number of wireless networks 100 may be deployed in a given geographic area. Each wireless network 100 may support a specific RAT and may operate on one or more frequencies. RAT may be referred to as a radio technology or air interface. Frequency may be referred to as a carrier or channel. Each frequency may support a single RAT in a given geographic area to avoid interference between wireless networks of different RATs. In some cases, NR or 5G RAT networks may be deployed.

[0043] In some examples, two or more UEs 120 (e.g., shown as UE 120a and UE 120e) may communicate directly using one or more sidelink channels (e.g., without using network node 110 as an intermediary for communicating with each other). For example, UE 120 may communicate using peer-to-peer (P2P) communication, device-to-device (D2D) communication, vehicle-to-everything (V2X) protocol (e.g., which may include vehicle-to-vehicle (V2V) protocol, vehicle-to-infrastructure (V2I) protocol, or vehicle-to-pedestrian (V2P) protocol), or mesh network. In such examples, UE 120 may perform scheduling operations, resource selection operations, and / or other operations described elsewhere herein as being performed by network node 110.

[0044] Devices of the wireless network 100 may communicate using an electromagnetic spectrum, which may be subdivided into various categories, bands, or channels by frequency or wavelength. For example, devices of the wireless network 100 may communicate using one or more operating bands. In 5G NR, two initial operating bands have been identified as frequency range names FR1 (410 MHz-7.125 GHz) and FR2 (24.25 GHz-52.6 GHz). Although a portion of FR1 is greater than 6 GHz, FR1 is often (interchangeably) referred to as a "sub-6 GHz" band in various documents and articles. A similar naming issue sometimes occurs with respect to FR2, which is often (interchangeably) referred to as a "millimeter wave" band in documents and articles, although it is different from the extremely high frequency (EHF) band (30 GHz-300 GHz) identified as a "millimeter wave" band by the International Telecommunication Union (ITU).

[0045] Frequencies between FR1 and FR2 are generally referred to as mid-band frequencies. Recent 5G NR research has identified the operating bands of these mid-band frequencies as the frequency range name FR3 (7.125GHz-24.25GHz). The frequency bands falling within FR3 can inherit the FR1 characteristics or the FR2 characteristics, and therefore the features of FR1 or FR2 can be effectively extended to the mid-band frequencies. In addition, higher frequency bands are currently being explored to extend 5G NR operations to above 52.6GHz. For example, three higher operating bands have been identified as the frequency range names FR4a or FR4-1 (52.6GHz-71 GHz), FR4 (52.6GHz-114.25GHz) and FR5 (114.25GHz-300GHz). Each of these higher frequency bands falls within the EHF band.

[0046] With these examples in mind, the term "sub-6 GHz" (if used herein) may broadly refer to frequencies that may be less than 6 GHz, may be within FR1, or may include mid-band frequencies, unless otherwise specifically stated. Additionally, the term "millimeter wave" (if used herein) may broadly refer to frequencies that may include mid-band frequencies, may be within FR2, FR4, FR4a, FR4-1, or FR5, or may be within the EHF band, unless otherwise specifically stated. It is contemplated that the frequencies included in these operating bands (e.g., FR1, FR2, FR3, FR4, FR4-a, FR4-1, or FR5) may be modified, and the techniques described herein are applicable to those modified frequency ranges.

[0047] In some aspects, UE 120 may include a communication manager 140. As described in greater detail elsewhere herein, the communication manager 140 may identify channel information based at least in part on a set of channel state information reference signals (CSI-RS); generate compressed channel state information (CSI) using a model, wherein an input to the neural network model is based at least in part on the channel information, and wherein the compressed CSI is compressed in the frequency domain; and send the compressed CSI. Additionally or alternatively, the communication manager 140 may perform one or more other operations described herein.

[0048] In some aspects, the network node 110 may include a communication manager 150. As described in more detail elsewhere herein, the communication manager 150 may receive compressed CSI compressed in the frequency domain; decompress the compressed CSI using a model to obtain channel information; and configure communications based at least in part on the channel information. Additionally or alternatively, the communication manager 150 may perform one or more other operations described herein.

[0049] Figure 2 2 is a diagram illustrating an example 200 in which a network node 110 communicates with a UE 120 in a wireless network 100. The network node 110 may be equipped with a set of antennas 234a to 234t, such as T antennas (T≥1). The UE 120 may be equipped with a set of antennas 252a to 252r, such as R antennas (R≥1). The network node 110 of example 200 includes one or more radio frequency components, such as an antenna 234 and a modem 254. In some examples, the network node 110 may include an interface, a communication component, or another component that facilitates communication with the UE 120 or another network node. Some network nodes 110 may not include a radio frequency component that facilitates direct communication with the UE 120, such as one or more CUs or one or more DUs.

[0050] At the network node 110, a transmit processor 220 may receive data intended for a UE 120 (or a set of UEs 120) from a data source 212. The transmit processor 220 may select one or more modulation and coding schemes (MCS) for the UE 120 using one or more channel quality indicators (CQIs) received from the UE 120. The network node 110 may process (e.g., encode and modulate) the data for the UE 120 using the MCS selected for the UE 120, and may provide data symbols for the UE 120. The transmit processor 220 may process system information (e.g., for semi-static resource partitioning information (SRPI)) and control information (e.g., CQI requests, grants, or upper layer signaling), and provide overhead symbols and control symbols. The transmit processor 220 may generate reference symbols for reference signals (e.g., cell-specific reference signals (CRS) or demodulation reference signals (DMRS)) and synchronization signals (e.g., primary synchronization signals (PSS) or secondary synchronization signals (SSS)). The transmit (TX) multiple-input multiple-output (MIMO) processor 230 may perform spatial processing (e.g., precoding) on ​​data symbols, control symbols, overhead symbols, or reference symbols, if applicable, and may provide a set of output symbol streams (e.g., T output symbol streams) to a corresponding set of modems 232 (e.g., T modems) (shown as modems 232a to 232t). For example, each output symbol stream may be provided to a modulator component (shown as MOD) of the modem 232. Each modem 232 may process a corresponding output symbol stream (e.g., for OFDM) using a corresponding modulator component to obtain an output sample stream. Each modem 232 may also process (e.g., convert to analog, amplify, filter, or upconvert) the output sample stream using a corresponding modulator component to obtain a downlink signal. The modems 232a to 232t may transmit a set of downlink signals (e.g., T downlink signals) via a corresponding set of antennas 234 (e.g., T antennas) (shown as antennas 234a to 234t).

[0051] At the UE 120, a set of antennas 252 (shown as antennas 252a to 252r) may receive downlink signals from the network node 110 or other network nodes 110, and may provide a set of received signals (e.g., R received signals) to a set of modems 254 (e.g., R modems) (shown as modems 254a to 254r). For example, each received signal may be provided to a demodulator component (shown as DEMOD) of the modem 254. Each modem 254 may use a corresponding demodulator component to condition (e.g., filter, amplify, downconvert, or digitize) the received signal to obtain input samples. Each modem 254 may use a demodulator component to further process the input samples (e.g., for OFDM) to obtain received symbols. A MIMO detector 256 may obtain received symbols from the modulator 254, perform MIMO detection on the received symbols (if applicable), and may provide detected symbols. The receive processor 258 may process (e.g., demodulate and decode) the detected symbols, may provide decoded data for the UE 120 to the data sink 260, and may provide decoded control information and system information to the controller / processor 280. The term "controller / processor" may refer to one or more controllers, one or more processors, or a combination thereof. The channel processor may determine a reference signal received power (RSRP) parameter, a received signal strength indicator (RSSI) parameter, a reference signal received quality (RSRQ) parameter, or a CQI parameter, among other examples. In some examples, one or more components of the UE 120 may be included in a housing.

[0052] The network controller 130 may include a communication unit 294, a controller / processor 290, and a memory 292. The network controller 130 may include, for example, one or more devices in a core network. The network controller 130 may communicate with the network node 110 via the communication unit 294.

[0053] One or more antennas (e.g., antennas 234a to 234t or antennas 252a to 252r) may include or may be included in one or more antenna panels, one or more antenna groups, one or more antenna element sets, or one or more antenna arrays, among other examples. An antenna panel, antenna group, antenna element set, or antenna array may include one or more antenna elements (in a single housing or multiple housings), a coplanar antenna element set, a non-coplanar antenna element set, or coupled to one or more transmit or receive components (such as Figure 2 One or more antenna elements of one or more components).

[0054] On the uplink, at the UE 120, a transmit processor 264 may receive and process data from a data source 262 and control information from a controller / processor 280 (e.g., for reports including RSRP, RSSI, RSRQ, or CQI). The transmit processor 264 may also generate reference symbols for one or more reference signals. The symbols from the transmit processor 264 may be precoded by a TX MIMO processor 266 (if applicable), further processed by a modem 254 (e.g., for DFT-s-OFDM or CP-OFDM), and sent to the network node 110. In some examples, the modem 254 of the UE 120 may include a modulator and a demodulator. In some examples, the UE 120 includes a transceiver. The transceiver may include any combination of an antenna 252, a modem 254, a MIMO detector 256, a receive processor 258, a transmit processor 264, or a TX MIMO processor 266. The transceiver may be used by a processor (eg, controller / processor 280) and memory 282 to perform aspects of any of the processes described herein.

[0055] At the network node 110, uplink signals from the UE 120 or other UEs may be received by the antenna 234, processed by the modem 232 (e.g., a demodulator component of the modem 232, shown as DEMOD), detected by the MIMO detector 236 (if applicable), and further processed by the receive processor 238 to obtain decoded data and control information sent by the UE 120. The receive processor 238 may provide the decoded data to the data sink 239 and the decoded control information to the controller / processor 240. The network node 110 may include a communication unit 244 and may communicate with the network controller 130 via the communication unit 244. The network node 110 may include a scheduler 246 to schedule one or more UEs 120 for downlink or uplink communication. In some examples, the modem 232 of the network node 110 may include a modulator and a demodulator. In some examples, the network node 110 includes a transceiver. The transceiver may include any combination of antenna 234, modem 232, MIMO detector 236, receive processor 238, transmit processor 220, or TX MIMO processor 230. The transceiver may be used by a processor (e.g., controller / processor 240) and memory 242 to perform aspects of any of the processes described herein.

[0056] In some aspects, the controller / processor 280 may be a component of a processing system. A processing system may generally be a system or series of machines or components that receives inputs and processes the inputs to produce a set of outputs that may be communicated to other systems or components, such as the UE 120. For example, the processing system of the UE 120 may be a system that includes various other components or subcomponents of the UE 120.

[0057] The processing system of UE 120 may interface with one or more other components of UE 120, may process information (such as input or signal) received from one or more other components, or may output information to one or more other components. For example, a chip or modem of UE 120 may include a processing system, a first interface for receiving or obtaining information, and a second interface for outputting, sending or providing information. In some examples, the first interface may be an interface between a processing system of a chip or modem and a receiver, so that UE 120 may receive information or signal input, and may pass information to the processing system. In some examples, the second interface may be an interface between a processing system of a chip or modem and a transmitter, so that UE 120 may send information output from a chip or modem. One of ordinary skill in the art will readily recognize that the second interface may also obtain or receive information or signal input, and the first interface may also output, send or provide information.

[0058] In some aspects, controller / processor 240 may be a component of a processing system. A processing system may generally be a system or series of machines or components that receives inputs and processes the inputs to produce a set of outputs that may be passed to other systems or components, such as network node 110. For example, a processing system of network node 110 may be a system that includes various other components or subcomponents of network node 110.

[0059] The processing system of the network node 110 may interface with one or more other components of the network node 110, may process information (such as input or signal) received from one or more other components, or may output information to one or more other components. For example, a chip or modem of the network node 110 may include a processing system, a first interface for receiving or obtaining information, and a second interface for outputting, sending or providing information. In some examples, the first interface may be an interface between the processing system of the chip or modem and a receiver, so that the network node 110 may receive information or signal input, and may pass information to the processing system. In some examples, the second interface may be an interface between the processing system of the chip or modem and a transmitter, so that the network node 110 may send information output from the chip or modem. Those of ordinary skill in the art will readily recognize that the second interface may also obtain or receive information or signal input, and the first interface may also output, send or provide information.

[0060] The controller / processor 240 of the network node 110, the controller / processor 280 of the UE 120, and / or Figure 2 Any other component in the network node 110 may perform one or more techniques associated with CSI reporting, as described in greater detail elsewhere herein. For example, the controller / processor 240 of the network node 110, the controller / processor 280 of the UE 120, and / or Figure 2 Any other component (or combination of components) in may perform or direct, for example, Fig. 9 The process of 900 Fig.10 1000, or operations of other processes as described herein. Memory 242 and memory 282 may store data and program codes for network node 110 and UE 120, respectively. In some examples, memory 242 and memory 282 may include non-transitory computer-readable media storing one or more instructions (e.g., code or program code) for wireless communication. For example, one or more instructions, when executed by one or more processors of network node 110 or UE 120 (e.g., directly, or after compilation, conversion, or interpretation), may cause one or more processors, UE 120, or network node 110 to perform or direct, for example, Fig. 9 The process of 900 Fig.10 The process 1000 of , or other processes described herein, may be performed. In some examples, executing instructions may include running instructions, converting instructions, compiling instructions, or interpreting instructions.

[0061] In some aspects, the UE 120 includes means for identifying channel information based at least in part on a CSI-RS set; means for generating compressed CSI using a model, wherein an input to the neural network model is based at least in part on the channel information, and wherein the compressed CSI is compressed in the frequency domain; and / or means for transmitting the compressed CSI. The means for the UE 120 to perform the operations described herein may include, for example, one or more of the communication manager 140, the antenna 252, the modem 254, the MIMO detector 256, the receive processor 258, the transmit processor 264, the TX MIMO processor 266, the controller / processor 280, or the memory 282.

[0062] In some aspects, the network node includes means for receiving compressed CSI compressed in the frequency domain; means for decompressing the compressed CSI using a model to obtain channel information; and / or means for configuring communications based at least in part on the channel information. In some aspects, the means for the network node to perform operations described herein may include, for example, one or more of the communication manager 150, the transmit processor 220, the TX MIMO processor 230, the modem 232, the antenna 234, the MIMO detector 236, the receive processor 238, the controller / processor 240, the memory 242, or the scheduler 246.

[0063] Although Figure 2 The blocks in the 200 and 210 are shown as distinct components, but the functionality described with respect to the blocks may be implemented in a single hardware, software, or combined component or in various combinations of components. For example, the functionality described with respect to the transmit processor 264, the receive processor 258, the TX MIMO processor 266, or another processor may be performed by or under the control of the controller / processor 280.

[0064] The deployment of a communication system (such as a 5G NR system) can be arranged with various components or components in a variety of ways. In a 5G NR system or network, a network node, a network node, a mobility element of a network, a RAN node, a core network node, a network element, a base station or a network device can be implemented in an aggregated or decomposed architecture. For example, a base station (such as a Node B (NB), an evolved NB (eNB), an NR BS, a 5G NB, an access point (AP), a TRP or a cell, etc.) or one or more units (or one or more components) that perform base station functionality can be implemented as an aggregated base station (also referred to as an independent base station or a single-chip base station) or a decomposed base station.

[0065] An aggregated base station may be configured to utilize a wireless protocol stack that is physically or logically integrated within a single RAN node (e.g., within a single device or unit). A decomposed base station may be configured to utilize a protocol stack that is physically or logically distributed between two or more units (such as a CU, one or more DUs, or one or more RUs). In some examples, a CU may be implemented within a RAN node, and one or more DUs may be co-located with the CU, or alternatively, may be geographically or virtually distributed in one or more other RAN nodes. A DU may be implemented to communicate with one or more RUs. Each of a CU, a DU, and a RU may also be implemented as a virtual unit, such as a virtual central unit (VCU), a virtual distributed unit (VDU), or a virtual radio unit (VRU).

[0066] Base station type operations or network designs may take into account the aggregated nature of base station functionality. For example, a decomposed base station may be utilized in an IAB network, an open radio access network (O-RAN (such as a network configuration initiated by the O-RAN Alliance)), or a virtualized radio access network (vRAN, also referred to as a cloud radio access network (C-RAN)) to facilitate scaling of a communication system by separating base station functionality into one or more units that may be deployed separately. A decomposed base station may include functionality implemented across two or more units at various physical locations, as well as functionality implemented virtually for at least one unit, which may enable flexibility in network design. Individual units of a decomposed base station may be configured for wired or wireless communication with at least one other unit of the decomposed base station.

[0067] Figure 3 3 is a diagram illustrating an example decomposed base station architecture 300 according to the present disclosure. The decomposed base station architecture 300 may include a CU 310, which may communicate directly with a core network 320 via a backhaul link, or indirectly with the core network 320 through one or more decomposed control units (e.g., a near RT RIC 325 via an E2 link, or a non-RT RIC 315 associated with a service management and orchestration (SMO) framework 305, or both). The CU 310 may communicate with one or more DUs 330 via corresponding mid-haul links (e.g., via an F1 interface). Each of the DUs 330 may communicate with one or more RUs 340 via corresponding fronthaul links. Each of the RUs 340 may communicate with one or more UEs 120 via corresponding radio frequency (RF) access links. In some implementations, a UE 120 may be served by multiple RUs 340 simultaneously.

[0068] Each of the units (including CU 310, DU 330, RU 340, near-RT RIC 325, non-RT RIC 315, and SMO framework 305) may include one or more interfaces, or be coupled to one or more interfaces, and the one or more interfaces are configured to receive or send signals, data, or information (collectively referred to as signals) via a wired or wireless transmission medium. Each of the units or an associated processor or controller that provides instructions to one or more communication interfaces of the corresponding unit may be configured to communicate with one or more of the other units via a transmission medium. In some examples, each of the units may include a wired interface and a wireless interface, the wired interface is configured to receive or send signals to one or more of the other units through a wired transmission medium, and the wireless interface may include a receiver, a transmitter, or a transceiver (e.g., an RF transceiver), which is configured to receive signals, or send signals, or both, to one or more of the other units through a wireless transmission medium.

[0069] In some aspects, the CU 310 may host one or more higher-layer control functions. Such control functions may include radio resource control (RRC) functions, packet data convergence protocol (PDCP) functions, or service data adaptation protocol (SDAP) functions, etc. Each control function may be implemented using an interface configured to transmit signals with other control functions hosted by the CU 310. The CU 310 may be configured to process user plane functionality (e.g., central unit-user plane (CU-UP) functionality), control plane functionality (e.g., central unit-control plane (CU-CP) functionality), or a combination thereof. In some implementations, the CU 310 may be logically split into one or more CU-UP units and one or more CU-CP units. When implemented in an O-RAN configuration, the CU-UP unit may communicate bidirectionally with the CU-CP unit via an interface (such as an E1 interface). The CU 310 may be implemented to communicate with the DU 330 when necessary for network control and signaling.

[0070] Each DU 330 may correspond to a logical unit that includes one or more base station functions to control the operation of one or more RUs 340. In some aspects, the DU 330 may host one or more of a radio link control (RLC) layer, a MAC layer, and one or more high physical (PHY) layers, at least in part according to a functional partition (e.g., a functional partition defined by 3GPP). In some aspects, one or more high PHY layers may be implemented by one or more modules for forward error correction (FEC) encoding and decoding, scrambling, modulation and demodulation, etc. In some aspects, the DU 330 may further host one or more low PHY layers (e.g., implemented by one or more modules for fast Fourier transform (FFT), inverse FFT (iFFT), digital beamforming, or physical random access channel (PRACH) extraction and filtering, etc.). Each layer (which may also be referred to as a module) may be implemented using an interface configured to transmit signals with other layers (and modules) hosted by the DU 330 or with a control function hosted by the CU 310.

[0071] Each RU 340 may implement lower layer functionality. In some deployments, the RU 340 controlled by the DU 330 may correspond to a logical node that hosts RF processing functions or low PHY layer functions (e.g., performing FFT, performing iFFT, digital beamforming, or PRACH extraction and filtering, etc.) based on functional partitioning (e.g., functional partitioning defined by 3GPP) (e.g., lower layer functional partitioning). In such an architecture, each RU 340 may be operated to handle over-the-air (OTA) communications with one or more UEs 120. In some implementations, real-time and non-real-time aspects of control and user plane communications with the RU 340 may be controlled by the corresponding DU 330. In some scenarios, this configuration may enable each DU 330 and CU 310 to be implemented in a cloud-based RAN architecture (e.g., a vRAN architecture).

[0072] The SMO framework 305 may be configured to support RAN deployment and provisioning of non-virtualized and virtualized network elements. For non-virtualized network elements, the SMO framework 305 may be configured to support the deployment of dedicated physical resources for RAN coverage requirements, which may be managed via an operation and maintenance interface, such as an O1 interface. For virtualized network elements, the SMO framework 305 may be configured to interact with a cloud computing platform, such as an open cloud (O-cloud) platform 390, to perform network element lifecycle management (such as instantiating virtualized network elements) via a cloud computing platform interface, such as an O2 interface. Such virtualized network elements may include, but are not limited to, CU 310, DU 330, RU 340, non-RT RIC 315, and near-RT RIC 325. In some implementations, the SMO framework 305 may communicate with hardware aspects of the 4G RAN, such as an open eNB (O-eNB) 311, via an O1 interface. Additionally, in some implementations, the SMO framework 305 can communicate directly with each of the one or more RUs 340 via a corresponding O1 interface. The SMO framework 305 can also include a non-RT RIC 315 configured to support the functionality of the SMO framework 305.

[0073] The non-RT RIC 315 may be configured to include logic functions that implement non-real-time control and optimization of RAN elements and resources, artificial intelligence / machine learning (AI / ML) workflows including model training and updating, or policy-based guidance of applications / features in the near-RT RIC 325. The non-RT RIC 315 may be coupled to or in communication with the near-RT RIC 325 (e.g., via an A1 interface). The near-RT RIC 325 may be configured to include logic functions that implement near-real-time control and optimization of RAN elements and resources via data collection and actions over an interface (e.g., via an E2 interface) that connects one or more CUs 310, one or more DUs 330, or both, and the O-eNB with the near-RT RIC 325.

[0074] In some implementations, in order to generate an AI / ML model to be deployed in the near-RT RIC 325, the non-RT RIC 315 may receive parameters or external enrichment information from an external server. Such information may be utilized by the near-RT RIC 325 and may be received at the SMO framework 305 or the non-RT RIC 315 from a non-network data source or from a network function. In some examples, the non-RT RIC 315 or the near-RT RIC 325 may be configured to tune RAN behavior or performance. For example, the non-RT RIC 315 may monitor long-term trends and patterns of performance and employ AI / ML models to perform corrective actions through the SMO framework 305 (such as via reconfiguration of the O1 interface) or via the creation of RAN management policies (such as A1 interface policies).

[0075] As indicated above, Figure 3 are provided as examples. Other examples may be relevant to Figure 3 The examples described are different.

[0076] Figure 4 4 is a diagram illustrating examples 400, 410, and 420 of a CSI-RS beam management process according to the present disclosure. Figure 4 As shown in , examples 400, 410, and 420 include UE 120 communicating with network node 110 in a wireless network (e.g., wireless network 100). However, Figure 4 The devices shown in are provided as examples, and the wireless network may support communication and beam management between other devices (e.g., between UE 120 and network node 110 or a transmit receive point (TRP), between a mobile terminal node and a control node, between an integrated access and backhaul (IAB) child node and an IAB parent node, and / or between a scheduled node and a scheduling node). In some aspects, UE 120 and network node 110 may be in a connected state (e.g., an RRC connected state).

[0077] like Figure 4 As shown, example 400 may include network node 110 and UE 120 communicating to perform beam management using CSI-RS. Example 400 depicts a first beam management procedure (e.g., P1 CSI-RS beam management). The first beam management procedure may be referred to as a beam selection procedure, an initial beam acquisition procedure, a beam scanning procedure, a cell search procedure, and / or a beam search procedure. Figure 4As shown in example 400, CSI-RS may be configured to be sent from network node 110 to UE 120. CSI-RS may be configured to be periodic (e.g., using RRC signaling), semi-persistent (e.g., using medium access control (MAC) control element (MAC-CE) signaling), and / or aperiodic (e.g., using downlink control information (DCI)).

[0078] The first beam management process may include the network node 110 performing beam scanning on multiple transmit (Tx) beams. The network node 110 may use each transmit beam for beam management to send CSI-RS. In order to enable the UE 120 to perform receive (Rx) beam scanning, the base station may use the transmit beam to send each CSI-RS multiple times (e.g., using repetition) within the same RS resource set, so that the UE 120 can sweep the receive beam in multiple transmission instances. For example, if the network node 110 has a set of N transmit beams and the UE 120 has a set of M receive beams, the CSI-RS may be sent M times on each of the N transmit beams, so that the UE 120 can receive M instances of the CSI-RS for each transmit beam. In other words, for each transmit beam of the network node 110, the UE 120 may perform beam scanning through the receive beam of the UE 120. Thus, the first beam management procedure may enable UE 120 to measure CSI-RS on different transmit beams using different receive beams to support selection of network node 110 transmit beam / UE 120 receive beam pairs. UE 120 may report the measurements to network node 110 to enable network node 110 to select one or more beam pairs for communication between network node 110 and UE 120. Although example 400 has been described in conjunction with CSI-RS, the first beam management procedure may also use synchronization signal blocks for beam management in a manner similar to that described above.

[0079] like Figure 4 As shown, example 410 may include network node 110 and UE 120 communicating to perform beam management using CSI-RS. Example 410 depicts a second beam management procedure (e.g., P2 CSI-RS beam management). The second beam management procedure may be referred to as a beam refinement procedure, a base station beam refinement procedure, a TRP beam refinement procedure, and / or a transmit beam refinement procedure. Figure 4As shown in example 410, the CSI-RS may be configured to be sent from the network node 110 to the UE 120. The CSI-RS may be configured to be periodic (e.g., using DCI). The second beam management process may include the network node 110 performing beam scanning on one or more transmit beams. The one or more transmit beams may be a subset of all transmit beams associated with the network node 110 (e.g., determined at least in part based on measurements reported by the UE 120 in conjunction with the first beam management process). The network node 110 may use each of the one or more transmit beams for beam management to send the CSI-RS. The UE 120 may measure each CSI-RS using a single (e.g., the same) receive beam (e.g., determined at least in part based on measurements performed in conjunction with the first beam management process). The second beam management process may enable the network node 110 to select the best transmit beam based at least in part on measurements of the CSI-RS reported by the UE 120 (e.g., measured by the UE 120 using a single receive beam).

[0080] like Figure 4 As shown, example 420 depicts a third beam management process (e.g., P3 CSI-RS beam management). The third beam management process may be referred to as a beam refinement process, a UE beam refinement process, and / or a receive beam refinement process. Figure 4 As shown in example 420, one or more CSI-RS may be configured to be sent from the network node 110 to the UE 120. The CSI-RS may be configured to be periodic (e.g., using DCI). The third beam management process may include the network node 110 sending one or more CSI-RS using a single transmit beam (e.g., determined at least in part based on measurements reported by the UE 120 in conjunction with the first beam management process and / or the second beam management process). In order to enable the UE 120 to perform receive beam scanning, the base station may use the transmit beam to send (e.g., using repetition) the CSI-RS multiple times within the same RS resource set, so that the UE 120 can sweep across one or more receive beams in multiple transmission instances. The one or more receive beams may be a subset of all receive beams associated with the UE 120 (e.g., determined at least in part based on measurements performed in conjunction with the first beam management process and / or the second beam management process). The third beam management procedure may enable the network node 110 and / or the UE 120 to select the best receive beam based at least in part on reported measurements received from the UE 120 (eg, measurements of the CSI-RS of the transmit beam using one or more receive beams).

[0081] In some examples described herein, reporting of information about CSI-RS may involve compression and decompression of the CSI (eg, using an autoencoder).

[0082] As indicated above, Figure 4 is provided as an example of a beam management process. Other examples of beam management processes may differ from those described above. Figure 4 For example, UE 120 and network node 110 may perform the third beam management procedure before performing the second beam management procedure, and / or UE 120 and network node 110 may perform a similar beam management procedure to select a UE transmit beam.

[0083] UE 120 may measure CSI-RS and send a CSI report indicating CSI determined at least in part based on CSI-RS. CSI provides information about the channel so that network node 110 can manage the connection between network node 110 and UE 120. In general, a channel may be represented by a channel matrix (generally represented by variable H). For example, a channel matrix may represent a channel in the time domain (generally referred to as a channel impulse response) or a channel in the frequency domain (generally referred to as a channel frequency response) or a combination thereof. Further representations may also be used, such as in the delay-Doppler domain. In the present disclosure, "channel information" (in contrast to "channel state information") refers to an uncompressed (e.g., original or unprocessed) characterization or representation or state of a channel, such as the representation by a channel matrix mentioned above. Channel information may also refer to information derived from the original channel without dimensionality loss / dimensionality reduction. Therefore, the channel information may include sufficient information to allow the channel matrix to be derived from the channel information without loss of accuracy. In this context, note that the channel matrix itself may not be completely known, but only estimated. Furthermore, the rank of the channel matrix may not be full (eg, due to correlations between transmit-receive paths). For frequency domain compression, the channel information may be based on the channel frequency response. In other words, the channel matrix may represent the channel in the frequency domain.

[0084] In a time division duplex (TDD) system, the channels in the uplink (UL) and downlink (DL) directions are the same, and therefore, the transmitter (e.g., gNB) can infer the (e.g., DL) CSI from the corresponding (e.g., UL) pilots. This inference is sometimes referred to as implicit feedback. However, in a frequency division duplex (FDD) system, explicit feedback is typically required because reciprocity between the UL and DL channels cannot be guaranteed. For example, for DL ​​CSI, the UE can measure the CSI-RS and send a CSI report indicating the CSI to the transmitter (e.g., gNB).

[0085] The CSI included in the CSI report may include processed channel information that reduces the amount of (UL) data involved in the CSI feedback compared to feedback of full channel information. Such feedback of processed channel information is sometimes referred to as implicit CSI feedback and is widely used, for example, in LTE systems, where it may give satisfactory performance.

[0086] For example, in some systems (such as systems utilizing beamformed communications), the CSI may indicate a set of beams selected by the UE 120, or may include information that facilitates the selection of beams or communication parameters by the network node 110. The information included in the CSI report may include a channel quality indicator (CQI), a rank indicator (RI), a W1 codebook payload, and / or a precoding matrix indicator (PMI), etc. The CQI may indicate downlink radio conditions for a bandwidth portion, such as in terms of signal to interference plus noise ratio (SINR). The RI may indicate the number of requested MIMO layers for the UE 120. The W1 codebook payload may indicate a set of L beams selected from an oversampled codebook, and may further indicate a subspace defined by the L beams per polarization.

[0087] In NR Release 15, NR Type-II codebook CSI is standardized, which can provide significant gains over the LTE codebook at the expense of a significant increase in UL overhead. In order to gain the benefits of the increased number of antenna ports, advanced precoding schemes such as nonlinear precoding or multi-TRP transmission can be used. For these advanced schemes, high-precision CSI feedback is required. In other words, so-called explicit CSI feedback may be desired, which can allow the original channel matrix (as measured, determined, or estimated at the UE) to be derived from the CSI feedback with little or no information loss. In an ideal scenario, complete channel information (e.g., channel matrix) would be fed back by the UE.

[0088] Explicit CSI feedback may be associated with significant overhead. For example, subband information that may be provided as part of a Type II codebook described below may have a large payload. Given the limited UL channel capacity, the large size and complexity of (e.g., explicit) CSI may limit the implementation of more complex forms of CSI, thereby limiting the gains in feedback reporting and use. Furthermore, the transmission of large CSI may use a significant amount of UE and network resources.

[0089] Some techniques described herein enable compression and decompression of CSI in the frequency domain using a model (such as an artificial neural network model). In some examples, compression in the frequency domain may occur after, for example, transforming a channel impulse response (CIR) measured, determined, or estimated at a UE into a channel frequency response (CFR) via a fast Fourier transform (FFT). Therefore, the overhead associated with the transmission and reception of CSI is reduced. Some techniques described herein enable compression and decompression of CSI including explicit CSI and / or implicit CSI as described above and according to a specific example of the present disclosure. Some techniques described herein provide frequency domain compression of CQI, RI, and / or PMI, for example, by using an artificial neural network model. In some aspects described herein, the output of the artificial neural network model may replace one or more of CQI, RI, W1 codebook payload, or PMI. For example, compressed CSI may include one or more of CQI, RI, W1 codebook payload, or PMI.

[0090] In some aspects, compression and decompression use an autoencoder, which is a type of neural network for learning efficient encoding and decoding of unlabeled data. The use of an autoencoder can improve the efficiency of compression and decompression relative to some other techniques.

[0091] Compression in the frequency domain may be beneficial over compression in other domains (eg, time domain) because schedulers often schedule communications at a subband granularity, so frequency domain compression may provide satisfactory compression results for subband granularity scheduling, as well as other benefits.

[0092] Figure 5 are diagrams illustrating examples 500 and 505 of end-to-end CSI compression according to the present disclosure.

[0093] Examples 500 and 505 illustrate end-to-end compression of CSI. As indicated by reference numeral 510, UE 120 may identify channel information, which may include a channel matrix (generally represented by H). For example, UE 120 may identify the original channel by measuring a CSI-RS set sent by network node 110. In some examples, the channel information may be associated with a number of antennas, CSI ports, and / or RBs. For example, a CSI-RS set may be associated with a first number (e.g., 4) of receive ports and a second number (e.g., 32) of CSI-RS ports, and may be distributed across a third number of RBs (e.g., 50 RBs in some examples). In one example, the size of the original channel may be 4×32×50. As mentioned above, channel information may refer to information derived from the original channel without dimensionality loss (i.e., in this example, the size of the channel information (e.g., the channel matrix) may be 4×32×50).

[0094] In some aspects, the UE 120 can process the original channel. For example, the UE 120 can pre-whiten the original channel based at least in part on the observed interference before averaging.

[0095] In some aspects, as indicated by reference numeral 515, the UE may average the channel information. For example, the UE may average the original channel (e.g., channel matrix or CFR) over n RBs. For example, the UE may average the channel information (e.g., original channel) over multiple RBs carrying CSI-RS. The multiple RBs over which the channel information is averaged may be used, for example, according to Figure 4 A subset of the set of RBs that transmit the CSI-RS during any beam management process in the beam management process. In some examples, the parameter n may indicate the granularity used to average the original channel (channel matrix or CFR). For example, if there are a total of 50 RBs associated with the original channel and a granularity of n=25 is used, the UE 120 may average the original channel across a first group of 25 RBs and a second group of 25 RBs, respectively, thereby simplifying the compression of the channel information. In this example, the averaging reduces the dimensionality by a factor of 25. In other words, the size of the averaged channel information is reduced from 4×32×50 to 4×32×2, significantly reducing overhead. Therefore, the UE 120 may perform compression over n RBs. Averaging the channel information may include calculating the average Avg(H H H), where H H is the conjugate transpose of H. In some aspects, UE 120 may not average the channel information. For example, UE 120 may perform per-RB compression. In other words, the input of the artificial neural network encoder may include (unaveraged) channel information, such as Figure 5 As shown in Example 505.

[0096] In example 500, UE 120 receives channel information (e.g., channel matrix H, square channel matrix H H H or the averaged channel information Avg(H H H)) performs singular value decomposition (SVD), as shown by reference numeral 520. SVD is the factorization of a matrix (e.g., a channel matrix). The SVD of an m×n complex matrix M is of the form M=UΣV HFactorization of , where U is an m×m complex unitary matrix, Σ is an m×n rectangular diagonal matrix with non-negative real numbers on the diagonal, and V is an n×n complex unitary matrix. The diagonal entries of Σ are called the singular values ​​of M. The columns of U are called the left singular vectors of M. The columns of V are called the right singular vectors of M. The left singular vectors form a standard orthogonal basis set, and the right singular vectors form a standard orthogonal basis set. The right singular vectors after SVD may be referred to as a beamformer or SVD beam in this article. The right singular vector may indicate a beam set derived from the channel information. In example 505, UE120 does not perform (i.e., omits) SVD on the channel information. Performing SVD may provide CSI with finer granularity than providing implicit feedback, and not performing SVD may save processing resources at UE 120.

[0097] As indicated by reference numeral 525, the UE may provide input to the model. The input may be based at least in part on channel information. For example, in example 500, the input to the model may include one or more (e.g., all) SVD beams generated at least in part based on channel information, as shown in association with reference numeral 520. As another example, in example 505, the input to the model may include channel information (e.g., an averaged channel matrix, an unaveraged channel matrix such as an original channel, etc.). In some aspects, the input to the model may include one or more (e.g., all) singular values ​​of the singular values ​​of the SVD. In examples 500 and 505, the model includes an artificial neural network encoder or is an artificial neural network encoder, described in more detail below.

[0098] As indicated by reference numeral 530, the model may output compressed CSI, and thus, the UE 120 may generate compressed CSI. The CSI may be compressed in the frequency domain, as described above. In some aspects, the model may use position encoding to identify each subband included in the compressed CSI. The final linear layer of the model that performs compression of the CSI may work across the subband (e.g., frequency) dimension as well as the gNB antenna dimension. The model may generate a reduced-dimensional version of the CSI such that the CSI is compressed in the frequency domain. The compressed CSI may include a compressed version of the SVD beam (in example 500) or a compressed version of the channel information (in example 505). The UE 120 may send the compressed CSI to the network node 110.

[0099] In some aspects, the compressed CSI may include a PMI payload. For example, the compressed CSI may indicate a precoder set or one or more PMIs once decompressed by the network node 110. In this example, the output of the model may be a compressed version of the precoder set or one or more PMIs. For example, the output of the model may provide one or more PMIs and / or one or more precoders derived from a singular vector (e.g., right singular vector) set determined at least in part based on channel information used to generate the compressed CSI. In some aspects, the compressed CSI may include a PMI payload and a W1 codebook payload. For example (e.g., in example 500), the compressed CSI may indicate a precoder set, one or more PMIs, a selection of a DFT oversampling codebook, and / or a subspace defined by L beams per polarization once decompressed by the network node 110. In this example, the output of the model may be a compressed version of a precoder set, one or more PMIs, a selection of a DFT oversampling codebook, and / or a subspace defined by L beams per polarization. In some aspects, the compressed CSI may include a PMI payload, a W1 codebook payload, a CQI, and a RI. For example (e.g., in example 505), the compressed CSI once decompressed by the network node 110 may indicate a precoder set, a PMI payload, a W1 codebook payload, a CQI, and a RI.

[0100] In some aspects, the compressed CSI may include one or more singular values ​​derived from the channel matrix or the averaged channel matrix via SVD. One or more singular values ​​may be the main (largest) singular values ​​obtained by SVD. For example, UE 120 may send one or more singular values, and network node 110 may receive one or more singular values. In some aspects, in addition to one or more singular values, UE 120 may also send broadband and / or subband CQI values ​​of CSI. Additionally or alternatively, in addition to one or more singular values, UE 120 may also send one or more PMI values ​​and / or one or more RI values. In some aspects, UE 120 may send one or more singular values ​​without sending broadband and / or subband CQI values. In some aspects, one or more singular values ​​may be considered as broadband and / or subband CQI values. In some aspects, singular values ​​may be differentially encoded, for example, relative to a baseline value. In some aspects, UE 120 may quantize singular values. For example, the network node 110 may send and the UE 120 may receive information indicating quantization for the singular values. The UE 120 may quantize the singular values ​​according to the quantization.

[0101] The network node 110 may decompress the compressed CSI. For example, the network node 110 may use a model (e.g., an artificial neural network model) to decompress the compressed CSI. In some aspects, the model used by the network node 110 and the model used by the UE 120 may be associated with an autoencoder, as described elsewhere herein. Thus, the network node 110 may recover the CSI sent by the UE 120. The network node 110 may communicate based at least in part on the decompressed CSI. For example, the network node 110 may schedule communications with the UE 120 based at least in part on the decompressed CSI, or configure the UE 120 for communications.

[0102] As indicated above, Figure 5 are provided as examples. Other examples may be relevant to Figure 5 The examples described are different.

[0103] Figure 6 600 is a diagram illustrating an example of hybrid CSI compression according to the present disclosure. The operations of example 600 may be performed by UE 120. Example 600 is referred to as hybrid CSI compression because example 600 incorporates aspects of model-based compression of enhanced type II (eType-II) CSI feedback and CSI feedback. A CSI report may include a codebook, which is a set of precoders or one or more PMIs. A type I codebook may include a predefined matrix. A type-II codebook may include a more detailed CSI report for multi-user MIMO and may include a set of beams. In some cases, type IICSI feedback may use a compressed type II precoder. This may reduce the overhead of type IICSI feedback. A compressed precoder may exploit sparsity in the spatial domain and / or frequency domain. In Release 17 of 3GPP Technical Specification (TS) 38.214 (eg, at sections 5.2.2.2.5 and 5.2.2.2.6), procedures for eType-IICSI feedback (including eType-IICSI codebook, compression method, and reporting) are defined.

[0104] like Figure 6As shown by reference numeral 605, UE 120 can identify channel information, which can include a channel matrix (generally represented by H). For example, UE 120 can identify the original channel by measuring a CSI-RS set sent by network node 110. In some examples, the channel information can be associated with a number of antennas, CSI ports, and / or RBs. For example, a CSI-RS set can be associated with 4 receive ports and 32 CSI-RS ports, and can be distributed across a number of RBs (e.g., 50 RBs in some examples). Therefore, the size of the original channel can be 4×32×50. In some aspects, UE 120 can process the original channel. For example, UE 120 can pre-whiten the original channel based at least in part on observed interference before averaging.

[0105] In some aspects, as shown in reference numeral 610, the UE may select the best codebook and the top L beams of the codebook (where L is an integer) for the wideband. For example, the UE 120 may select the top L beams based at least in part on CSI-RS measurements associated with the top L beams. In addition, the UE may average the channel as projected onto the top L beams over n RBs. The projection onto the top L beams reduces the (spatial) dimension of the channel (information), thereby reducing the overhead of CSI signaling. The parameter n may indicate the granularity for averaging the original channel. For example, if there are a total of 50 RBs associated with the original channel and a granularity of n=25 is used, the UE 120 may average the projected channel across a first group of 25 RBs and a second group of 25 RBs, respectively, thereby simplifying the compression of the channel information. Therefore, the UE 120 may perform compression over n RBs. Averaging the projected channel information may include calculating the average Avg((HB) H (HB)), where (HB) represents the projection of channel H onto the top L beams. In some aspects, UE 120 may not average the projected channels (information). For example, UE 120 may perform per-RB compression.

[0106] As indicated by reference numeral 615, in some aspects, the UE 120 performs SVD on the projected channel information (eg, the averaged channel information). As mentioned above, the SVD of an m×n complex matrix M is of the form M=UΣV HFactorization of , where U is an m×m complex unitary matrix, Σ is an m×n rectangular diagonal matrix with non-negative real numbers on the diagonal, and V is an n×n complex unitary matrix. The diagonal entries of Σ are called the singular values ​​of M. The columns of U are called the left singular vectors of M. The columns of V are called the right singular vectors of M (e.g., beamformer, SVD beams). As shown in the reference numeral 620, for simplicity, finding the best codebook, selecting the top L beams of the best codebook, projecting the original channel (e.g., channel matrix) onto L beams (i.e., subspaces), averaging the projected channels, and / or performing SVD on the averaged projected channels may be referred to herein as projecting the channel H onto the subspace. Therefore, the UE 120 may project the received channel onto the subspace. The projection of the received channel may have a lower dimension than the channel matrix of the received channel.

[0107] As indicated by reference numeral 625, the UE may provide input to a model (e.g., an artificial neural network). The input may be based at least in part on the projected channel information. For example, in example 600, the input to the model may include an averaged projected channel that is projected onto the top L beams of the optimal codebook. In some aspects (e.g., for type IICSI), the input may be the original channel of subspace HW1 (e.g., the UE 120 may not perform SVD on the (averaged or unaveraged) projected channel information). In some aspects (e.g., for type IICSI), the input may be a set of SVD beams of subspace HW1 (with or without averaging). In some aspects, (e.g., for eType-IICSI), the input may be subspace HW1W f The original channel (eg, UE 120 may not perform SVD on the (averaged or unaveraged) projected channel information). In some aspects, (eg, for eType-IICSI), the input may be the subspace HW1W f A set of SVD beams of (with or without averaging). In some aspects (eg, for Doppler CSI), the input may be the subspace HW1W f W t The original channel (eg, UE 120 may not perform SVD on the (averaged or unaveraged) projected channel information). In some aspects, (eg, for Doppler CSI), the input may be the subspace HW1W f W t The Doppler CSI may include the SVD beams of W t The time domain correlation or Doppler domain information represented by the channel information is used to assist in downlink precoding. Therefore, in some aspects, the input of the model is based at least in part on a matrix (W) containing one or more frequency domain bases derived from the channel information.f )of.

[0108] As shown in the reference numeral 630, the model may output compressed CSI, and thus, the UE 120 may generate compressed CSI. The CSI may be compressed in the frequency domain, as described above. For example, the model may generate a reduced-dimensional version of the CSI so that the CSI is compressed in the frequency domain. The compressed CSI may include a compressed version of the SVD beam or a compressed version of the projected channel information. The UE 120 may send the compressed CSI to the network node 110. In some aspects, the compressed CSI may include a PMI payload. For example, the compressed CSI may indicate a precoder set or one or more PMIs once it is decompressed by the network node 110. In this example, the output of the model may be a compressed version of the precoder set or one or more PMIs.

[0109] In some aspects, the compressed CSI may include one or more singular values ​​derived from the channel matrix or the averaged channel matrix after being projected onto the subspace via SVD. One or more singular values ​​may be the main (largest) singular values ​​obtained by SVD. For example, UE 120 may send one or more singular values, and network node 110 may receive one or more singular values. In some aspects, in addition to one or more singular values, UE 120 may also send broadband and / or subband CQI values ​​of CSI (and / or PMI, and / or RI). In some aspects, UE 120 may send one or more singular values ​​without sending broadband and / or subband CQI values. In some aspects, one or more singular values ​​may be considered as broadband and / or subband CQI values. In some aspects, singular values ​​may be differentially encoded, for example, relative to a baseline value. In some aspects, UE 120 may quantize singular values. For example, the network node 110 may transmit information indicating quantization for the singular values, and the UE 120 may receive the information. The UE 120 may quantize the singular values ​​according to the quantization.

[0110] The network node 110 may decompress the compressed CSI. For example, the network node 110 may use a model (e.g., a neural network model) to decompress the compressed CSI. In some aspects, the model used by the network node 110 and the model used by the UE 120 may be associated with an autoencoder, as described elsewhere herein. Thus, the network node 110 may recover the CSI sent by the UE 120. The network node 110 may communicate based at least in part on the decompressed CSI. For example, the network node 110 may schedule communications with the UE 120 based at least in part on the decompressed CSI, or configure the UE 120 for communication.

[0111] As indicated above, Figure 6 are provided as examples. Other examples may be relevant to Figure 6 The examples described are different.

[0112] Figure 7 700 is a diagram illustrating an example of signaling associated with compressed CSI feedback according to the present disclosure. As shown, the example 700 includes a UE 120 and a network node 110. In some aspects, different operations of the example 700 may be performed by different network nodes 110. For example, a first network node 110 may send configuration information, and a second network node 110 may send a CSI-RS set.

[0113] As indicated by reference numeral 710, the network node 110 may send configuration information, and the UE 120 may receive the configuration information. The configuration information may be signaled via radio resource control (RRC) signaling, medium access control (MAC) signaling, downlink control information (DCI), system information, or a combination thereof. Although example 700 shows a single transmission of configuration information, in some aspects, the configuration information may be sent via multiple different transmissions (e.g., different portions of the configuration information may be sent in different transmissions). The configuration information may include one or more parameters, which are shown by dashed boxes in example 700.

[0114] In some aspects, the configuration information may include information indicating a maximum payload size of the CSI. For example, the configuration information may indicate a maximum payload size of the compressed CSI, or a maximum payload size of a CSI report carrying the compressed CSI. In some aspects, the UE 120 may determine one or more parameters for compressing the CSI based at least in part on the maximum payload size. For example, the UE 120 may determine the dimension of the output of the model based at least in part on the maximum payload size (e.g., so that the output of the model does not exceed the maximum payload size). As another example, the UE 120 may determine the quantization of the output of the model based at least in part on the maximum payload size (e.g., so that the quantized output of the model does not exceed the maximum payload size). In some aspects, the configuration information may include a set of parameters for the model. For example, the configuration information may indicate a set of parameters of an artificial neural network model, such as quantization, dimension, hyperparameters (e.g., parameters indicating the structure of the neural network model), etc. of the neural network model.

[0115] In some aspects, the configuration information may include an indication of the granularity of the averaged channel matrix. For example, the configuration may indicate the number n of RBs over which the UE 120 should average the channel matrix (e.g., the original channel), such as in conjunction with Figure 5 and Figure 6 Described in more detail.

[0116] In some aspects, the configuration information may indicate a CSI compression technique for the CSI. For example, the configuration information may indicate whether the UE 120 should generate eType-IICSI, compressed CSI using end-to-end compression (as in example 500 or example 505), or compressed CSI using hybrid compression (as in example 600). As another example, the configuration information may indicate whether to perform SVD. According to the indicated CSI compression technique, the UE 120 may compress the CSI, and the network node 110 may decompress the CSI.

[0117] In some aspects, the configuration information may indicate a compression scheme associated with one or more (DL) layers of CSI. For example, CSI associated with a larger number of layers has a larger payload than CSI associated with a smaller number of layers. UE 120 may compress CSI based at least in part on the rank associated with CSI. For example, the payload size of the compressed CSI may be fixed, regardless of the rank associated with the CSI, which simplifies uplink scheduling. As another example, each layer of the compressed CSI may have the same payload size, so that the payload size of the compressed CSI increases linearly with the rank of the CSI. As another example, there may be a first payload size for CSI with rank 1 and a second payload size for CSI with rank greater than 1 (or more generally, a first payload size for CSI with rank 1 to X and a second payload size for CSI with rank greater than X, where X is a rank threshold). In this example, the first payload size, the second payload size, the relationship between the first payload and the second payload, and / or the value of the rank threshold may be configured by the network node 110. In some aspects, the configuration information may indicate parameters associated with joint compression of multiple layers, such as payload size. In some aspects, the configuration information may indicate parameters associated with individual compression of one or more layers (e.g., one or more V vectors of CSI), such as relative to the strongest layer (e.g., the strongest V vector). For example, the parameters associated with individual compression of one or more layers may indicate a compression accuracy that is lower than the compression accuracy of the strongest layer.

[0118] The compressed CSI may be associated with an index configuration. The index configuration may indicate how to identify elements of the input to the model (such as elements of the channel matrix of the original channel, elements of the averaged channel matrix, elements of the right singular vectors, or elements of the SVD) by index. In some aspects, the index configuration may indicate a receive antenna index, followed by a CSI-RS transmit port index, followed by a subband index for identifying elements of the input to the model (e.g., when the original channel is input to the model). In some aspects, the index configuration may indicate a layer index, followed by a CSI-RS transmit port index, followed by a subband index for identifying elements of the input to the model (e.g., when a beamformer is input to the model).

[0119] As indicated by reference numeral 720, the network node 110 may transmit a CSI-RS set. The CSI-RS set may be transmitted in one or more RBs. As indicated by reference numeral 730, the UE 120 may identify (e.g., measure, determine, or estimate) channel information (e.g., an original channel) based at least in part on the CSI-RS set.

[0120] As indicated by reference numeral 740, UE 120 may generate compressed CSI using the model. UE 120 may generate compressed CSI according to one or more parameters indicated by the configuration information indicated by reference numeral 710. The generation of compressed CSI is combined with Figure 5 and Figure 6 Described in more detail.

[0121] As indicated by reference numeral 750, UE 120 may send compressed CSI and network node 110 may receive compressed CSI. For example, UE 120 may send a CSI report including the compressed CSI. In some aspects, the CSI report may include the compressed CSI and one or more CSI parameters (such as RI, CQI, W1 codebook payload, etc.) in addition to the compressed CSI. In some aspects, the CSI report may include only the compressed CSI (e.g., all CSI parameters may be provided via the compressed CSI).

[0122] As indicated by reference numeral 760, the network node 110 may decompress the compressed CSI. In some aspects, the network node 110 may use a model (e.g., an artificial neural network decoder associated with an artificial neural network encoder used to compress the CSI) to decompress the compressed CSI. The artificial neural network decoder may be associated with the artificial neural network encoder based on the artificial neural network decoder and the artificial neural network encoder forming an autoencoder.

[0123] As indicated at 770, network node 110 may configure communications for UE 120 based at least in part on the decompressed CSI. For example, network node 110 may select one or more beams for UE 120, one or more communication parameters for UE 120, and the like.

[0124] In this manner, the CSI is compressed and then decompressed (eg, using an autoencoder), which reduces the overhead associated with CSI transmission and increases the amount of data that can be transmitted via the CSI at a given overhead level.

[0125] As indicated above, Figure 7 are provided as examples. Other examples may be relevant to Figure 7 The examples described are different.

[0126] Figure 8 8 is a diagram illustrating an example 800 associated with an auto-encoder for CSI compression according to the present disclosure. As shown, UE 120 may communicate with network node 110. UE 120 and network node 110 may communicate via a wireless network (e.g., Figure 1 The wireless network 100 shown communicates with each other.

[0127] As shown, UE 120 may include a communication manager 140 that may be configured to utilize an autoencoder 808 to perform one or more wireless communication tasks. The communication manager 140 may be configured to utilize Figure 8 In some aspects, the communication manager 140 may be a combination of the following: Fig.11 The compression component 1108 described herein includes the compression component 1108 or is included in the compression component 1108.

[0128] As shown, the autoencoder 808 may include an encoder portion 810 configured to receive an input and provide compressed CSI as an output. The encoder portion 810 may be or may be similar to the Figure 7 The model is described by reference numeral 740. The autoencoder 808 may also include a decoder portion 812 configured to receive the compressed CSI and provide a decompressed CSI (eg, at least a portion or an approximation of the input to the encoder portion 810) as an output.

[0129] like Figure 8 As shown, the network node 110 may include a communication manager 150, which may be configured to utilize the autoencoder 816 to perform one or more wireless communication tasks. In some aspects, the communication manager 150 may be a communication manager 150 configured to utilize the autoencoder 816 to perform one or more wireless communication tasks. Fig.12 The decompression component 1208 described herein may include, include, or be included in the decompression component 1208. In some aspects, the autoencoder 816 may correspond to the autoencoder 808. In some aspects, the autoencoder 816 may be a replica of the autoencoder 808. In some aspects, the UE 120 may include only the encoder portion 810 and / or the network node 110 may include only the decoder portion 818, which is configured to receive the compressed CSI as input and provide the decompressed CSI as output. The communication manager 150 may be configured to utilize Figure 8 The autoencoder 816 may include an encoder portion 820 configured to receive an input and provide compressed CSI as an output.

[0130] like Figure 8 As shown, in some examples, the network node 110 may include or may be associated with a transceiver (shown as Tx / Rx 824) that may facilitate wireless communications with a transceiver 822 of the UE 120. In some aspects, the network node 110 may not include a transceiver that facilitates direct wireless communications with the UE 120. For example, the network node 110 may be associated with a radio that communicates directly with the UE 120. As shown at reference numeral 826, the network node 110 may use the transceiver 824 to send wireless communications to the UE 120. The wireless communications may include, for example, an encoder portion 810 of the autoencoder 808 and / or Figure 7 The transceiver 822 of the UE 120 may receive wireless communications. The communication manager 140 may decode and instantiate the encoder part 810.

[0131] As shown, the communication manager 140 can provide input based at least in part on the channel information to an encoder portion 810 of the autoencoder 808. The encoder portion 810 of the autoencoder 808 can determine the compressed CSI. As shown, the communication manager 140 can provide the compressed CSI to the transceiver 822 for transmission. As shown by reference numeral 828, the transceiver 822 can send the compressed CSI, and the transceiver 824 of the network node 110 can receive the compressed CSI.

[0132] As shown, the communication manager 150 of the network node 110 may provide the compressed CSI as input to a decoder portion 818 of the autoencoder 816. The decoder portion 818 may determine (e.g., decompress, reconstruct, approximate) the CSI or input to the encoder portion 810. In some aspects, the network node 110 may retrain the decoder portion 818 based at least in part on the CSI.

[0133] As indicated above, Figure 8 These are provided as examples only. Other examples may differ from those described herein. Figure 8 Examples described.

[0134] Fig. 9 9 is a diagram illustrating an example process 900 performed, for example, by a UE, in accordance with the present disclosure. Example process 900 is an example of a UE (eg, UE 120) performing operations associated with frequency domain compression of channel state information.

[0135] like Fig. 9 As shown, in some aspects, process 900 may include identifying channel information based at least in part on a set of channel state information reference signals (CSI-RS) (block 910). For example, a UE (e.g., using Fig. 9 The communication manager 140 and / or the measurement component 1110 depicted in FIG. 1 may identify channel information based at least in part on a set of channel state information reference signals (CSI-RS), as described above.

[0136] like Fig. 9 As further shown in FIG. 9 , in some aspects, process 900 may include: generating compressed CSI using a model, wherein an input to the model is based at least in part on channel information, and wherein the compressed CSI is compressed in the frequency domain (block 920). For example, a UE (e.g., using Fig.11 The communication manager 140 and / or compression component 1108 depicted in FIG. 1 may use a model to generate compressed CSI, wherein an input to the model is based at least in part on channel information, and wherein the compressed CSI is compressed in the frequency domain, as described above.

[0137] like Fig. 9 As further shown in FIG. 1 , in some aspects, process 900 may include sending compressed CSI (block 930). For example, a UE (e.g., using Fig.11 The communication manager 140 and / or the sending component 1104 depicted in FIG. 1 may send the compressed CSI as described above.

[0138] Process 900 may include additional aspects, such as any single aspect or any combination of the aspects described below and / or in conjunction with one or more other processes described elsewhere herein.

[0139] In a first aspect, the model is a neural network encoder.

[0140] In a second aspect, alone or in combination with the first aspect, the compressed CSI is based at least in part on right singular vectors associated with the channel information, and left singular vectors associated with the channel information are omitted.

[0141] In a third aspect, alone or in combination with one or more of the first and second aspects, the channel information comprises at least a portion of a singular value decomposition associated with a channel matrix.

[0142] In a fourth aspect, alone or in combination with one or more of the first to third aspects, the channel information comprises a channel matrix.

[0143] In a fifth aspect, alone or in combination with one or more of the first to fourth aspects, the channel information comprises at least a portion of a singular value decomposition associated with a projection of a channel matrix, wherein the projection is onto a top beam set of a codebook.

[0144] In a sixth aspect, alone or in combination with one or more of the first to fifth aspects, the compressed CSI includes a precoding matrix indicator payload output by the model.

[0145] In a seventh aspect, either alone or in combination with one or more of the first to sixth aspects, process 900 includes receiving information indicating a maximum payload size, wherein generating compressed CSI is based at least in part on the maximum payload size.

[0146] In an eighth aspect, alone or in combination with one or more of the first to seventh aspects, process 900 includes receiving an indication of a granularity of an average channel matrix, wherein identifying channel information further includes generating the average channel matrix using the granularity.

[0147] In a ninth aspect, alone or in combination with one or more of the first to eighth aspects, process 900 includes receiving a set of parameters for a model, wherein generating compressed CSI is based at least in part on the set of parameters.

[0148] In a tenth aspect, alone or in combination with one or more of the first to ninth aspects, the compressed CSI comprises one or more singular values ​​based at least in part on a channel matrix.

[0149] In an eleventh aspect, alone or in combination with one or more of the first to tenth aspects, a compression scheme for the compressed CSI is based at least in part on a number of layers associated with the compressed CSI.

[0150] Although Fig. 9 Example blocks of process 900 are shown, but in some aspects process 900 may include Fig. 9 In some embodiments, the process 900 may include additional blocks, fewer blocks, different blocks, or blocks arranged in a different manner than those depicted in the process 900. Additionally or alternatively, two or more blocks of the process 900 may be executed in parallel.

[0151] Fig.10 is a diagram illustrating an example process 1000, performed, for example, by a network node, in accordance with the present disclosure. Example process 1000 is an example of a network node (eg, network node 110, a plurality of network nodes) performing operations associated with frequency domain compression of channel state information.

[0152] As shown in the figure, Fig.10 As shown, in some aspects, process 1000 may include receiving compressed CSI compressed in the frequency domain (block 1010). For example, a network node (e.g., using Fig.12 The communication manager 150 and / or receiving component 1202 depicted in FIG. 1 may receive compressed CSI compressed in the frequency domain, as described above.

[0153] like Fig.10 As further shown, in some aspects, process 1000 may include: using the model to decompress the compressed CSI to obtain channel information (block 1020). For example, a network node (e.g., using Fig.12 The communication manager 150 and / or decompression component 1208 depicted in FIG. 1 may use the model to decompress the compressed CSI to obtain channel information, as described above.

[0154] like Fig.10 As further shown, in some aspects, process 1000 may include configuring communications based at least in part on the channel information (block 1030). For example, a network node (e.g., using Fig.12 The depicted communications manager 150 and / or configuration component 1210) can configure communications based at least in part on the channel information, as described above.

[0155] Process 1000 may include additional aspects, such as any single aspect or any combination of the aspects described below and / or in conjunction with one or more other processes described elsewhere herein.

[0156] In a first aspect, the model is a neural network decoder.

[0157] In a second aspect, alone or in combination with the first aspect, the decompressed CSI is based at least in part on right singular vectors associated with the channel information, and left singular vectors associated with the channel information are omitted.

[0158] In a third aspect, alone or in combination with one or more of the first and second aspects, the channel information comprises at least a portion of a singular value decomposition associated with a channel matrix.

[0159] In a fourth aspect, alone or in combination with one or more of the first to third aspects, the control information includes a channel matrix.

[0160] In a fifth aspect, alone or in combination with one or more of the first to fourth aspects, the channel information comprises at least a portion of a singular value decomposition associated with a projection of a channel matrix, wherein the projection is onto a top beam set of a codebook.

[0161] In a sixth aspect, alone or in combination with one or more of the first to fifth aspects, the compressed CSI includes a precoding matrix indicator payload.

[0162] In a seventh aspect, either alone or in combination with one or more of aspects 1 to 6, process 1000 includes sending information indicating a maximum payload size, wherein the compressed CSI is based at least in part on the maximum payload size.

[0163] In an eighth aspect, alone or in combination with one or more of the first to seventh aspects, the process 1000 includes sending an indication of a granularity of an average channel matrix of the channel information.

[0164] In a ninth aspect, either alone or in combination with one or more of aspects 1 to 8, process 1000 includes sending a set of parameters associated with generating compressed CSI for a neural network encoder.

[0165] In a tenth aspect, alone or in combination with one or more of the first to ninth aspects, the compressed CSI comprises one or more singular values ​​based at least in part on a channel matrix.

[0166] In an eleventh aspect, alone or in combination with one or more of the first to tenth aspects, a compression scheme for the compressed CSI is based at least in part on a number of layers associated with the compressed CSI.

[0167] Although Fig.10Example blocks of process 1000 are shown, but in some aspects, process 1000 may include Fig.10 In some embodiments, the process 1000 may include additional blocks, fewer blocks, different blocks, or blocks arranged in a different manner than those depicted in the process 1000. Additionally or alternatively, two or more blocks of the process 1000 may be executed in parallel.

[0168] Fig.11 1 is a diagram of an example apparatus 1100 for wireless communication according to the present disclosure. The apparatus 1100 may be a UE, or the UE may include the apparatus 1100. In some aspects, the apparatus 1100 includes a receiving component 1102 and a transmitting component 1104, which may communicate with each other (e.g., via one or more buses and / or one or more other components). As shown, the apparatus 1100 may communicate with another apparatus 1106 (e.g., a UE, a base station, or another wireless communication device) using the receiving component 1102 and the transmitting component 1104. As further shown, the apparatus 1100 may include a communication manager 140. The communication manager 140 may include one or more of a compression component 1108 or a measurement component 1110, as well as other examples.

[0169] In some aspects, the apparatus 1100 may be configured to perform the Figure 3-Figure 8 Additionally or alternatively, the apparatus 1100 may be configured to perform one or more processes described herein, such as Fig. 9 In some aspects, the apparatus 1100 and / or Fig.11 One or more of the components shown may include a combination of Figure 2 One or more components of the UE described. Additionally or alternatively, Fig.11 One or more of the components shown in the figure may be combined with Figure 2 In some embodiments, the present invention may be implemented in one or more components described herein. Additionally or alternatively, one or more components in the component set may be at least partially implemented as software stored in a memory. For example, a component (or a portion of a component) may be implemented as instructions or codes stored in a non-transitory computer-readable medium and may be executed by a controller or processor to perform the function or operation of the component.

[0170] The receiving component 1102 may receive communications from the device 1106, such as reference signals, control information, data communications, or a combination thereof. The receiving component 1102 may provide the received communications to one or more other components of the device 1100. In some aspects, the receiving component 1102 may perform signal processing (e.g., filtering, amplification, demodulation, analog-to-digital conversion, demultiplexing, deinterleaving, demapping, equalization, interference cancellation, or decoding, etc.) on the received communications and may provide the processed signals to one or more other components of the device 1100. In some aspects, the receiving component 1102 may include the above in combination with Figure 2 One or more antennas, modems, demodulators, MIMO detectors, receive processors, controllers / processors, memories, or combinations thereof of the described UE.

[0171] The transmitting component 1104 may transmit communications, such as reference signals, control information, data communications, or a combination thereof, to the device 1106. In some aspects, one or more other components of the device 1100 may generate communications and may provide the generated communications to the transmitting component 1104 for transmission to the device 1106. In some aspects, the transmitting component 1104 may perform signal processing (such as filtering, amplification, modulation, digital-to-analog conversion, multiplexing, interleaving, mapping, or encoding, among other examples) on the generated communications and may transmit the processed signals to the device 1106. In some aspects, the transmitting component 1104 may include the above in combination with Figure 2 One or more antennas, modems, modulators, transmit MIMO processors, transmit processors, controllers / processors, memories, or combinations thereof of the described UE. In some aspects, the transmit component 1104 can be co-located with the receive component 1102 in a transceiver.

[0172] The measuring component 1110 can identify channel information based at least in part on the CSI-RS set. The compressing component 1108 can generate compressed CSI using a model, wherein an input to the model is based at least in part on the channel information, and wherein the compressed CSI is compressed in the frequency domain. The transmitting component 1104 can transmit the compressed CSI.

[0173] Fig.11 The number and arrangement of components shown in are provided as examples. In practice, there may be Fig.11 The components shown may include additional components, fewer components, different components, or components arranged in a different manner than those shown. Fig.11 Two or more components shown may be implemented in a single component, or Fig.11 The single component shown may be implemented as multiple distributed components. Additionally or alternatively, Fig.11The collection of (eg, one or more) components shown in can perform the operations described by Fig.11 Another collection of components shown in the figure performs one or more functions.

[0174] Fig.12 1 is a diagram of an example apparatus 1200 for wireless communication according to the present disclosure. The apparatus 1200 may be a network node, or a network node may include the apparatus 1200. In some aspects, the apparatus 1200 includes a receiving component 1202 and a sending component 1204, which may communicate with each other (e.g., via one or more buses and / or one or more other components). As shown, the apparatus 1200 may communicate with another apparatus 1206 (e.g., a UE, a base station, or another wireless communication device) using the receiving component 1202 and the sending component 1204. As further shown, the apparatus 1200 may include a communication manager 150. The communication manager 150 may include one or more of a decompression component 1208 or a configuration component 1210, as well as other examples.

[0175] In some aspects, the apparatus 1200 may be configured to perform the Figure 3-Figure 9 Additionally or alternatively, the apparatus 1200 may be configured to perform one or more processes described herein, such as Fig.10 In some aspects, the apparatus 1200 and / or Fig.12 One or more of the components shown may include a combination of Figure 2 One or more components of the network node described. Additionally or alternatively, Fig.12 One or more of the components shown in the figure may be combined with Figure 2 In addition or alternatively, one or more components in the component set may be at least partially implemented as software stored in a memory. For example, a component (or a portion of a component) may be implemented as instructions or codes stored in a non-transitory computer-readable medium and may be executed by a controller or processor to perform the function or operation of the component.

[0176] The receiving component 1202 may receive communications from the apparatus 1206, such as reference signals, control information, data communications, or a combination thereof. The receiving component 1202 may provide the received communications to one or more other components of the apparatus 1200. In some aspects, the receiving component 1202 may perform signal processing (e.g., filtering, amplification, demodulation, analog-to-digital conversion, demultiplexing, deinterleaving, demapping, equalization, interference cancellation, or decoding, etc.) on the received communications and may provide the processed signals to one or more other components of the apparatus 1200. In some aspects, the receiving component 1202 may include combining Figure 2 One or more antennas, modems, demodulators, MIMO detectors, receive processors, controllers / processors, memories, or combinations thereof of the described network nodes.

[0177] The transmitting component 1204 may transmit communications, such as reference signals, control information, data communications, or a combination thereof, to the device 1206. In some aspects, one or more other components of the device 1200 may generate communications and may provide the generated communications to the transmitting component 1204 for transmission to the device 1206. In some aspects, the transmitting component 1204 may perform signal processing (such as filtering, amplification, modulation, digital-to-analog conversion, multiplexing, interleaving, mapping, or encoding, among other examples) on the generated communications and may transmit the processed signals to the device 1206. In some aspects, the transmitting component 1204 may include combining Figure 2 One or more antennas, modems, modulators, transmit MIMO processors, transmit processors, controllers / processors, memories, or combinations thereof of the described network nodes. In some aspects, the transmit component 1204 can be co-located with the receive component 1202 in a transceiver.

[0178] Receiving component 1202 can receive compressed CSI compressed in the frequency domain. Decompressing component 1208 can use the model to decompress the compressed CSI to obtain channel information. Configuring component 1210 or transmitting component 1204 can configure communications based at least in part on the channel information.

[0179] Fig.12 The number and arrangement of components shown in are provided as examples. In practice, there may be Fig.12 The components shown may include additional components, fewer components, different components, or components arranged in a different manner than those shown. Fig.12 Two or more components shown may be implemented in a single component, or Fig.12 The single component shown may be implemented as multiple distributed components. Additionally or alternatively, Fig.12 The assembly of components (e.g., one or more components) shown in FIG. 1 may perform the operations described by Fig.12 Another collection of components shown in the figure performs one or more functions.

[0180] A summary of some aspects of the disclosure is provided below.

[0181] Aspect 1: A method of wireless communication performed by a user equipment (UE), comprising: identifying channel information based at least in part on a channel state information reference signal (CSI-RS) set; generating compressed channel state information (CSI) using a neural network model, wherein an input of the neural network model is based at least in part on the channel information, and wherein the compressed CSI is compressed in the frequency domain; and sending the compressed CSI.

[0182] Aspect 2: The method according to aspect 1, wherein the channel information is based on a channel matrix representing the channel in the frequency domain.

[0183] Aspect 3: A method according to any one of Aspects 1-2, wherein the input to the neural network model is based at least in part on an average of the channel information across a set of resource blocks (RBs).

[0184] Aspect 4: The method according to aspect 3 further includes: receiving an indication of a granularity for indicating the number of RBs included in the RB set.

[0185] Aspect 5: The method according to aspect 3, wherein the compressed CSI replaces the precoding matrix indicator payload, the channel quality indicator payload, the rank indicator payload and the W1 codebook payload in the CSI report to be sent by the UE.

[0186] Aspect 6: The method according to any one of aspects 1-5, wherein the input to the model is based at least in part on a projection of a channel matrix, wherein the projection is onto a set of beams of a codebook.

[0187] Aspect 7: A method according to any one of Aspects 1-6, wherein the input to the neural network model is based at least in part on a singular value decomposition (SVD) associated with the channel information.

[0188] Aspect 8: The method according to Aspect 7, wherein the input of the neural network model includes one or more singular values ​​of the SVD associated with the channel information.

[0189] Aspect 9: The method according to Aspect 7, wherein the input of the neural network model includes one or more singular vectors of the SVD associated with the channel information.

[0190] Aspect 10: The method according to Aspect 9, wherein the one or more singular vectors of the SVD are right singular vectors.

[0191] Aspect 11: The method according to aspect 7, wherein the compressed CSI replaces at least a precoding matrix indicator payload and a W1 codebook payload in a CSI report to be sent by the UE.

[0192] Aspect 12: The method according to any one of aspects 1-11, wherein the compressed CSI replaces at least a precoding matrix indicator payload in a CSI report to be sent by the UE.

[0193] Aspect 13: A method according to any one of Aspects 1-12, wherein the neural network model is an artificial neural network encoder.

[0194] Aspect 14: The method according to any one of aspects 1-13 further comprises: receiving information indicating a maximum payload size, wherein generating the compressed CSI is based at least in part on the maximum payload size.

[0195] Aspect 15: The method according to any one of aspects 1-14, further comprising: receiving a set of parameters for the model, wherein generating the compressed CSI is based at least in part on the set of parameters.

[0196] Aspect 16: The method according to any one of aspects 1-15, wherein a compression scheme used for the compressed CSI is based at least in part on the number of layers associated with the compressed CSI.

[0197] Aspect 17: A method according to any one of Aspects 1-16, wherein the input to the neural network model is based at least in part on a matrix containing one or more frequency domain bases derived from the channel information.

[0198] Aspect 18: A method of wireless communication performed by a network node, comprising: receiving compressed channel state information (CSI) compressed in the frequency domain; using a neural network model to decompress the compressed CSI to obtain channel information; and configuring communication based at least in part on the channel information.

[0199] Aspect 19: The method according to aspect 18, wherein the channel information is based at least in part on a channel matrix representing the channel in the frequency domain.

[0200] Aspect 20: The method according to any one of aspects 18-19, wherein the compressed CSI is based at least in part on an average of the channel information across a set of resource blocks (RBs).

[0201] Aspect 21: The method according to aspect 20 further includes: sending an indication of a granularity for indicating the number of RBs included in the RB set.

[0202] Aspect 22: The method according to aspect 20, wherein the compressed CSI replaces the precoding matrix indicator payload, the channel quality indicator payload, the rank indicator payload and the W1 codebook payload in the CSI report.

[0203] Aspect 23: The method according to any one of aspects 18-22, wherein the compressed CSI is based at least in part on a projection of a channel matrix, wherein the projection is onto a set of beams of a codebook.

[0204] Aspect 24: The method according to any one of aspects 18-23, wherein the compressed CSI is based at least in part on a singular value decomposition (SVD) associated with the channel information.

[0205] Aspect 25: The method according to aspect 24, wherein the compressed CSI replaces at least a precoding matrix indicator payload and a W1 codebook payload in a CSI report.

[0206] Aspect 26: The method according to any one of aspects 18-25, wherein the compressed CSI replaces at least a precoding matrix indicator payload in a CSI report.

[0207] Aspect 27: A method according to any one of Aspects 18-26, wherein the neural network model is an artificial neural network decoder.

[0208] Aspect 28: The method according to any one of aspects 18-27 further comprises: sending information indicating a maximum payload size, wherein the compressed CSI is based at least in part on the maximum payload size.

[0209] Aspect 29: The method according to any one of Aspects 18-28 further includes: sending a parameter set associated with generating the compressed CSI for a neural network encoder.

[0210] Aspect 30: An apparatus for wireless communication at a device, comprising: a processor; a memory coupled to the processor; and instructions stored in the memory and executable by the processor to cause the apparatus to perform the method according to one or more of Aspects 1 to 29.

[0211] Aspect 31: A device for wireless communication, comprising a memory and one or more processors coupled to the memory, wherein the one or more processors are configured to execute the method according to one or more aspects of Aspect 1 to Aspect 29.

[0212] Aspect 32: An apparatus for wireless communication, comprising at least one unit for executing the method according to one or more aspects of Aspect 1-Aspect 29.

[0213] Aspect 33: A non-transitory computer-readable medium storing a code for wireless communication, the code comprising instructions executable by a processor to perform the method according to one or more aspects of aspects 1 to 29.

[0214] Aspect 34: A non-transitory computer-readable medium storing an instruction set for wireless communication, the instruction set comprising one or more instructions, which, when executed by one or more processors of a device, causes the device to perform a method according to one or more aspects of Aspect 1-Aspect 29.

[0215] The foregoing disclosure provides illustration and description, but is not intended to be exhaustive or to limit the aspects to the precise forms disclosed. Modifications and variations are possible in light of the above disclosure or may be acquired from practice of the various aspects.

[0216] As used herein, the term "component" is intended to be broadly interpreted as a combination of hardware and / or hardware and software. Whether referred to as software, firmware, middleware, microcode, hardware description language or other names, "software" should be broadly interpreted as meaning instructions, instruction sets, codes, code segments, program codes, programs, subroutines, software modules, applications, software applications, software packages, routines, subroutines, objects, executable files, threads of execution, processes and / or functions, etc. As used herein, a "processor" is implemented with a combination of hardware and / or hardware and software. It will be apparent that the system and / or method described herein can be implemented with a combination of hardware and / or hardware and software in different forms. The actual dedicated control hardware or software code for implementing these systems and / or methods is not limited to these aspects. Therefore, the operation and behavior of the system and / or method are described herein without citing a specific software code, because those skilled in the art will understand that software and hardware can be designed to implement the system and / or method based at least in part on the description herein.

[0217] As used herein, "satisfying a threshold" may refer to a value being greater than a threshold, greater than or equal to a threshold, less than a threshold, less than or equal to a threshold, equal to a threshold, not equal to a threshold, etc., depending on the context.

[0218] Although the specific combination of features is recorded in the claims and / or disclosed in the specification, these combinations are not intended to limit the disclosure of various aspects. Many features in these features can be combined in a manner not recorded in the claims and / or not disclosed in the specification. The disclosure of various aspects includes each dependent claim combined with each other claim in the claim set. As used herein, the phrase "at least one" related to the project list refers to any combination of these projects, including single members. For 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 and any combination with the multiple of the same element (for example, 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 sorting of a, b and c).

[0219] Elements, actions or instructions used herein should not be interpreted as key or necessary, unless clearly described as such. In addition, as used herein, the articles "a" and "an" are intended to include one or more projects, and can be used interchangeably with "one or more". In addition, as used herein, the article "the" is intended to include one or more projects quoted in conjunction with the article "the", and can be used interchangeably with "one or more". In addition, as used herein, the terms "set" and "group" are intended to include one or more projects, and can be used interchangeably with "one or more". In the case of being intended for only one project, "only one" or similar language is used. In addition, as used herein, the terms "has", "have", "having" etc. are intended to be open terms, which do not limit the elements (for example, "having" A elements can also have B) that they modify. In addition, the phrase "based on" is intended to mean "based on at least in part", unless otherwise explicitly stated. Furthermore, as used herein, the term "or" when used in series is intended to be inclusive and may be used interchangeably with "and / or" unless expressly stated otherwise (eg, if used in combination with "each" or "only one of").

Claims

1. A user equipment (UE) for wireless communication, comprising: Memory; as well as One or more processors coupled to the memory, configured to: identifying channel information based at least in part on a set of channel state information reference signals (CSI-RS); generating compressed channel state information (CSI) using a neural network model, wherein an input to the neural network model is based at least in part on the channel information, and wherein the compressed CSI is compressed in the frequency domain; and The compressed CSI is transmitted.

2. The UE according to claim 1, wherein: The channel information is based at least in part on a channel matrix representing the channel in the frequency domain.

3. The UE according to claim 1, wherein: The input to the neural network model is based at least in part on an average of the channel information across a set of resource blocks (RBs).

4. The UE according to claim 3, wherein: The one or more processors are further configured to: An indication of a granularity for indicating a number of RBs included in the set of RBs is received.

5. The UE according to claim 3, wherein: The compressed CSI replaces a precoding matrix indicator payload, a channel quality indicator payload, a rank indicator payload, and a W1 codebook payload in a CSI report to be sent by the UE.

6. The UE according to claim 1, wherein: The input to the neural network model is based at least in part on a projection of a channel matrix, wherein the projection is onto a set of beams of a codebook.

7. The UE according to claim 1, wherein: The input to the neural network model is based at least in part on a singular value decomposition (SVD) associated with the channel information.

8. The UE according to claim 7, wherein: The input to the neural network model includes one or more singular values ​​of the SVD associated with the channel information.

9. The UE according to claim 7, wherein: The input to the neural network model includes one or more singular vectors of the SVD associated with the channel information.

10. The UE according to claim 9, wherein: The one or more singular vectors of the SVD are right singular vectors.

11. The UE according to claim 7, wherein: The compressed CSI replaces at least a precoding matrix indicator payload and a W1 codebook payload in a CSI report to be sent by the UE.

12. The UE according to claim 1, wherein: The compressed CSI replaces at least a precoding matrix indicator payload in a CSI report to be sent by the UE.

13. The UE according to claim 1, wherein: The neural network model is an artificial neural network encoder.

14. The UE according to claim 1, wherein: The one or more processors are further configured to: Information indicating a maximum payload size is received, wherein generating the compressed CSI is based at least in part on the maximum payload size.

15. The UE according to claim 1, wherein: The one or more processors are further configured to: A set of parameters for the neural network model is received, wherein generating the compressed CSI is based at least in part on the set of parameters.

16. The UE according to claim 1, wherein: A compression scheme for the compressed CSI is based at least in part on a number of layers associated with the compressed CSI.

17. The UE according to claim 1, wherein: The input to the neural network model is based at least in part on a matrix containing one or more frequency domain bases derived from the channel information.

18. A network node for wireless communication, comprising: Memory; as well as One or more processors coupled to the memory, configured to: receiving compressed channel state information (CSI) compressed in the frequency domain; Decompressing the compressed CSI using a neural network model to obtain channel information; as well as Communications are configured based at least in part on the channel information.

19. The network node according to claim 18, wherein: The channel information is based at least in part on a channel matrix representing the channel in the frequency domain.

20. The network node according to claim 18, wherein: The compressed CSI is based at least in part on an average of the channel information across a set of resource blocks (RBs).

21. The network node according to claim 20, wherein: The one or more processors are further configured to: An indication of a granularity for indicating the number of RBs included in the RB set is sent.

22. The network node according to claim 20, wherein: The compressed CSI replaces the precoding matrix indicator payload, the channel quality indicator payload, the rank indicator payload, and the W1 codebook payload in the CSI report.

23. The network node according to claim 18, wherein: The compressed CSI is based at least in part on a projection of a channel matrix, wherein the projection is onto a set of beams of a codebook.

24. The network node according to claim 18, wherein: The compressed CSI is based at least in part on a singular value decomposition (SVD) associated with the channel information.

25. The network node according to claim 18, wherein: The compressed CSI replaces at least the precoding matrix indicator payload in the CSI report.

26. The network node according to claim 18, wherein: The neural network model is an artificial neural network decoder.

27. The network node according to claim 18, wherein: The one or more processors are further configured to: Information indicating a maximum payload size is sent, wherein the compressed CSI is based at least in part on the maximum payload size.

28. The network node according to claim 18, wherein: The one or more processors are further configured to: A set of parameters associated with generating the compressed CSI is sent for a neural network encoder.

29. A method of wireless communication performed by a user equipment (UE), comprising: identifying channel information based at least in part on a set of channel state information reference signals (CSI-RS); generating compressed channel state information (CSI) using a neural network model, wherein an input to the neural network model is based at least in part on the channel information, and wherein the compressed CSI is compressed in the frequency domain; and The compressed CSI is transmitted.

30. A method of wireless communication performed by a network node, comprising: receiving compressed channel state information (CSI) compressed in the frequency domain; Decompressing the compressed CSI using a neural network model to obtain channel information; as well as Communications are configured based at least in part on the channel information.

Images