Model updating for channel state information compression
The method of transmitting updated parameters between model sides addresses the challenge of updating two-sided AI/ML-based CSI compression models, enhancing adaptability and reducing overhead in wireless communication systems.
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- LENOVO (BEIJING) LTD
- Filing Date
- 2025-10-03
- Publication Date
- 2026-07-30
AI Technical Summary
Existing wireless communication systems face challenges in efficiently updating and tuning two-sided models for AI/ML-based channel state information (CSI) compression, necessitating improved methods for model updating to adapt to varying application scenarios and system configurations.
A method for transmitting updated parameters from one side of a two-sided model to the other, enabling model updating and tuning for AI/ML-based CSI compression, utilizing low-rank adaptation (LoRA) and supporting efficient parameter transfer with low overhead.
Enables effective adaptation of CSI models to varying scenarios, reducing CSI report overhead and ensuring performance targets are met, facilitating wide deployment and inter-vendor collaboration.
Smart Images

Figure CN2025126337_30072026_PF_FP_ABST
Abstract
Description
MODEL UPDATING FOR CHANNEL STATE INFORMATION COMPRESSIONTECHNICAL FIELD
[0001] The present disclosure relates to wireless communications, and more specifically to methods and apparatuses for model updating for channel state information (CSI) compression.BACKGROUND
[0002] A wireless communications system may include one or multiple network communication devices. Each network communication devices may support wireless communications for one or multiple terminal devices. The wireless communications system may support wireless communications with one or multiple terminal devices by utilizing resources of the wireless communication system (e.g., time resources (e.g., symbols, slots, subframes, frames, or the like) or frequency resources (e.g., subcarriers, carriers) ) . Additionally, the wireless communications system may support wireless communications across various radio access technologies including third generation (3G) radio access technology, fourth generation (4G) radio access technology, fifth generation (5G) radio access technology, other suitable radio access technologies beyond 5G (e.g., sixth generation (6G) ) , Institute of Electrical and Electronics Engineers (IEEE) 802.11 (Wi-Fi) , IEEE 802.16 (WiMAX) , or IEEE 802.20.
[0003] A two-sided model is used for artificial intelligence (AI) / machine learning (ML) -based CSI compression. In this use case, CSI estimated from configured CSI-RS resources is pre-processed (e.g., to derive eigen vectors) , compressed and quantized with an AI / ML model, termed as Encoder (or a CSI generation part) in a user equipment (UE) . In a next-generation NodeB (gNB) , CSI (or eigen vectors) is recovered from a received CSI report via another AI / ML model, termed as Decoder (or a CSI reconstruction part) . How to support model updating / tuning for the two-sided model for AI / ML-based CSI compression needs to be further studied.SUMMARY
[0004] The present disclosure relates to methods and apparatuses that support model updating for CSI compression. By transmitting updated parameters from one side of a two-sided model to the other side, model updating for AI / ML-based CSI compression with a two-sided model is supported.
[0005] In the context of the present disclosure, an apparatus may be implemented as a network entity or UE, or a part of the network entity or UE. In some implementations, the apparatus may be implemented as a processor at the network entity or UE.
[0006] In one aspect, some implementations of a UE described herein may include a processor; and a transceiver coupled to the processor. The processor is configured to: receive, from a base station via the transceiver, first information of at least one set of updated parameters of at least one layer or block of a first model for CSI generation; and update the first model based on the first information.
[0007] Some implementations of a method performed at a UE described herein may comprise: receiving, from a base station, first information of at least one set of updated parameters of at least one layer or block of a first model for CSI generation; and updating the first model based on the first information.
[0008] Some implementations of a processor described herein may include at least one memory and a controller. The controller is coupled with the at least one memory and configured to cause the controller to: receive, from a base station, first information of at least one set of updated parameters of at least one layer or block of a first model for CSI generation; and update the first model based on the first information.
[0009] In some implementations, the at least one set of updated parameters may correspond to a second model for CSI reconstruction. In some implementations, the first model may be a part of a two-sided model applied in the UE, and the second model may be another part of the two-sided model applied in the base station.
[0010] In some implementations, the first information may comprise at least one of the following: an index indicating the at least one set of updated parameters, at least one set of quantized updated parameters, or an indication of the at least one layer or block of the first model.
[0011] In some implementations, the index may be an identity corresponding to an applicable scenario.
[0012] In some implementations, the at least one set of updated parameters may be represented by matrices or vectors generated by a low-rank adaptation (LoRA) tuning scheme.
[0013] In some implementations, the UE may transmit, to the base station, an acknowledgement on the updating of the first model.
[0014] In some implementations, the UE may receive, from the base station, a request for data collection for model updating, wherein the request comprises an identity corresponding to an applicable scenario; and transmit, to the base station, collected data for the applicable scenario.
[0015] In some implementations, the request may be involved in a CSI report configuration.
[0016] In some implementations, the UE may transmit, to the base station, a request for data collection, wherein the data collection is for monitoring a condition for triggering model updating; receive, from the base station, at least one CSI reference signal resource set for data collection; determine whether to trigger model updating based on collected data; and transmit, to the base station, an indication of triggering model updating based on determining to trigger model updating.
[0017] In some implementations, the UE may receive, from a base station or a core network, second information of the first model, wherein the second information comprises at least one of the following: parameters of the first model, a reference model, an indication of whether the first model is updatable, or an indication of at least one layer or block which includes at least one set of updatable parameters.
[0018] In some implementations, the second information may be transmitted via an access stratum protocol layer dedicated for the second information, or a protocol data unit (PDU) session based on an Internet protocol (IP) address.
[0019] In another aspect, some implementations of a base station described herein may include a processor; and a transceiver coupled to the processor. The processor is configured to: update a second model for CSI reconstruction; determine first information of at least one updated parameter of at least one layer or block of a first model for CSI generation based on the updated second model; and transmit the first information to a UE via the transceiver.
[0020] Some implementations of a method performed at a base station described herein may comprise: updating a second model for CSI reconstruction; determining first information of at least one updated parameter of at least one layer or block of a first model for CSI generation based on the updated second model; and transmitting the first information to a UE.
[0021] Some implementations of a processor described herein may include at least one memory and a controller. The controller is coupled with the at least one memory and configured to cause the controller to: update a second model for CSI reconstruction; determine first information of at least one updated parameter of at least one layer or block of a first model for CSI generation based on the updated second model; and transmit the first information to a UE.
[0022] In some implementations, the at least one updated parameter corresponds to the second model, and wherein the first model is a part of a two-sided model applied in the UE, and the second model is another part of the two-sided model applied in the base station.
[0023] In some implementations, the first information comprises at least one of the following: an index indicating the at least one updated parameter, the at least one updated parameter, or an indication of the at least one layer or block.
[0024] In some implementations, the index corresponds to an applicable scenario.
[0025] In some implementations, the at least one updated parameter is represented by matrices or vectors generated by a LoRA tuning scheme.
[0026] In some implementations, the base station may receive, from the UE, an acknowledgement on updating of the first model.
[0027] In some implementations, the base station may transmit, to the UE, a request for data collection for model updating, wherein the request comprises an index corresponding to an applicable scenario; receive, from the UE, collected data for the applicable scenario; and determine whether to trigger the model updating based on the collected data.
[0028] In some implementations, the base station may generate the first information with the collected data based on determining to trigger the model updating.
[0029] In some implementations, the base station may receive, from the UE, a request for data collection, wherein the data collection is for monitoring a condition for triggering model updating; and transmit, to the UE, at least one CSI reference signal resource for data collection.
[0030] In some implementations, the base station may receive, from the UE, an indication of triggering model updating.
[0031] In some implementations, the base station may transmit, to the UE, second information of the first model, wherein the second information comprises at least one of the following: parameters of the first model, a reference model, an indication of whether the first model is updatable, or an indication of at least one layer or block which includes at least one updatable parameter.
[0032] In some implementations, the second information is transmitted via an access stratum protocol layer dedicated for the second information, or a PDU session based on IP address.
[0033] In some implementations, the base station may determine the first information by:determining the first information based on an applicable scenario.
[0034] It is to be understood that the summary section is not intended to identify key or essential features of embodiments of the present disclosure, nor is it intended to be used to limit the scope of the present disclosure. Other features of the present disclosure will become easily comprehensible through the following description.BRIEF DESCRIPTION OF THE DRAWINGS
[0035] Fig. 1 illustrates an example of a wireless communications system that supports model updating for CSI compression in accordance with aspects of the present disclosure;
[0036] Fig. 2 illustrates a diagram illustrating an AI / ML-based CSI compression scheme with a two-sided model in which embodiments of the present disclosure may be implemented;
[0037] Fig. 3 illustrates a diagram illustrating a general AI / ML model description in which embodiments of the present disclosure may be implemented;
[0038] Fig. 4 illustrates a diagram illustrating an example procedure for CSI compression model in which embodiments of the present disclosure may be implemented;
[0039] Fig. 5 illustrates a diagram illustrating another example procedure for CSI compression model in which embodiments of the present disclosure may be implemented;
[0040] Fig. 6 illustrates a diagram illustrating another example procedure for CSI compression model in which embodiments of the present disclosure may be implemented;
[0041] Fig. 7 illustrates a signaling diagram illustrating an example process that supports model updating for CSI compression in accordance with aspects of the present disclosure;
[0042] Fig. 8 illustrates a diagram illustrating an example tuneable model in accordance with aspects of the present disclosure;
[0043] Fig. 9 illustrates a signaling diagram illustrating an example process of base station triggering model updating in accordance with aspects of the present disclosure;
[0044] Fig. 10 illustrates a signaling diagram illustrating an example process of UE triggering model updating in accordance with aspects of the present disclosure;
[0045] Fig. 11 illustrates a signaling diagram illustrating an example process of model updating in accordance with aspects of the present disclosure;
[0046] Fig. 12 illustrates an example of a device that supports model updating for CSI compression in accordance with some aspects of the present disclosure;
[0047] Fig. 13 illustrates an example of a processor that supports model updating for CSI compression in accordance with some aspects of the present disclosure; and
[0048] Figs. 14 and 15 illustrate a flowchart of an example method that supports model updating for CSI compression in accordance with aspects of the present disclosure, respectively.DETAILED DESCRIPTION
[0049] Principles of the present disclosure will now be described with reference to some embodiments. It is to be understood that these embodiments are described only for the purpose of illustration and help those skilled in the art to understand and implement the present disclosure, without suggesting any limitation as to the scope of the disclosure. The disclosure described herein may be implemented in various manners other than the ones described less than or equal to.
[0050] In the following description and claims, unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skills in the art to which this disclosure belongs.
[0051] References in the present disclosure to “one embodiment, ” “an example embodiment, ” “an embodiment, ” “some embodiments, ” and the like indicate that the embodiment (s) described may include a particular feature, structure, or characteristic, but it is not necessary that every embodiment includes the particular feature, structure, or characteristic. Moreover, such phrases do not necessarily refer to the same embodiment (s) . Further, when a particular feature, structure, or characteristic is described in connection with an embodiment, it is submitted that it is within the knowledge of one skilled in the art to affect such feature, structure, or characteristic in connection with other embodiments whether or not explicitly described.
[0052] It shall be understood that although the terms “first” and “second” or the like may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, a first element could also be termed as a second element, and similarly, a second element could also be termed as a first element, without departing from the scope of embodiments. As used herein, the term “and / or” includes any and all combinations of one or more of the listed terms.
[0053] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments. As used herein, the singular forms “a” , “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “comprises” , “comprising” , “has” , “having” , “includes” and / or “including” , when used herein, specify the presence of stated features, elements, and / or components etc., but do not preclude the presence or addition of one or more other features, elements, components and / or combinations thereof.
[0054] As mentioned above, how to support model updating / tuning for the two-sided model for AI / ML-based CSI compression needs to be further studied.
[0055] Thus, the present disclosure provides solutions that support model updating for CSI compression. In one aspect, a UE may receive, from a base station, first information of at least one set of updated parameters of at least one layer or block of a first model for CSI generation. The UE may update the first model based on the first information. As such, model updating for AI / ML-based CSI compression with a two-sided model is supported.
[0056] Aspects of the present disclosure are described in the context of a wireless communications system.
[0057] Fig. 1 illustrates an example of a wireless communications system 100 that supports model updating for CSI compression in accordance with aspects of the present disclosure. The wireless communications system 100 may include one or more network entities (also referred to as network equipment (NE) ) . For convenience, network entities 102-1, 102-2 and 102-3 are shown and are collectively referred to as one or more network entities 102 hereinafter. The wireless communications system 100 may further include one or more terminal devices 104, a core network 106, and a packet data network 108. The wireless communications system 100 may support various radio access technologies. In some implementations, the wireless communications system 100 may be a 4G network, such as an LTE network or an LTE-Advanced (LTE-A) network. In some other implementations, the wireless communications system 100 may be a 5G network, such as an NR network. In other implementations, the wireless communications system 100 may be a combination of a 4G network and a 5G network, or other suitable radio access technology including Institute of Electrical and Electronics Engineers (IEEE) 802.11 (Wi-Fi) , IEEE 802.16 (WiMAX) , IEEE 802.20. The wireless communications system 100 may support radio access technologies beyond 5G. Additionally, the wireless communications system 100 may support technologies, such as time division multiple access (TDMA) , frequency division multiple access (FDMA) , or code division multiple access (CDMA) , etc.
[0058] The one or more network entities 102 may be dispersed throughout a geographic region to form the wireless communications system 100. One or more of the network entities 102 described herein may be or include or may be referred to as a network device, a network node, a base station, a network element, a radio access network (RAN) , a base transceiver station, an access point, a NodeB, an eNodeB (eNB) , a next-generation NodeB (gNB) , or other suitable terminology. A network entity 102 and a terminal device 104 may communicate via a communication link 110, which may be a wireless or wired connection. For example, a network entity 102 and a terminal device 104 may perform wireless communication (e.g., receive signaling, transmit signaling) over a Uu interface. The one or more network entities 102 may be collectively referred to as network entities 102 or individually referred to as a network entity 102.
[0059] A network entity 102 may provide one or more geographic coverage areas (also referred to as cells) for which the network entity 102 may support services (e.g., voice, video, packet data, messaging, broadcast, etc. ) for one or more terminal devices 104 within a geographic coverage area. For example, a network entity 102 and a terminal device 104 may support wireless communication of signals related to services (e.g., voice, video, packet data, messaging, broadcast, etc. ) according to one or multiple radio access technologies. In some implementations, a network entity 102 may be moveable, for example, a satellite associated with a non-terrestrial network. In some implementations, different geographic coverage areas associated with the same or different radio access technologies may overlap, but the different geographic coverage areas may be associated with different network entities 102. Information and signals described herein may be represented using any of a variety of different technologies and techniques. For example, data, instructions, commands, information, signals, bits, symbols, and chips that may be referenced throughout the description may be represented by voltages, currents, electromagnetic waves, magnetic fields or particles, optical fields or particles, or any combination thereof.
[0060] The one or more terminal devices 104 may be dispersed throughout a geographic region of the wireless communications system 100. A terminal device 104 may include or may be referred to as a UE, a mobile device, a wireless device, a remote device, a remote unit, a handheld device, or a subscriber device, or some other suitable terminology. In some implementations, the terminal device 104 may be referred to as a unit, a station, a terminal, or a client, among other examples. Additionally, or alternatively, the terminal device 104 may be referred to as an IoT device, an Internet-of-Everything (IoE) device, or machine-type communication (MTC) device, among other examples. In some implementations, a terminal device 104 may be stationary in the wireless communications system 100. In some other implementations, a terminal device 104 may be mobile in the wireless communications system 100.
[0061] The one or more terminal devices 104 may be devices in different forms or having different capabilities. Some examples of terminal devices 104 are illustrated in Fig. 1. A terminal device 104 may be capable of communicating with various types of devices, such as the network entities 102, other terminal devices 104, or network equipment (e.g., the core network 106, the packet data network 108, a relay device, an integrated access and backhaul (IAB) node, or another network equipment) , as shown in Fig. 1. Additionally, or alternatively, a terminal device 104 may support communication with other network entities 102 or terminal devices 104, which may act as relays in the wireless communications system 100.
[0062] A terminal device 104 may also be able to support wireless communication directly with other terminal devices 104 over a communication link 114. For example, a terminal device 104 may support wireless communication directly with another terminal device 104 over a device-to-device (D2D) communication link. In some implementations, such as vehicle-to-vehicle (V2V) deployments, vehicle-to-everything (V2X) deployments, or cellular-V2X deployments, the communication link 114 may be referred to as a sidelink. For example, a terminal device 104 may support wireless communication directly with another terminal device 104 over a PC5 interface.
[0063] A network entity 102 may support communications with the core network 106, or with another network entity 102, or both. For example, a network entity 102 may interface with the core network 106 through one or more backhaul links 116 (e.g., via an S1, N2, N2, or another network interface) . The network entities 102 may communicate with each other over the backhaul links 116 (e.g., via an X2, Xn, or another network interface) . In some implementations, the network entities 102 may communicate with each other directly (e.g., between the network entities 102) . In some other implementations, the network entities 102 may communicate with each other or indirectly (e.g., via the core network 106) . In some implementations, one or more network entities 102 may include subcomponents, such as an access network entity, which may be an example of an access node controller (ANC) . An ANC may communicate with the one or more terminal devices 104 through one or more other access network transmission entities, which may be referred to as a radio heads, smart radio heads, or transmission-reception points (TRPs) .
[0064] As an example, the network entity 102-1 may provide a cell 112-1 and the network entity 102-2 may provide a cell 112-2. It is to be understood that each of the network entities 102-1 and 102-2 may provide more cells (not shown) .
[0065] In an example, the network entity may be a satellite, for example, the network entity 102-3. The network entity 102-3 may have full or part of an eNB / gNB on board. The communication link 110 between the network entity 102-3 and the terminal device 104, the communication link 116 between the network entity 102-3 and the network entity 102-2, and the communication link 116 between the network entity 102-2 and the core network 106 may be used for an NTN transparent mode. The communication link 110 between the satellite 102-3 and the terminal device 104, and the communication link 116 between the network entity 102-3 (e.g., with a base station on board) and the core network 106 may be used for a NTN regenerative mode.
[0066] In some implementations, a network entity 102 may be configured in a disaggregated architecture, which may be configured to utilize a protocol stack physically or logically distributed among two or more network entities 102, such as an integrated access backhaul (IAB) network, an open RAN (O-RAN) (e.g., a network configuration sponsored by the O-RAN Alliance) , or a virtualized RAN (vRAN) (e.g., a cloud RAN (C-RAN) ) . For example, a network entity 102 may include one or more of a central unit (CU) , a distributed unit (DU) , a radio unit (RU) , a RAN intelligent controller (RIC) (e.g., a near-real time RIC (Near-RT RIC) , a non-real time RIC (Non-RT RIC) ) , a service management and orchestration (SMO) system, or any combination thereof.
[0067] An RU may also be referred to as a radio head, a smart radio head, a remote radio head (RRH) , a remote radio unit (RRU) , or a TRP. One or more components of the network entities 102 in a disaggregated RAN architecture may be co-located, or one or more components of the network entities 102 may be located in distributed locations (e.g., separate physical locations) . In some implementations, one or more network entities 102 of a disaggregated RAN architecture may be implemented as virtual units (e.g., a virtual CU (VCU) , a virtual DU (VDU) , a virtual RU (VRU) ) .
[0068] Split of functionality between a CU, a DU, and an RU may be flexible and may support different functionalities depending upon which functions (e.g., network layer functions, protocol layer functions, baseband functions, radio frequency functions, and any combinations thereof) are performed at a CU, a DU, or an RU. For example, a functional split of a protocol stack may be employed between a CU and a DU such that the CU may support one or more layers of the protocol stack and the DU may support one or more different layers of the protocol stack. In some implementations, the CU may host upper protocol layer (e.g., a layer 3 (L3) , a layer 2 (L2) ) functionality and signaling (e.g., radio resource control (RRC) , service data adaption protocol (SDAP) , PDCP) . The CU may be connected to one or more DUs or RUs, and the one or more DUs or RUs may host lower protocol layers, such as a layer 1 (L1) (e.g., physical (PHY) layer) or an L2 (e.g., radio link control (RLC) layer, MAC) layer functionality and signaling, and may each be at least partially controlled by the CU 160.
[0069] Additionally, or alternatively, a functional split of the protocol stack may be employed between a DU and an RU such that the DU may support one or more layers of the protocol stack and the RU may support one or more different layers of the protocol stack. The DU may support one or multiple different cells (e.g., via one or more RUs) . In some implementations, a functional split between a CU and a DU, or between a DU and an RU may be within a protocol layer (e.g., some functions for a protocol layer may be performed by one of a CU, a DU, or an RU, while other functions of the protocol layer are performed by a different one of the CU, the DU, or the RU) .
[0070] A CU may be functionally split further into CU control plane (CU-CP) and CU user plane (CU-UP) functions. A CU may be connected to one or more DUs via a midhaul communication link (e.g., F1, F1-c, F1-u) , and a DU may be connected to one or more RUs via a fronthaul communication link (e.g., open fronthaul (FH) interface) . In some implementations, a midhaul communication link or a fronthaul communication link may be implemented in accordance with an interface (e.g., a channel) between layers of a protocol stack supported by respective network entities 102 that are in communication via such communication links.
[0071] The core network 106 may support user authentication, access authorization, tracking, connectivity, and other access, routing, or mobility functions. The core network 106 may be an evolved packet core (EPC) , or a 5G core (5GC) , which may include a control plane entity that manages access and mobility (e.g., a mobility management entity (MME) , an access and mobility management functions (AMF) ) and a user plane entity that routes packets or interconnects to external networks (e.g., a serving gateway (S-GW) , a packet data network (PDN) gateway (P-GW) , or a user plane function (UPF) ) . In some implementations, the control plane entity may manage non-access stratum (NAS) functions, such as mobility, authentication, and bearer management (e.g., data bearers, signal bearers, etc. ) for the one or more terminal devices 104 served by the one or more network entities 102 associated with the core network 106.
[0072] The core network 106 may communicate with the packet data network 108 over one or more backhaul links 116 (e.g., via an S1, N2, N2, or another network interface) . The packet data network 108 may include an application server 118. In some implementations, one or more terminal devices 104 may communicate with the application server 118. A terminal device 104 may establish a session (e.g., a PDU session, or the like) with the core network 106 via a network entity 102. The core network 106 may route traffic (e.g., control information, data, and the like) between the terminal device 104 and the application server 118 using the established session (e.g., the established PDU session) . The PDU session may be an example of a logical connection between the terminal device 104 and the core network 106 (e.g., one or more network functions of the core network 106) . The core network 106 may comprise a sensing function.
[0073] In the wireless communications system 100, the network entities 102 and the terminal devices 104 may use resources of the wireless communications system 100 (e.g., time resources (e.g., symbols, slots, subframes, frames, or the like) or frequency resources (e.g., subcarriers, carriers) ) to perform various operations (e.g., wireless communications) . In some implementations, the network entities 102 and the terminal devices 104 may support different resource structures. For example, the network entities 102 and the terminal devices 104 may support different frame structures. In some implementations, such as in 4G, the network entities 102 and the terminal devices 104 may support a single frame structure. In some other implementations, such as in 5G and among other suitable radio access technologies, the network entities 102 and the terminal devices 104 may support various frame structures (i.e., multiple frame structures) . The network entities 102 and the terminal devices 104 may support various frame structures based on one or more numerologies.
[0074] One or more numerologies may be supported in the wireless communications system 100, and a numerology may include a subcarrier spacing and a cyclic prefix. A first numerology (e.g., μ=0) may be associated with a first subcarrier spacing (e.g., 15 kHz) and a normal cyclic prefix. In some implementations, the first numerology (e.g., μ=0) associated with the first subcarrier spacing (e.g., 15 kHz) may utilize one slot per subframe. A second numerology (e.g., μ=1) may be associated with a second subcarrier spacing (e.g., 30 kHz) and a normal cyclic prefix. A third numerology (e.g., μ=2) may be associated with a third subcarrier spacing (e.g., 60 kHz) and a normal cyclic prefix or an extended cyclic prefix. A fourth numerology (e.g., μ=3) may be associated with a fourth subcarrier spacing (e.g., 120 kHz) and a normal cyclic prefix. A fifth numerology (e.g., μ=4) may be associated with a fifth subcarrier spacing (e.g., 240 kHz) and a normal cyclic prefix.
[0075] A time interval of a resource (e.g., a communication resource) may be organized according to frames (also referred to as radio frames) . Each frame may have a duration, for example, a 10 millisecond (ms) duration. In some implementations, each frame may include multiple subframes. For example, each frame may include 10 subframes, and each subframe may have a duration, for example, a 1 ms duration. In some implementations, each frame may have the same duration. In some implementations, each subframe of a frame may have the same duration.
[0076] Additionally or alternatively, a time interval of a resource (e.g., a communication resource) may be organized according to slots. For example, a subframe may include a number (e.g., quantity) of slots. The number of slots in each subframe may also depend on the one or more numerologies supported in the wireless communications system 100. For instance, the first, second, third, fourth, and fifth numerologies (i.e., μ=0, μ=1, μ=2, μ=3, μ=4) associated with respective subcarrier spacings of 15 kHz, 30 kHz, 60 kHz, 120 kHz, and 240 kHz may utilize a single slot per subframe, two slots per subframe, four slots per subframe, eight slots per subframe, and 16 slots per subframe, respectively. Each slot may include a number (e.g., quantity) of symbols (e.g., orthogonal frequency division multiplexing (OFDM) symbols) . In some implementations, the number (e.g., quantity) of slots for a subframe may depend on a numerology. For a normal cyclic prefix, a slot may include 14 symbols. For an extended cyclic prefix (e.g., applicable for 60 kHz subcarrier spacing) , a slot may include 12 symbols. The relationship between the number of symbols per slot, the number of slots per subframe, and the number of slots per frame for a normal cyclic prefix and an extended cyclic prefix may depend on a numerology. It should be understood that reference to a first numerology (e.g., μ=0) associated with a first subcarrier spacing (e.g., 15 kHz) may be used interchangeably between subframes and slots.
[0077] In the wireless communications system 100, an electromagnetic (EM) spectrum may be split, based on frequency or wavelength, into various classes, frequency bands, frequency channels, etc. By way of example, the wireless communications system 100 may support one or multiple operating frequency bands, such as frequency range designations FR1 (410 MHz –7.125 GHz) , FR2 (24.25 GHz –52.6 GHz) , FR3 (7.125 GHz –24.25 GHz) , FR4 (52.6 GHz –114.25 GHz) , FR4a or FR4-1 (52.6 GHz –71 GHz) , and FR5 (114.25 GHz –300 GHz) . In some implementations, the network entities 102 and the terminal devices 104 may perform wireless communications over one or more of the operating frequency bands. In some implementations, FR1 may be used by the network entities 102 and the terminal devices 104, among other equipment or devices for cellular communications traffic (e.g., control information, data) . In some implementations, FR2 may be used by the network entities 102 and the terminal devices 104, among other equipment or devices for short-range, high data rate capabilities.
[0078] FR1 may be associated with one or multiple numerologies (e.g., at least three numerologies) . For example, FR1 may be associated with a first numerology (e.g., μ=0) , which includes 15 kHz subcarrier spacing; a second numerology (e.g., μ=1) , which includes 30 kHz subcarrier spacing; and a third numerology (e.g., μ=2) , which includes 60 kHz subcarrier spacing. FR2 may be associated with one or multiple numerologies (e.g., at least 2 numerologies) . For example, FR2 may be associated with a third numerology (e.g., μ=2) , which includes 60 kHz subcarrier spacing; and a fourth numerology (e.g., μ=3) , which includes 120 kHz subcarrier spacing.
[0079] Fig. 2 illustrates a diagram illustrating an AI / ML-based CSI feedback compression scheme 200 with a two-sided model in which embodiments of the present disclosure may be implemented. In new radio (NR) Rel-18 and Rel-19, the AI / ML-based CSI compression based on the two-sided model was fully studied to identify the benefit, i.e., less CSI report overhead than a legacy one. In this use case, the CSI estimated from configured CSI-RS resources is pre-processed (e.g., to derive eigen vectors) , compressed and quantized with an AI / ML model, termed as Encoder (or the CSI generation part) in a UE.In a gNB, the CSI (or the eigen vectors) is recovered from the received CSI report via another AI / ML model, termed as Decoder (or the CSI reconstruction part) .
[0080] To obtain the expected benefit, i.e., less CSI report overhead, it is necessary to well pair the models, Encoder in the UE and Decoder in the gNB, to support inter-vendor collaboration. For that purpose, the transfer issues of model and / or dataset between network (NW) and UE have been studied during the study item.
[0081] Fig. 3 illustrates a diagram 300 illustrating a general AI / ML model description in which embodiments of the present disclosure may be implemented. A general AI / ML model may be described with a combination of model structure and parameters, together with the related dataset to train the model.
[0082] At the end of the study item in Rel-19, it has been agreed to further specify the signaling and procedures to support the AI / ML-based CSI compression in NR Rel-20 for some selected directions, for example, Option 3a-1 and Option 4-1 in Direction A and Direction C which are described in connection with Figs. 4 to 6 below.
[0083] Fig. 4 illustrates a diagram illustrating an example procedure 400 for CSI compression model in which embodiments of the present disclosure may be implemented. For Direction A, Option 3a-1, parameters exchanged from the NW-side and ends at the UE-side are those of the CSI generation part, which are received at the UE or UE-side goes through offline engineering at the UE-side (e.g., UE-side over-the-top (OTT) server) , e.g., potential re-training, re-development of a different model, and / or offline testing.
[0084] Some additional information, if necessary, may be shared from the NW-side to help UE-side offline engineering and provide performance guidance: performance target, dataset or information related to collecting dataset.
[0085] Fig. 5 illustrates a diagram illustrating another example procedure 500 for CSI compression model in which embodiments of the present disclosure may be implemented. For Direction A, Option 4-1, the dataset exchanged from the NW-side and ends at the UE-side consists of (target CSI, CSI feedback) , which are received at the UE or UE-side goes through offline engineering at the UE-side (e.g., UE-side OTT server) , e.g., model training or offline testing.
[0086] Some additional information, if necessary, may be shared from the NW-side to help UE-side offline engineering and provide performance guidance: performance target
[0087] Fig. 6 illustrates a diagram illustrating another example procedure 600 for CSI compression model in which embodiments of the present disclosure may be implemented. For Direction C, fully standardized reference model (s) and parameters with specified CSI generation part and / or CSI reconstruction part are exchanged.
[0088] Note that the applied models, especially the parameters, needs to well adapt with the varying application scenarios, system configuration and even vendors. For the two-sided models applied for AI / ML-based CSI compression scheme, any updating / tuning on the model on one side should indicate the other side for corresponding updating / tuning to guarantee the performance target.
[0089] The concept of ‘pairing ID’ has been introduced to support model pairing to associate some applications, e.g., dataset, scenarios, which is supposed to be a high level identity for the model used in a broad scenario to have some basic performance satisfying the related RAN4 test cases.
[0090] Thus, though a reference model and / or a set of parameters can be defined, it is still necessary to further indicate whether the model can be tuned and what parameters in which layer (s) in the model can be updated / tuned for the better performance. In this way, model updating / tuning for the two-sided model for AI / ML-based CSI compression may be supported.
[0091] To support wide deployment of a reference model, it is expected to update / tune the parameters of the model on one side, which requires the indication and updating the model on the other side as well for pairing.
[0092] If the updated parameters of a model need to be transferred over the air, the overhead of the related data should be low enough in an efficient way.
[0093] A scheme to support the model tuning for the AI / ML-based CSI compression with a two-sided model with low overhead is proposed.
[0094] Fig. 7 illustrates a signaling diagram illustrating an example process 700 that supports model updating for CSI compression in accordance with aspects of the present disclosure. The process 700 may involve the UE 104 and the base station 102 in Fig. 1. For the purpose of discussion, the process 700 will be described with reference to Fig. 1.
[0095] At step 710, the base station 102 may update a second model for CSI reconstruction.
[0096] At step 720, the base station 102 may determine first information of at least one set of updated parameters of at least one layer or block of a first model for CSI generation based on the updated second model.
[0097] At step 730, the base station 102 may transmit the first information to the UE 104.
[0098] At step 740, the UE 104 may update the first model based on the first information.
[0099] In some implementations, at step 750, the UE may transmit, to the base station, an acknowledgement on the updating of the first model.
[0100] In some implementations, the UE 104 may update the first model and transmits updated parameters to the base station 102 for updating the second model. The procedure is similar as the base station 102 transmitting updated parameters to the UE 104.
[0101] In some implementations, the first model may be a part of a two-sided model applied in the UE, and the second model may be another part of the two-sided model applied in the base station. In some implementations, the two-sided model may be defined with an identity, e.g., model ID or pairing ID. The identity of the model recognized by the base station 102 or the core network 106 will not be changed after updating.
[0102] In some implementations, the at least one set of updated parameters may correspond to the second model.
[0103] In some implementations, a set of updated parameters may be identified via a layer indication or a block indication. Fig. 8 illustrates a diagram illustrating an example tuneable model 800 in accordance with aspects of the present disclosure. A tunable layer is identified and parameters of the tunable layer may be updated.
[0104] In some implementations, updating a model may comprise update parameters of either some layers or blocks via replacing (i.e., all new parameters) or overlapping (i.e., the bias of each parameters to add) . In some implementations, updating may also be referred to as tuning.
[0105] In some implementations, the first information may comprise at least one set of quantized updated parameters. The at least one set of updated parameters may be quantized with a pre-defined format, since the parameters are generated remotely.
[0106] The at least one set of quantized updated parameters are transferred to the UE 104 to apply on the indicated tunable layers / blocks, followed by a confirmation message to the base station 102. Then, the time slot to activate the updated models on both sides can be configured for the UE 104 by the base station 102.
[0107] In some implementations, the first information may comprise an index indicating the at least one set of updated parameters. In some implementations, the index may correspond to a pre-defined codebook containing a set of parameters of the indicated tunable layer / block. the parameters for different scenarios are pre-defined in a codebook. Thus, the index, i.e., the temporal ID, can be involved in an RRC signaling or CSI report configuration related with the model updating.
[0108] In some implementations, the set of parameters of the tunable layers / blocks, i.e., the codebook containing the candidate sets of parameters can be sent from the base station 102 or the core network 106 to the UE 104, e.g., via an RRC signaling or a dedicated protocol layer signaling.
[0109] In some implementations, the index may be an identity corresponding to an applicable scenario. For example, the identity may be referred to as a temporary ID. The identity is associated with the model ID.
[0110] The association between the temporal ID and the tunable parameters for different application target scenarios is illustrated in Table 1, including two models identified with ID k and ID q. Model k is tunable, but Model q is not. Table 1
[0111] For either kind of AI / ML model, there is corresponding applicable scenario, which can be identified with a set of measurement (or sensing) results on the deployment and / or propagation environment.
[0112] For example, the propagation environment can be described by a set of ranges of the statistic values of the spatial channel, such as signal to interference and noise ratio (SINR) , delay spread (DS) , Doppler spread, angle spread (AS) , or power radio of the line-of-sight (LoS) and non-line-of-sight (NLoS) paths. For example, Scenario k means the propagation environment with a combination of {SINR= [-10, 20] dB, DS= [3, 20] ns, Doppler spread= [3, 100] Hz, AS= [2, 10] degree} .
[0113] For example, the propagation environment can be described by a geometric range / area identified by a set of positions together with the duration with time stamps.
[0114] The scenarios with smaller granularity of the range defined for the applicable scenario may better adapt with the target scenario as mentioned above. For example, the smaller range of SINR or smaller area.
[0115] It is to be noted that for each target sub-scenario, there are associated tunable parameter set for the indicated tunable layers / blocks.
[0116] When the UE 104 receives the index, the corresponding parameters may be selected to apply on the indicated tunable layers / blocks, followed by the confirmation message to the base station 102. Then, the time slot to activate the updated models on both sides may be configured for the UE 104 by the base station 102.
[0117] In some implementations, the first information may comprise an indication of the at least one layer or block of the first model. It is to be understood that the at least one layer or block of the first model can be flexibly selected and indicated. It is to be noted that the indication on the tunable layers / blocks can be referred to the legacy method used for the AI / ML model, which needs to be defined and transferred via air interface with some new signaling to support the options on the AI / ML model used for the air interface.
[0118] In some implementations, the first information may comprise the model ID of the first model to indicate which model is used.
[0119] In some implementations, the first information may be transmitted via an RRC signaling.
[0120] In some implementations, the at least one set of updated parameters may be represented by matrices or vectors generated by a LoRA tuning scheme, for example, a multiplying of two matrices or vectors.
[0121] For the LoRA tuning scheme, the base station 102 may indicate the LoRA configurations to the UE 104 with at least following information: - a rank of low-rank decomposition matrices (e.g., 1, 2, 4, 8, ... ) , - a scaling factor (alpha) applied to LoRA weights when adding them to base model weights (e.g., 16, 32, 64…) , - a dropout probability applied to the LoRA layers during training (e.g., 0, 0.1) , - a bias value that specifies how (or whether) to apply LoRA to bias terms (e.g., none, all, lora_only) , - which layers in the model to apply LoRA (e.g., q_proj, v_proj, all-linear) .
[0122] An example of such configuration is shown as below:
[0123] In some implementations, the configuration on the updating with LoRA may be defined as the signaling within a life-cycle management (LCM) framework for any deployed AI / ML model.
[0124] Model updating procedure may be triggered by the base station 102 or the UE 104.
[0125] Fig. 9 illustrates a signaling diagram illustrating an example process 900 of base station triggering model updating in accordance with aspects of the present disclosure. The process 900 may involve the UE 104 and the base station 102 in Fig. 1. For the purpose of discussion, the process 900 will be described with reference to Fig. 1.
[0126] At step 910, the base station 102 may transmit, to the UE 104, a request for data collection for model updating. The request may comprise an identity corresponding to an applicable scenario. In some implementations, the request may be involved in a CSI report configuration. In some implementations, the data collection is performed with CSI RS resources.
[0127] At step 920, the base station 102 may transmit CSI RS resources to the UE 104.
[0128] At step 930, the UE 104 may transmit, to the base station 102, collected data for the applicable scenario. In some implementations, the UE 104 may transmit the collected data via a CSI report. In some implementations, the collected data may be associated with the temporary ID.
[0129] In some implementations, the UE 104 may deliver the collected data (i.e., target CSI and CSI feedback) from physical layer to upper layer (e.g., applicable layer) in a streaming manner, e.g., deliver once available.
[0130] At step 940, the base station 102 may determine whether to trigger the model updating based on the collected data. For example, the base station 102 may evaluate the data to decide trigger tuning or not.
[0131] In some implementations, the base station 102 may generate the first information with the collected data based on determining to trigger the model updating. The parameters of the tunable layers / blocks are generated, i.e., re-trained / tuned, by collecting data. If being generated locally, the updated parameters can be deployed directly and inform the other side for updating. If being generated on the other side, the updated parameters need to be indicated and transferred over the air.
[0132] In some implementations, the base station 102 may determine the first information based on an applicable scenario. The candidate parameters of the tunable layers / blocks of a model, including encoder and decoder, is pre-defined for different applicable scenarios. The parameters for an applicable scenario can be identified via the index of a pre-defined codebook, and the index can be regarded as the temporary ID.
[0133] Fig. 10 illustrates a signaling diagram illustrating an example process 1000 of UE triggering model updating in accordance with aspects of the present disclosure. The process 1000 may involve the UE 104 and the base station 102 in Fig. 1. For the purpose of discussion, the process 1000 will be described with reference to Fig. 1.
[0134] At step 1010, the UE 104 may transmit, to the base station 102, a request for data collection.
[0135] At step 1020, the base station 102 may transmit, to the UE 104, a configuration of at least one CSI reference signal resource set for data collection and / or CSI report.
[0136] At step 1030, the base station 102 may transmit, to the UE 104, the at least one CSI reference signal resource set.
[0137] At step 1040, the UE 104 may determine whether to trigger model updating based on collected data. For example, the UE 104 may evaluate the data to decide trigger tuning or not.
[0138] In some implementations, the UE 104 may transmit, to the base station, an indication of triggering model updating based on determining to trigger model updating. The base station 102 may trigger model updating accordingly.
[0139] In some implementations, the indication of triggering model updating may comprise an indication of applicable scenario changing.
[0140] In some implementations, the purpose of the data collection is to collect enough data to identify the applicable scenario or evaluate the performance to decide whether the model tuning is needed or not.
[0141] In some implementations, most of the current data collection procedure defined for the AI / ML-based beam management and CSI prediction in NR Rel-19 can be reused, i.e., to indicate such CSI collection in the enhanced CSI report configuration. There may be some potential enhancements, such as the report on the target CSI with new formats (e.g., higher resolution than legacy, quantization levels) and the configurations (e.g., periodicity, time / frequency resource and UCI format) .
[0142] In some implementations, the purpose of the data collectionis to collect enough data to re-train / tune the model to derive the updated parameters of the tunable layers.
[0143] To guarantee the consistence between the re-training / tuning and inference, the temporary ID need to be assigned and associated with the scenario for data collection, which can be also indicated and involved in the CSI report configuration when requesting.
[0144] In some implementations, the data collection is for monitoring a condition for triggering model updating. In some implementations, the condition for triggering model updating comprises performance of the deployed model, e.g., a squared generalized cosine similariy (SGCS) value. The SGCS between the recovered CSI (i.e., the output of the decoder at the base station 102 or the output of the nominal decoder at the UE 104) and the true target CSI (i.e., calculated and reported from the UE 104) , is calculated to derive the statistic values with a configured duration.
[0145] If the SGCS value satisfies a criterion, e.g., less than a configured threshold / performance target, the model updating procedure can be triggered by the base station 102 or requested by the UE 104. The configurations can be provided and transmitted by the base station 102 to the UE 104 with corresponding signaling together with the parameters transfer (for Option 3a-1 in Direction A) , dataset transfer (for Option 4-1 in Direction A) or reference model transfer (Direction C) as one of the additional information.
[0146] In some implementations, the condition for triggering model updating comprises an applicable scenarios identification. For any reference model defined for the AI / ML-based CSI feedback compression, there would be some associated typical applicable scenarios identified by some identities, e.g., Model IDs / Pairing IDs, which, as explained above, would have large granularity to cover some typical scenarios. To obtain better performance, the smaller granularity on the scenario may be used with the temporal ID, and the scenarios with the smaller granularity can be identified by analyzing the collected data, i.e., CSI.
[0147] If the applicable scenario changing is detected among the smaller granularity, the model tuning procedure is triggered or requested. If the applicable scenario changing is detected among the larger granularity, the model switching procedure is triggered or requested.
[0148] In some implementations, the UE may receive, from the base station 102 or the core network 106, second information of the first model.
[0149] In some implementations, the second information may comprise at least one of the following: parameters of the first model, a reference model, an indication of whether the first model is updatable, or an indication of at least one layer or block which includes at least one set of updatable parameters.
[0150] For example, the indication of whether the first model is updatable may be involved in a model description as the additional information when the model is requested or received by the base station 102 or the UE 104 for deployment if the reference model is transferred or indicated. (e.g., Direction C)
[0151] For another example, the indication may be involved with a transmission of model parameters when the model is requested or received by the base station 102 or the UE 104 for deployment. (e.g., Option 3a-1 in Direction A)
[0152] In some implementations, the second information may be transmitted via an access stratum protocol layer dedicated for the second information, or a PDU session based on an IP address. In some implementations, the access stratum protocol layer is end to end between the base station 102 or the core network 106 and the UE 104.
[0153] Fig. 11 illustrates a signaling diagram illustrating an example process 1100 of model updating in accordance with aspects of the present disclosure. The process 1100 may involve the UE 104, the base station 102, and the core network 106 in Fig. 1. For the purpose of discussion, the process 1100 will be described with reference to Fig. 1.
[0154] As shown in Fig. 11, at step 1110, models on both sides are transferred from the core network 106 to the base station 102 and the UE 104 with the assigned Model ID / Pairing ID. At step 1120, no matter the model can be tunable or not, the parameters or the dataset for training can be exchanged between the base station 102 and the UE 104 based on the ID. At step 1130, if the model can be tunable, the model can be tuned / updated based on the temporal IDs.
[0155] In this way, the parameters of some dedicated or selected layer (s) / block (s) to be updated in a tunable model can be identified by the temporal IDs, associated with different target applicable scenarios. The tunable parameters can be pre-defined within a codebook, where the temporal ID can be used as the index for different set associated with different target application scenarios. The tunable parameters can be generated with the data collected within a target application scenario associated with the temporal ID.
[0156] The model tuning can be triggered by the base station or the UE via monitoring the performance and evaluate the application scenario based on the data collection. The updated parameters of the tunable layers / blocks can be transferred from NW to UE via quantization or pre-defined index of a codebook of the matrices / vectors.
[0157] The parameters to support the model tuning for the AI / ML-based CSI compression can be much less than the traditional scheme to transfer the whole parameters. It is to be understood that the process can be also extended to support any other AI / ML-based use cases with a two-sided model.
[0158] Fig. 12 illustrates an example of a device 1200 for model updating for CSI compression in accordance with aspects of the present disclosure. The device 1200 may be an example of a base station 102 or a UE 104 as described herein. The device 1200 may support wireless communication with one or more base stations 102, UEs 104, or any combination thereof. The device 1200 may include components for bi-directional communications including components for transmitting and receiving communications, such as a processor 1202, a memory 1204, a transceiver 1206, and, optionally, an I / O controller 1208. These components may be in electronic communication or otherwise coupled (e.g., operatively, communicatively, functionally, electronically, electrically) via one or more interfaces (e.g., buses) .
[0159] The processor 1202, the memory 1204, the transceiver 1206, or various combinations thereof or various components thereof may be examples of means for performing various aspects of the present disclosure as described herein. For example, the processor 1202, the memory 1204, the transceiver 1206, or various combinations or components thereof may support a method for performing one or more of the operations described herein.
[0160] In some implementations, the processor 1202, the memory 1204, the transceiver 1206, or various combinations or components thereof may be implemented in hardware (e.g., in communications management circuitry) . The hardware may include a processor, a digital signal processor (DSP) , an application-specific integrated circuit (ASIC) , a field-programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic, discrete hardware components, or any combination thereof configured as or otherwise supporting a means for performing the functions described in the present disclosure. In some implementations, the processor 1202 and the memory 1204 coupled with the processor 1202 may be configured to perform one or more of the functions described herein (e.g., executing, by the processor 1202, instructions stored in the memory 1204) .
[0161] For example, the processor 1202 may support wireless communication at the device 1200 in accordance with examples as disclosed herein. In some implementations where the device 1200 is used to implement a UE (e.g., the UE 104) , the processor 1202 may be configured to operable to support a means for performing the following: receiving, from a base station, first information of at least one updated parameter of at least one layer or block of a first model for CSI generation; and updating the first model based on the first information.
[0162] In some implementations where the device 1200 is used to implement a base station (e.g., the base station 102) , the processor 1202 may be configured to operable to support a means for performing the following: updating a second model for CSI reconstruction; determining first information of at least one updated parameter of at least one layer or block of a first model for CSI generation based on the updated second model; and transmitting the first information to a UE.
[0163] The processor 1202 may include an intelligent hardware device (e.g., a general-purpose processor, a DSP, a CPU, a microcontroller, an ASIC, an FPGA, a programmable logic device, a discrete gate or transistor logic component, a discrete hardware component, or any combination thereof) . In some implementations, the processor 1202 may be configured to operate a memory array using a memory controller. In some other implementations, a memory controller may be integrated into the processor 1202. The processor 1202 may be configured to execute computer-readable instructions stored in a memory (e.g., the memory 1204) to cause the device 1200 to perform various functions of the present disclosure.
[0164] The memory 1204 may include random access memory (RAM) and read-only memory (ROM) . The memory 1204 may store computer-readable, computer-executable code including instructions that, when executed by the processor 1202 cause the device 1200 to perform various functions described herein. The code may be stored in a non-transitory computer-readable medium such as system memory or another type of memory. In some implementations, the code may not be directly executable by the processor 1202 but may cause a computer (e.g., when compiled and executed) to perform functions described herein. In some implementations, the memory 1204 may include, among other things, a basic I / O system (BIOS) which may control basic hardware or software operation such as the interaction with peripheral components or devices.
[0165] The I / O controller 1208 may manage input and output signals for the device 1200. The I / O controller 1208 may also manage peripherals not integrated into the device M02. In some implementations, the I / O controller 1208 may represent a physical connection or port to an external peripheral. In some implementations, the I / O controller 1208 may utilize an operating system such as or another known operating system. In some implementations, the I / O controller 1208 may be implemented as part of a processor, such as the processor 1206. In some implementations, a user may interact with the device 1200 via the I / O controller 1208 or via hardware components controlled by the I / O controller 1208.
[0166] In some implementations, the device 1200 may include a single antenna 1210. However, in some other implementations, the device 1200 may have more than one antenna 1210 (i.e., multiple antennas) , including multiple antenna panels or antenna arrays, which may be capable of concurrently transmitting or receiving multiple wireless transmissions. The transceiver 1206 may communicate bi-directionally, via the one or more antennas 1210, wired, or wireless links as described herein. For example, the transceiver 1206 may represent a wireless transceiver and may communicate bi-directionally with another wireless transceiver. The transceiver 1206 may also include a modem to modulate the packets, to provide the modulated packets to one or more antennas 1210 for transmission, and to demodulate packets received from the one or more antennas 1210. The transceiver 1206 may include one or more transmit chains, one or more receive chains, or a combination thereof.
[0167] A transmit chain may be configured to generate and transmit signals (e.g., control information, data, packets) . The transmit chain may include at least one modulator for modulating data onto a carrier signal, preparing the signal for transmission over a wireless medium. The at least one modulator may be configured to support one or more techniques such as amplitude modulation (AM) , frequency modulation (FM) , or digital modulation schemes like phase-shift keying (PSK) or quadrature amplitude modulation (QAM) . The transmit chain may also include at least one power amplifier configured to amplify the modulated signal to an appropriate power level suitable for transmission over the wireless medium. The transmit chain may also include one or more antennas 1210 for transmitting the amplified signal into the air or wireless medium.
[0168] A receive chain may be configured to receive signals (e.g., control information, data, packets) over a wireless medium. For example, the receive chain may include one or more antennas 1210 for receive the signal over the air or wireless medium. The receive chain may include at least one amplifier (e.g., a low-noise amplifier (LNA) ) configured to amplify the received signal. The receive chain may include at least one demodulator configured to demodulate the receive signal and obtain the transmitted data by reversing the modulation technique applied during transmission of the signal. The receive chain may include at least one decoder for decoding the processing the demodulated signal to receive the transmitted data.
[0169] Fig. 13 illustrates an example of a processor 1300 for model updating for CSI compression in accordance with aspects of the present disclosure. The processor 1300 may be an example of a processor configured to perform various operations in accordance with examples as described herein. The processor 1300 may include a controller 1302 configured to perform various operations in accordance with examples as described herein. The processor 1300 may optionally include at least one memory 1304, such as L1 / L2 / L3 cache. Additionally, or alternatively, the processor 1300 may optionally include one or more arithmetic-logic units (ALUs) 1306. One or more of these components may be in electronic communication or otherwise coupled (e.g., operatively, communicatively, functionally, electronically, electrically) via one or more interfaces (e.g., buses) .
[0170] The processor 1300 may be a processor chipset and include a protocol stack (e.g., a software stack) executed by the processor chipset to perform various operations (e.g., receiving, obtaining, retrieving, transmitting, outputting, forwarding, storing, determining, identifying, accessing, writing, reading) in accordance with examples as described herein. The processor chipset may include one or more cores, one or more caches (e.g., memory local to or included in the processor chipset (e.g., the processor 1300) or other memory (e.g., random access memory (RAM) , read-only memory (ROM) , dynamic RAM (DRAM) , synchronous dynamic RAM (SDRAM) , static RAM (SRAM) , ferroelectric RAM (FeRAM) , magnetic RAM (MRAM) , resistive RAM (RRAM) , flash memory, phase change memory (PCM) , and others) .
[0171] The controller 1302 may be configured to manage and coordinate various operations (e.g., signaling, receiving, obtaining, retrieving, transmitting, outputting, forwarding, storing, determining, identifying, accessing, writing, reading) of the processor 1300 to cause the processor 1300 to support various operations in accordance with examples as described herein. For example, the controller 1302 may operate as a control unit of the processor 1300, generating control signals that manage the operation of various components of the processor 1300. These control signals include enabling or disabling functional units, selecting data paths, initiating memory access, and coordinating timing of operations.
[0172] The controller 1302 may be configured to fetch (e.g., obtain, retrieve, receive) instructions from the memory 1304 and determine subsequent instruction (s) to be executed to cause the processor 1300 to support various operations in accordance with examples as described herein. The controller 1302 may be configured to track memory address of instructions associated with the memory 1304. The controller 1302 may be configured to decode instructions to determine the operation to be performed and the operands involved. For example, the controller 1302 may be configured to interpret the instruction and determine control signals to be output to other components of the processor 1300 to cause the processor 1300 to support various operations in accordance with examples as described herein. Additionally, or alternatively, the controller 1302 may be configured to manage flow of data within the processor 1300. The controller 1302 may be configured to control transfer of data between registers, arithmetic logic units (ALUs) , and other functional units of the processor 1300.
[0173] The memory 1304 may include one or more caches (e.g., memory local to or included in the processor 1300 or other memory, such RAM, ROM, DRAM, SDRAM, SRAM, MRAM, flash memory, etc. In some implementation, the memory 1304 may reside within or on a processor chipset (e.g., local to the processor 1300) . In some other implementations, the memory 1304 may reside external to the processor chipset (e.g., remote to the processor 1300) .
[0174] The memory 1304 may store computer-readable, computer-executable code including instructions that, when executed by the processor 1300, cause the processor 1300 to perform various functions described herein. The code may be stored in a non-transitory computer-readable medium such as system memory or another type of memory. The controller 1302 and / or the processor 1300 may be configured to execute computer-readable instructions stored in the memory 1304 to cause the processor 1300 to perform various functions. For example, the processor 1300 and / or the controller 1302 may be coupled with or to the memory 1304, the processor 1300, the controller 1302, and the memory 1304 may be configured to perform various functions described herein. In some examples, the processor 1300 may include multiple processors and the memory 1304 may include multiple memories. One or more of the multiple processors may be coupled with one or more of the multiple memories, which may, individually or collectively, be configured to perform various functions herein.
[0175] The one or more ALUs 1306 may be configured to support various operations in accordance with examples as described herein. In some implementation, the one or more ALUs 1306 may reside within or on a processor chipset (e.g., the processor 1300) . In some other implementations, the one or more ALUs 1306 may reside external to the processor chipset (e.g., the processor 1300) . One or more ALUs 1306 may perform one or more computations such as addition, subtraction, multiplication, and division on data. For example, one or more ALUs 1306 may receive input operands and an operation code, which determines an operation to be executed. One or more ALUs 1306 be configured with a variety of logical and arithmetic circuits, including adders, subtractors, shifters, and logic gates, to process and manipulate the data according to the operation. Additionally, or alternatively, the one or more ALUs 1306 may support logical operations such as AND, OR, exclusive-OR (XOR) , not-OR (NOR) , and not-AND (NAND) , enabling the one or more ALUs 1306 to handle conditional operations, comparisons, and bitwise operations.
[0176] The processor 1300 may support wireless communication in accordance with examples as disclosed herein. In some implementations where the processor 1300 is implemented at a UE (e.g., the UE 104) , the processor 1300 may be configured to operable to support a means for performing the following: receiving, from a base station, first information of at least one updated parameter of at least one layer or block of a first model for CSI generation; and updating the first model based on the first information.
[0177] In some implementations where the processor 1300 is implemented at a base station (e.g., the base station 102) , the processor 1300 may be configured to operable to support a means for performing the following: updating a second model for CSI reconstruction; determining first information of at least one updated parameter of at least one layer or block of a first model for CSI generation based on the updated second model; and transmitting the first information to a UE.
[0178] Fig. 14 illustrates a flowchart of a method 1400 for model updating for CSI compression in accordance with aspects of the present disclosure. The operations of the method 1400 may be implemented by a device or its components as described herein. For example, the operations of the method 1400 may be performed by the UE 104 as described herein. In some implementations, the device may execute a set of instructions to control the function elements of the device to perform the described functions. Additionally, or alternatively, the device may perform aspects of the described functions using special-purpose hardware.
[0179] At 1410, the method may include receiving, from a base station, first information of at least one updated parameter of at least one layer or block of a first model for CSI generation. The operations of 1410 may be performed in accordance with examples as described herein. In some implementations, aspects of the operations of 1410 may be performed by a device as described with reference to Fig. 1.
[0180] At 1420, the method may include updating the first model based on the first information. The operations of 1420 may be performed in accordance with examples as described herein. In some implementations, aspects of the operations of 1420 may be performed by a device as described with reference to Fig. 1.
[0181] Fig. 15 illustrates a flowchart of a method 1500 for model updating for CSI compression in accordance with aspects of the present disclosure. The operations of the method 1500 may be implemented by a device or its components as described herein. For example, the operations of the method 1500 may be performed by the base station 102 as described herein. In some implementations, the device may execute a set of instructions to control the function elements of the device to perform the described functions. Additionally, or alternatively, the device may perform aspects of the described functions using special-purpose hardware.
[0182] At 1510, the method may include updating a second model for CSI reconstruction. The operations of 1510 may be performed in accordance with examples as described herein. In some implementations, aspects of the operations of 1510 may be performed by a device as described with reference to Fig. 1.
[0183] At 1520, the method may include determining first information of at least one updated parameter of at least one layer or block of a first model for CSI generation based on the updated second model. The operations of 1520 may be performed in accordance with examples as described herein. In some implementations, aspects of the operations of 1520 may be performed by a device as described with reference to Fig. 1.
[0184] At 1530, the method may include transmitting the first information to a UE. The operations of 1530 may be performed in accordance with examples as described herein. In some implementations, aspects of the operations of 1530 may be performed by a device as described with reference to Fig. 1.
[0185] It shall be noted that implementations of the present disclosure which have been described with reference to Figs. 1 to 11 are also applicable to the device 1200, the processor 1300 as well as the methods 1400 and 1500.
[0186] It shall be noted that the methods described herein describes possible implementations, and that the operations and the steps may be rearranged or otherwise modified and that other implementations are possible. Further, aspects from two or more of the methods may be combined.
[0187] The various illustrative blocks and components described in connection with the disclosure herein may be implemented or performed with a general-purpose processor, a DSP, an ASIC, a CPU, an FPGA or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. A general-purpose processor may be a microprocessor, but in the alternative, the processor may be any processor, controller, microcontroller, or state machine. A processor may also be implemented as a combination of computing devices (e.g., a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration.
[0188] The functions described herein may be implemented in hardware, software executed by a processor, firmware, or any combination thereof. If implemented in software executed by a processor, the functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium. Other examples and implementations are within the scope of the disclosure and appended claims. For example, due to the nature of software, functions described herein may be implemented using software executed by a processor, hardware, firmware, hardwiring, or combinations of any of these. Features implementing functions may also be physically located at various positions, including being distributed such that portions of functions are implemented at different physical locations.
[0189] Computer-readable media includes both non-transitory computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. A non-transitory storage medium may be any available medium that may be accessed by a general-purpose or special-purpose computer. By way of example, non-transitory computer-readable media may include RAM, ROM, electrically erasable programmable ROM (EEPROM) , flash memory, compact disk (CD) ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other non-transitory medium that may be used to carry or store desired program code means in the form of instructions or data structures and that may be accessed by a general-purpose or special-purpose computer, or a general-purpose or special-purpose processor.
[0190] As used herein, including in the claims, an article “a” before an element is unrestricted and understood to refer to “at least one” of those elements or “one or more” of those elements. The terms “a, ” “at least one, ” “one or more, ” and “at least one of one or more” may be interchangeable. As used herein, including in the claims, “or” as used in a list of items (e.g., a list of items prefaced by a phrase such as “at least one of” or “one or more of” or “one or both of” ) indicates an inclusive list such that, for example, a list of at least one of A, B, or C means A or B or C or AB or AC or BC or ABC (i.e., A and B and C) . Also, as used herein, the phrase “based on” shall not be construed as a reference to a closed set of conditions. For example, an example step that is described as “based on condition A” may be based on both a condition A and a condition B without departing from the scope of the present disclosure. In other words, as used herein, the phrase “based on” shall be construed in the same manner as the phrase “based at least in part on. Further, as used herein, including in the claims, a “set” may include one or more elements.
[0191] The description herein is provided to enable a person having ordinary skill in the art to make or use the disclosure. Various modifications to the disclosure will be apparent to a person having ordinary skill in the art, and the generic principles defined herein may be applied to other variations without departing from the scope of the disclosure. Thus, the disclosure is not limited to the examples and designs described herein but is to be accorded the broadest scope consistent with the principles and novel features disclosed herein.
Claims
1.A user equipment (UE) , comprising:a processor; anda transceiver coupled to the processor,wherein the processor is configured to:receive, from a base station via the transceiver, first information of at least one set of updated parameters of at least one layer or block of a first model for channel state information (CSI) generation; andupdate the first model based on the first information.2.The UE of claim 1, wherein the at least one set of updated parameters corresponds to a second model for CSI reconstruction, andwherein the first model is a part of a two-sided model applied in the UE, and the second model is another part of the two-sided model applied in the base station.3.The UE of claim 1, wherein the first information comprises at least one of the following:an index indicating the at least one set of updated parameters,at least one set of quantized updated parameters, oran indication of the at least one layer or block of the first model.4.The UE of claim 3, wherein the index is an identity corresponding to an applicable scenario.5.The UE of claim 1, wherein the at least one set of updated parameters is represented by matrices or vectors generated by a low-rank adaptation (LoRA) tuning scheme.6.The UE of claim 1, wherein the processor is further configured to:transmit, to the base station via the transceiver, an acknowledgement on the updating of the first model.7.The UE of claim 1, wherein the processor is further configured to:receive, from the base station via the transceiver, a request for data collection for model updating, wherein the request comprises an identity corresponding to an applicable scenario; andtransmit, to the base station via the transceiver, collected data for the applicable scenario.8.The UE of claim 1, wherein the processor is further configured to:transmit, to the base station via the transceiver, a request for data collection, wherein the data collection is for monitoring a condition for triggering model updating;receive, from the base station via the transceiver, at least one CSI reference signal resource set for data collection;determine whether to trigger model updating based on collected data; andtransmit, to the base station via the transceiver, an indication of triggering model updating based on determining to trigger model updating.9.The UE of claim 1, wherein the processor is further configured to:receive, from the base station or a core network via the transceiver, second information of the first model, wherein the second information comprises at least one of the following:parameters of the first model,a reference model,an indication of whether the first model is updatable, oran indication of at least one layer or block which includes at least one set of updatable parameters.10.The UE of claim 9, wherein the second information is transmitted via an access stratum protocol layer dedicated for the second information, or a protocol data unit (PDU) session based on an Internet protocol (IP) address.11.A base station, comprising:a processor; anda transceiver coupled to the processor,wherein the processor is configured to:update a second model for channel state information (CSI) reconstruction;determine first information of at least one set of updated parameters of at least one layer or block of a first model for CSI generation based on the updated second model; andtransmit the first information to a user equipment (UE) via the transceiver.12.The base station of claim 11, wherein the first information comprises at least one of the following:an index indicating the at least one set of updated parameters,at least one set of quantized updated parameters, oran indication of the at least one layer or block of the first model.13.The base station of claim 11, wherein the processor is further configured to:transmit, to the UE via the transceiver, a request for data collection for model updating, wherein the request comprises an identity corresponding to an applicable scenario;receive, from the UE via the transceiver, collected data for the applicable scenario; anddetermine whether to trigger the model updating based on the collected data.14.The base station of claim 13, wherein the request is involved in a CSI report configuration.15.The base station of claim 11, wherein the processor is further configured to:receive, from the UE via the transceiver, a request for data collection, wherein the data collection is for monitoring a condition for triggering model updating; andtransmit, to the UE via the transceiver, at least one CSI reference signal resource set for data collection.16.The base station of claim 15, wherein the processor is further configured to:receive, from the UE via the transceiver, an indication of triggering model updating.17.The base station of claim 11, wherein the processor is further configured to:transmit, to the UE via the transceiver, second information of the first model, wherein the second information comprises at least one of the following:parameters of the first model,a reference model,an indication of whether the first model is updatable, oran indication of at least one layer or block which includes at least one set of updatable parameters.18.The base station of claim 11, wherein the processor is configured to determine the first information by:determining the first information based on an applicable scenario.19.A processor, comprising:at least one controller coupled with at least one memory, and configured to cause the processor to:receive, from a base station via the transceiver, first information of at least one set of updated parameters of at least one layer or block of a first model for channel state information (CSI) generation; andupdate the first model based on the first information.20.A method performed by a user equipment (UE) , comprising:receiving, from a base station via the transceiver, first information of at least one set of updated parameters of at least one layer or block of a first model for channel state information (CSI) generation; andupdating the first model based on the first information.