Data set transmission
Patent Information
- Application Number
- PCT/CN2025/136881
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2025-11-21
- Publication Date
- 2026-10-01
Smart Images

Figure CN2025136881_01102026_PF_FP_ABST
Abstract
Description
DATA SET TRANSMISSIONTECHNICAL FIELD
[0001] The present disclosure relates to wireless communications, and more specifically to user equipment (UE) , base station and methods supporting data set transmission.BACKGROUND
[0002] A wireless communications system may include one or multiple network communication devices, such as base stations, which may be otherwise known as an eNodeB (eNB) , a next-generation NodeB (gNB) , or other suitable terminology. Each network communication devices, such as a base station may support wireless communications for one or multiple user communication devices, which may be otherwise known as user equipment (UE) , or other suitable terminology. The wireless communications system may support wireless communications with one or multiple user communication 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, among other suitable radio access technologies beyond 5G (e.g., sixth generation (6G) ) .
[0003] For artificial intelligence (AI) based channel state information (CSI) compression, pairs of target CSI and CSI feedback may be transmitted from gNB to UE for UE to train its own encoder or model.
[0004] The CSI feedback can be based on scalar quantization or vector quantization, and the quantization granularity may be different. It is agreed to support 1: M mapping between the target CSI and the CSI feedback. Thus, there is a need to study how to perform the 1: M mapping between target CSI and CSI feedback, how to determine the performance target, etcSUMMARY
[0005] The present disclosure relates to UE, base station and methods supporting data set transmission. With the present disclosure, the model training at the UE side may be achieved.
[0006] Some implementations of a UE described herein may include a processor and a transceiver coupled to the processor, wherein the processor is configured to: receive, via the transceiver from a base station, a data set comprising data samples related to first CSI and second CSI, wherein the first CSI is an input of a model, and the second CSI is an output of the model; receive configurations of the first CSI and the second CSI via the transceiver from the base station; and perform training of the model based on the data set and the configurations of the first CSI and the second CSI.
[0007] In some implementations, the configurations of the first CSI and the second CSI comprise a first configuration of the first CSI, wherein the first configuration of the first CSI comprises at least one of the following: the number of antenna ports; the number of subbands; a rank value; an indication indicating whether format of the first CSI is a precoding matrix or a channel matrix; or a quantization type of the first CSI.
[0008] In some implementations, the configurations of the first CSI and the second CSI comprise a second configuration of the second CSI, wherein the second configuration of the second CSI comprises at least one of the following: the number of an output value of the model, a size of segments of the second CSI, wherein each of the segments of the second CSI comprises one or more output values of the model, a quantization type of each of the segments of the second CSI, or the number of bits for quantization of each of the segments of the second CSI.
[0009] In some implementations, the quantization type of the first CSI or quantization type of each of the segments of the second CSI comprises one of the following: scalar quantization, and vector quantization.
[0010] In some implementations, the quantization type of the first CSI or quantization type of each of the segments of the second CSI comprises the vector quantization. In such implementations, the processor is further configured to: receive, via the transceiver from the base station, a configuration of a codebook or parameters related to the codebook.
[0011] In some implementations, the configurations of the first CSI and the second CSI comprise at least one of the following: a first configuration of the first CSI, one or more second configurations of the second CSI associated with the first configuration of the first CSI, or the number of second configurations of the second CSI associated with the first configuration of the first CSI.
[0012] In some implementations, the data samples comprise a first data sample of the first CSI and one or more second data samples of the second CSI, and mapping between the first data sample and the one or more second data samples is based on association between the first configuration and the one or more second configurations.
[0013] In some implementations, the configurations of the first CSI and the second CSI are configured for each data set.
[0014] In some implementations, the processor is further configured to: receive, via the transceiver from the base station, a configuration of the following: mapping between a first index of the first configuration of the first CSI and the first configuration; and mapping between one or more second indexes of the one or more second configurations of the second CSI and the one or more second configurations. In such implementations, the first index and the one or more second indexes are configured for the data set.
[0015] In some implementations, the processor is further configured to: receive, via the transceiver from the base station, a configuration of mapping between an index and a combination of the first configuration of the first CSI and the one or more second configurations of the second CSI. In such implementations, the index of the combination is configured for the data set.
[0016] In some implementations, the combination indicates the first configuration of the first CSI, the number of the one or more second configurations of the second CSI, and the one or more second configurations of the second CSI.
[0017] In some implementations, the data set comprises a first subset and a second subset; The processor is configured to receive the data set by: receiving the second subset after receiving the first subset. The second subset is configured with an indicator, wherein the indicator indicates whether at least one of the first configuration, the number of the one or more second configurations associated with the first configuration, and the one or more second configurations are same as or different from those in the first subset.
[0018] In some implementations, the data set is configured with an index of the data set, each of the first subset and the second subset is associated with the index of the data set.
[0019] In some implementations, the indicator indicates that at least one of the first configuration, the number of the one or more second configurations associated with the first configuration, the one or more second configurations are same as those in the first subset set. In such implementations, the first configuration and the one or more second configurations are absent from the second subset.
[0020] In some implementations, the indicator indicates that at least one of the first configuration, the number of the one or more second configurations associated with the first configuration, and the one or more second configurations are different from those in the first subset set. In such implementations, at least one of the first configuration and the one or more second configurations are present in the second subset.
[0021] In some implementations, the processor is further configured to: determine a squared generalized cosine similarity (SGCS) or a normalized mean squared error (NMSE) as a performance target of the model.
[0022] In some implementations, the SGCS or NMSE is configured for the data set.
[0023] In some implementations, the SGCS or NMSE is based on at least one of the configurations of the first CSI and the second CSI.
[0024] In some implementations, the data set is associated with a pair identity (ID) for pairing between the model and a second model at the base station.
[0025] In some implementations, the processor is configured to determine the SGCS or NMSE based on a pair ID.
[0026] In some implementations, the data set comprises multiple subsets, and a group of the subsets is associated with a group index.
[0027] In some implementations, the processor is configured to determine the SGCS or NMSE based on the group index.
[0028] In some implementations, the SGCS or NMSE is configured or determined per layer.
[0029] In some implementations, the processor is further configured to: transmit an indication indicating whether the SGCS or NMSE is supported.
[0030] In some implementations, the processor is further configured to: transmit an indication indicating whether the configurations of the first CSI or the second CSI are supported.
[0031] In some implementations, the indication indicates that the configurations of the first CSI and the second CSI are supported. In such implementations, the processor is further configured to: receive, via the transceiver from the base station, a configuration of CSI report.
[0032] In some implementations, the processor is further configured to: after receiving the data set, receive a further data set via the transceiver from the base station; and determine whether the further data set can be combined with the data set for training.
[0033] In some implementations, the processor is configured to determine whether the further data set can be combined with the data set for training of the model: based on whether the further data set and the data set are associated with a same pair ID or not.
[0034] In some implementations, the processor is configured to determine whether the further data set can be combined with the data set for training of the model: based on whether the first configuration is the same as the third configuration or not, and / or whether the second configuration is the same as the fourth configuration or not.
[0035] In some implementations, the processor is further configured to: receive, via the transceiver from the base station, an indicator indicating whether the further data set can be combined with the data set.
[0036] In some implementations, the processor is further configured to: receive an indication of a group index, and the further data set can be combined with a data set associated with the group index.
[0037] In some implementations, the further data set is added to the group and associated with the group index.
[0038] In some implementations, the processor is further configured to: receive a first index of a first data set; receive a second index of a second data set or determine the second index of the second data set based on the first index of the first data set and a first number of data sets; and determine that the further data set can be combined with one or more data sets between the first data set and the second data set.
[0039] In some implementations, each data set is configured with a group index, and data sets with same group index can be combined for training.
[0040] In some implementations, the data set is associated with a first SGCS or a first NMSE as a first performance target of the model. In such implementations, the further data set is associated with a second SGCS or a second NMSE as a second performance target of the model. In such implementations, the processor is further configured to: after combining the further data set with the data set for training of the model, release the data set; or update the first SGCS with the second SGCS or update the first NMSE with the second NMSE.
[0041] In some implementations, the processor is further configured to: after finishing the training of the model, release the data set.
[0042] Some implementations of a base station described herein may include a processor and a transceiver coupled to the processor, wherein the processor is configured to:transmit, via the transceiver to a UE, a data set comprising data samples related to first CSI and second CSI, wherein the first CSI is an input of a model, and the second CSI is an output of the model; and transmit configurations of the first CSI and the second CSI via the transceiver to the UE.
[0043] In some implementations, the configurations of the first CSI and the second CSI comprise a first configuration of the first CSI, wherein the first configuration of the first CSI comprises at least one of the following: the number of antenna ports; the number of subbands; a rank value; an indication indicating whether format of the first CSI is a precoding matrix or a channel matrix; or a quantization type of the first CSI.
[0044] In some implementations, the configurations of the first CSI and the second CSI comprise a second configuration of the second CSI, wherein the second configuration of the second CSI comprises at least one of the following: the number of output values of the model, a size of segments of the second CSI, wherein each of the segments of the second CSI comprises one or more output values of the model, a quantization type of each of the segments of the second CSI, or the number of bits for quantization of each of the segments of the second CSI.
[0045] In some implementations, the quantization type of the first CSI or quantization type of each of the segments of the second CSI comprises one of the following: scalar quantization, and vector quantization.
[0046] In some implementations, the quantization type of the first CSI or quantization type of each of the segments of the second CSI comprises the vector quantization. In such implementations, the processor is further configured to: receive, via the transceiver from the base station, a configuration of a codebook or parameters related to the codebook.
[0047] In some implementations, the configurations of the first CSI and the second CSI comprise at least one of the following: a first configuration of the first CSI, one or more second configurations of the second CSI associated with the first configuration of the first CSI, or the number of second configurations of the second CSI associated with the first configuration of the first CSI.
[0048] In some implementations, the data samples comprise a first data sample of the first CSI and one or more second data samples of the second CSI, and mapping between the first data sample and the one or more second data samples is based on association between the first configuration and the one or more second configurations.
[0049] In some implementations, the configurations of the first CSI and the second CSI are configured for each data set.
[0050] In some implementations, the processor is further configured to: transmit, via the transceiver to the UE, a configuration of the following: mapping between a first index of the first configuration of the first CSI and the first configuration; and mapping between one or more second indexes of the one or more second configurations of the second CSI and the one or more second configurations. In such implementations, the first index and the one or more second indexes are configured for the data set.
[0051] In some implementations, the processor is further configured to: transmit, via the transceiver to the UE, a configuration of mapping between an index and a combination of the first configuration of the first CSI and the one or more second configurations of the second CSI. In such implementations, the index of the combination is configured for the data set.
[0052] In some implementations, the combination indicates the first configuration of the first CSI, the number of the one or more second configurations of the second CSI, and the one or more second configurations of the second CSI.
[0053] In some implementations, the data set comprises a first subset and a second subset. In such implementations, the processor is configured to transmit the data set by: transmitting the second subset after transmitting the first subset. The second subset is configured with an indicator, wherein the indicator indicates whether at least one of the first configuration, the number of the one or more second configurations associated with the first configuration, and the one or more second configurations are same as or different from those in the first subset.
[0054] In some implementations, the data set is configured with an index of the data set, each of the first subset and the second subset is associated with the index of the data set.
[0055] In some implementations, the indicator indicates that at least one of the first configuration, the number of the one or more second configurations associated with the first configuration, the one or more second configurations are same as those in the first subset set. In such implementations, the first configuration and the one or more second configurations are absent from the second subset.
[0056] In some implementations, the indicator indicates that at least one of the first configuration, the number of the one or more second configurations associated with the first configuration, and the one or more second configurations are different from those in the first subset set. In such implementations, at least one of the first configuration and the one or more second configurations are present in the second subset.
[0057] In some implementations, an SGCS or an NMSE is configured for the data set.
[0058] In some implementations, the SGCS or NMSE is based on at least one of the configurations of the first CSI and the second CSI.
[0059] In some implementations, the data set is associated with a pair ID for pairing between the model and a second model at the base station.
[0060] In some implementations, the SGCS or NMSE is determined based on a pair ID.
[0061] In some implementations, the data set comprises multiple subsets, and a group of the subsets is associated with a group index.
[0062] In some implementations, the SGCS or NMSE is determined based on the group index.
[0063] In some implementations, the SGCS or NMSE is configured or determined per layer.
[0064] In some implementations, the processor is further configured to: receive an indication indicating whether the SGCS or NMSE is supported.
[0065] In some implementations, the processor is further configured to: receive an indication indicating whether the configurations of the first CSI or the second CSI are supported.
[0066] In some implementations, the indication indicates that the configurations of the first CSI and the second CSI are supported. In such implementations, the processor is further configured to: transmit, via the transceiver to the UE, a configuration of CSI report.
[0067] In some implementations, the processor is further configured to: after transmitting the data set, transmit a further data set via the transceiver to the UE.
[0068] Some implementations of a method described herein may include: receiving, from a base station, a data set comprising data samples related to first CSI and second CSI, wherein the first CSI is an input of a model, and the second CSI is an output of the model; receiving configurations of the first CSI and the second CSI from the base station; and performing training of the model based on the data set and the configurations of the first CSI and the second CSI.
[0069] Some implementations of a method described herein may include: transmitting, to a UE, a data set comprising data samples related to first CSI and second CSI, wherein the first CSI is an input of a model, and the second CSI is an output of the model; and transmitting configurations of the first CSI and the second CSI to the UE.
[0070] Some implementations of a processor described herein may include at least one memory and a controller coupled with the at least one memory and configured to cause the controller to: receive, via the transceiver from a base station, a data set comprising data samples related to first CSI and second CSI, wherein the first CSI is an input of a model, and the second CSI is an output of the model; receive configurations of the first CSI and the second CSI via the transceiver from the base station; and perform training of the model based on the data set and the configurations of the first CSI and the second CSI.
[0071] Some implementations of a processor described herein may include at least one memory and a controller coupled with the at least one memory and configured to cause the controller to: transmit, via the transceiver to a UE, a data set comprising data samples related to first CSI and second CSI, wherein the first CSI is an input of a model, and the second CSI is an output of the model; and transmit configurations of the first CSI and the second CSI via the transceiver to the UE.
[0072] 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
[0073] Fig. 1 illustrates an example of a wireless communications system that supports data set transmission in accordance with aspects of the present disclosure;
[0074] Fig. 2 illustrates a signaling diagram illustrating an example process that supports data set transmission in accordance with aspects of the present disclosure;
[0075] Figs. 3 to 6 illustrate another example of the configurations of the first CSI or the second CSI in accordance with aspects of the present disclosure, respectively;
[0076] Figs. 7 to 10 illustrate an example of combination of the data set and the new data set for training of the model in accordance with aspects of the present disclosure, respectively;
[0077] Fig. 11 illustrates an example of a device that supports data set transmission in accordance with aspects of the present disclosure;
[0078] Fig. 12 illustrates an example of a processor that supports data set transmission in accordance with aspects of the present disclosure; and
[0079] Figs. 13 and 14 illustrate a flowchart of a method that supports data set transmission in accordance with aspects of the present disclosure, respectively.DETAILED DESCRIPTION
[0080] 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 below.
[0081] 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.
[0082] 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.
[0083] 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.
[0084] 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.
[0085] As described above, it is agreed to support 1: M mapping between the target CSI and the CSI feedback. Thus, there is a need to study how to perform the 1: M mapping between target CSI and CSI feedback.
[0086] In view of the above, the present disclosure provides a solution supporting data set transmission. In this solution, a UE receives, from a base station, a data set comprising data samples related to first CSI and second CSI. The first CSI is an input of a model, and the second CSI is an output of the model. The UE also receives configurations of the first CSI and the second CSI from the base station. In turn, the UE performs training of the model based on the data set and the configurations of the first CSI and the second CSI.
[0087] Aspects of the present disclosure are described in the context of a wireless communications system.
[0088] Fig. 1 illustrates an example of a wireless communications system 100 that supports data set transmission in accordance with aspects of the present disclosure. The wireless communications system 100 may include one at least one of network entities 102 (also referred to as network equipment (NE) ) , one or more terminal devices or UEs 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 wires communications system 100 may be a 6G network. 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.
[0089] The network entities 102 may be collectively referred to as network entities 102 or individually referred to as a network entity 102. Hereinafter, some implementations of the present disclosure will be described by taking a base station as an example of the network entity 102. Thus, the network entity 102 may be used interchangeably with the base station 102. For example, base stations 102 may comprise a first base station 102-1 and a second base station 102-2.
[0090] The 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 node, a base station (BS) , a network element, a radio access network (RAN) node, a base transceiver station, an access point, a NodeB, an eNodeB (eNB) , a next-generation NodeB (gNB) , a base station that will be used in 6G, or other suitable terminology. A network entity 102 and a UE 104 may communicate via a communication link 110, which may be a wireless or wired connection. For example, a network entity 102 and a UE 104 may perform wireless communication (e.g., receive signaling, transmit signaling) over a Uu interface. The network entities 102 may be collectively referred to as network entities 102 or individually referred to as a network entity 102.
[0091] A network entity 102 may provide a geographic coverage area 112 for which the network entity 102 may support services (e.g., voice, video, packet data, messaging, broadcast, etc. ) for one or more UEs 104 within the geographic coverage area 112. For example, a network entity 102 and a UE 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 112 associated with the same or different radio access technologies may overlap, but the different geographic coverage areas 112 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.
[0092] The one or more UEs 104 may be dispersed throughout a geographic region of the wireless communications system 100. A UE 104 may include or may be referred to as 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 UE 104 may be referred to as a unit, a station, a terminal, or a client, among other examples. Additionally, or alternatively, the UE 104 may be referred to as an internet-of-things (IoT) device, an internet-of-everything (IoE) device, or machine-type communication (MTC) device, among other examples. In some implementations, a UE 104 may be stationary in the wireless communications system 100. In some other implementations, a UE 104 may be mobile in the wireless communications system 100.
[0093] The one or more UEs 104 may be devices in different forms or having different capabilities. Some examples of UEs 104 are illustrated in Fig. 1. A UE 104 may be capable of communicating with various types of devices, such as the network entities 102, other UEs 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 UE 104 may support communication with other network entities 102 or UEs 104, which may act as relays in the wireless communications system 100.
[0094] A UE 104 may also be able to support wireless communication directly with other UEs 104 over a communication link 114. For example, a UE 104 may support wireless communication directly with another UE 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 UE 104 may support wireless communication directly with another UE 104 over a PC5 interface.
[0095] 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 UEs 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) .
[0096] 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 radio access network (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.
[0097] 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 transmission reception point (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) ) .
[0098] 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) , packet data convergence protocol (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, medium access control (MAC) layer) functionality and signaling, and may each be at least partially controlled by the CU.
[0099] 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) .
[0100] 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.
[0101] 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 UEs 104 served by the one or more network entities 102 associated with the core network 106.
[0102] 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 UEs 104 may communicate with the application server 118. A UE 104 may establish a session (e.g., a protocol data unit (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 UE 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 UE 104 and the core network 106 (e.g., one or more network functions of the core network 106) .
[0103] In the wireless communications system 100, the network entities 102 and the UEs 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 UEs 104 may support different resource structures. For example, the network entities 102 and the UEs 104 may support different frame structures. In some implementations, such as in 4G, the network entities 102 and the UEs 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 UEs 104 may support various frame structures (i.e., multiple frame structures) . The network entities 102 and the UEs 104 may support various frame structures based on one or more numerologies.
[0104] 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.
[0105] 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.
[0106] 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., 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.
[0107] 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 UEs 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 UEs 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 UEs 104, among other equipment or devices for short-range, high data rate capabilities.
[0108] 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.
[0109] Fig. 2 illustrates a signaling diagram illustrating an example process 200 that supports data set transmission in accordance with aspects of the present disclosure. The process 200 may involve the UE 104 and the base station 102 in Fig. 1. For the purpose of discussion, the process 200 will be described with reference to Fig. 1.
[0110] As shown in Fig. 2, the base station 102 transmits 210 a data set to the UE 104. The data set comprises data samples related to first CSI and second CSI. The first CSI is an input of a model at the UE 104, and the second CSI is an output of the model.
[0111] Hereinafter, the first CSI is also referred to as target CSI, and the second CSI is also referred to as CSI feedback.
[0112] In some implementations, the model at the UE 104 may be a CSI compression model. The model at the UE 104 or the CSI compression model is also referred to as an encoder. At the side of the base station 102, there will be a corresponding CSI decompression model, also referred as a decoder. The compression model and the decompression model are linked or paired based on a pair ID.
[0113] In some implementations, the base station 102 may transmit the data set over physical layer, RRC layer or a dedicated protocol layer for exchanging data or model between the UE 104 and the base station 102 over air interface. If the base station 102 transmits the data set over RRC layer, the data set can be in an L2 or L3 message encoded in ASN. 1.
[0114] Alternatively, in some implementations, the base station 102 may transmit the data set over IP as IP packet to UE. For example, the base station 102 may transmit the data set to a CN node, the CN node may transmit the data set to a UPF and then to the UE 104.
[0115] In some implementations, the first CSI can be based on UE measurement with / without combination with sounding reference signal (SRS) measured at the base station 102.
[0116] In some implementations, a format of the first CSI may be a precoding matrix or a channel matrix.
[0117] In some implementations, if the format of the first CSI is the precoding matrix, it will correspond to a vector or a matrix in a codebook.
[0118] In some implementations, if the format of the first CSI is the channel matrix, it can be a raw matrix or based on singular value decomposition (SVD) division of the raw channel matrix based on measurement. The channel matrix is more suitable for combination of SRS based measurement.
[0119] In some implementations, the size of the first CSI is based on a configuration of the number of antenna ports (represented by P) , the number of subbands (represented by N3) , and a rank value (represented by R) . In some implementations, for Release 20 AI based CSI compression, for each layer l, the first CSI is N3 vectors, and the size of the vector is P*1. Therefore, there will be R sets of N3 vectors, then the total size of vectors will be R*N3 for different layer and different subbands, and each vector is associated with all the antenna ports.
[0120] In some implementations, a type of the first CSI may be based on scalar quantization or based on etypeII codebook. If the type of the first CSI is based on scalar quantization, it is to quantize the real / imaginary part or amplitude / phase of a precoding matrix. If the type of the first CSI is based on enhancement of etypeII codebook, it is to use several parameters to calculate a matrix. The type of the first CSI for data set transmission may be configured to be either scalar quantization or based on enhancement of etypeII codebook. No matter which type the first CSI is, the size of the first CSI may be based on the configured antenna port number P, subband number N3 and the rank value R.
[0121] In some implementations, the second CSI may be based on float 32, i.e., each floating value is quantized with 32 bits. Alternatively, the second CSI may be associated with different configurations of the number of output values of the model, the size of segments of the second CSI, and the number of bits for quantization of each of the segments of the second CSI. Each of the segments of the second CSI comprises one or more output value.
[0122] Hereinafter, the number of output values of the model is represented by d, the size of segments of the second CSI is represented by L, and the number of bits for quantization of each of the segments of the second CSI is represented by Q.
[0123] For example, suppose the model output corresponding to a precoding vector as model input is also a vector z, then d is the number of elements of the vector.
[0124] In some implementations, L can be equal to or larger than 1, and it corresponds how to determine different segments of the output vector z.
[0125] In some implementations, if a quantization type of each of the segments of the second CSI is scalar quantization, L is 1; if the quantization type of each of the segments of the second CSI is vector quantization, L can be larger than 1, e.g. 2 or 4. If the quantization type is vector quantization, an index of a codebook or parameters related to the codebook also need to be indicated to the UE 104 so that UE 104 or the base station 102 can perform mapping between the bits and a precoding vector.
[0126] Consequently, there will be different number of bits and different interpretation for different configurations of d, Q and L.
[0127] With continued reference to Fig. 2, the base station 102 transmits 220 configurations of the first CSI and the second CSI to the UE 104.
[0128] In turn, the UE 104 performs 230 training of the model based on the data set and the configurations of the first CSI and the second CSI.
[0129] In some implementations, the configurations of the first CSI and the second CSI may comprise a first configuration of the first CSI. The first configuration of the first CSI may comprise at least one of the following: the number of antenna ports (P) ; the number of subbands (N3) ; a rank value (R) ; an indication indicating whether format of the first CSI is a precoding matrix or a channel matrix; or a quantization type of the first CSI. Hereinafter, a configuration of the first CSI is also referred to as “target CSI configuration” .
[0130] In some implementations, if the format of the first CSI is a raw channel matrix, the number of the second CSI is based on the indicated rank value R or the size of the channel matrix will be indicated.
[0131] In some implementations, the configurations of the first CSI and the second CSI may comprise a second configuration of the second CSI. Hereinafter, a configuration of the second CSI is also referred to as “CSI feedback configuration” .
[0132] In some implementations, the second configuration of the second CSI may comprise at least one of the following: the number of output values of the model (d) ; a size of segments of the second CSI (L) , wherein each of the segments of the second CSI may comprise one or more output values; a quantization type of each of the segments of the second CSI; or the number of bits for quantization of each of the segments of the second CSI (Q) .
[0133] In some implementations, the quantization type of the first CSI or the quantization type of each of the segments of the second CSI may be scalar quantization or vector quantization.
[0134] In some implementations, there may be single or multiple configurations of each of the first CSI and the second CSI.
[0135] In some implementations, if there are multiple configurations of the second CSI, they are associated with a single first configuration of the first CSI. The association can be explicitly configured or implicitly determined if they are in a same data set.
[0136] In some implementations, after the base station 102 transmits the data set to the UE 104, the UE 104 may transmit, to the base station 102, an indication indicating whether the configurations of the first CSI or the second CSI are supported.
[0137] Hereinafter, some examples of the configurations of the first CSI or the second CSI will be described with reference to Figs. 3 to 6.
[0138] Fig. 3 illustrates an example of the configurations of the first CSI or the second CSI in accordance with aspects of the present disclosure. In the example of Fig. 3, the configuration for the data set transmitted from the base station 102 to the UE 104 can be separately configured for each data set transmission.
[0139] In case there are M CSI feedback (i.e., second CSI) configurations corresponding to one target CSI configuration, all the M CSI feedback configurations will be indicated together with the data set. For example, in the example of Fig. 3, there are three CSI feedback configurations corresponding to one target CSI configuration.
[0140] After the configurations of the target CSI and the CSI feedback, there will be one target CSI, M CSI feedbacks corresponding to different CSI feedback configurations and associated with the same target CSI configuration.
[0141] If there are different target CSI configurations, they can be located at different data sets or with different data set indices. That is, a data set containing multiple target CSI samples and CSI feedback samples are associated with a same target CSI configuration.
[0142] If there are not enough data set samples with the same target CSI configuration, 0 can be padded.
[0143] Fig. 4 illustrates another example of the configurations of the first CSI or the second CSI in accordance with aspects of the present disclosure. In the example of Fig. 4, each of the target CSI configuration and the CSI feedback configuration are associated with an index.
[0144] Each first index can be associated with a combination of P, N3, R and the type of the target CSI.
[0145] Each second index can be associated with a combination of d, Q, L or float 32 (i.e., FL32) .
[0146] Each second index can also be associated with multiple combinations of d, Q, L or FL32. For example, as shown in Fig. 4, an index #1 means CSI feedback configuaritons#1 and #2, and an index#3 means CSI feedback configurations#1, #3 and #4.
[0147] The corresponding configuration in Fig. 3 can be replaced with a first index and a second index.
[0148] Mapping between the index and the corresponding configuration of the target CSI and mapping between the index and the corresponding configuration of the CSI feedback or mapping between the index and the combination can be pre-configured.
[0149] Figs. 5 and 6 illustrate an example of the configurations of the first CSI or the second CSI in accordance with aspects of the present disclosure, respectively. In the examples of Figs. 5 and 6, each data set is configured with a pair identity (ID) . The pair ID may be used for pairing between the model at the UE 104 and a model at the base station 102.
[0150] In some implementations, the model at the UE 104 may be a CSI compression model. The model at the UE 104 or the CSI compression model is also referred to as an encoder.
[0151] In some implementations, the model at the base station 102 may be a CSI decompression model. The model at the base station 102 or the CSI decompression model is also referred to as a decoder.
[0152] In some implementations, the data set may comprise a first subset and a second subset. The UE 104 may receive the second subset after receiving the first subset. The data set may also be configured with a subset index.
[0153] The second subset may be configured with an indicator. The indicator indicates whether at least one of the target CSI configuration, the number of the one or more CSI feedback configurations associated with the target CSI configuration, and the one or more CSI feedback configurations are same as or different from those in the first subset set.
[0154] The indicator can also indicate whether at least one of the target CSI configuration, the number of the one or more CSI feedback configurations associated with the target CSI configuration, and the one or more CSI feedback configurations are same as or different from those in the previous data set. In this case, there is no subset, and there are only data sets with different indices.
[0155] In such implementations, the first data subset or data set needs to have an explicit indication of the target CSI configuration and CSI feedback configuration.
[0156] As shown in Fig. 5, for other subset or data set, if the indicator indicates at least one of the target CSI configuration, the number of the one or more CSI feedback configurations associated with the target CSI configuration, and the one or more CSI feedback configurations are same as those in the first subset or data set, the following can be target CSI samples and CSI feedback samples.
[0157] As shown in Fig. 6, for other subset, if the indicator indicates at least one of the target CSI configuration, the number of the one or more CSI feedback configurations associated with the target CSI configuration, and the one or more CSI feedback configurations are different from those in the first subset or data set, there will be followed by the target CSI and CSI feedback detailed configuration or index as in the examples of Figs. 3 and 4.
[0158] In some implementations, the indicator can apply to both target CSI configuration and CSI feedback configuration, or there will be two indicators, and each of the two indicators is for target CSI and CSI feedback respectively. In such implementations, if only one of the target CSI configuration and CSI feedback configuration is changed, there will be only configuration or index of the configuration for the target CSI or CSI feedback respectively.
[0159] In some implementations, the indicator can also indicate that the number of CSI feedback configurations associated with a same target CSI configuration is changed. In such implementations, only CSI feedback configurations will be present in the current data set.
[0160] In some implementations, only when the target CSI configuration or CSI feedback configuration is changed, there will be corresponding configuration in the transmitted data set.
[0161] The subset index is different for different data sets associated with a same pair ID.That is, the subset index is indexed with the same pair ID. When the pair ID is changed, the subset index can be reused or reindexed.
[0162] Alternatively, in some implementations, the UE 104 may receive, from the base station 102, a configuration of mapping between an index and a combination of the first configuration of the first CSI and the one or more second configurations of the second CSI. The index of the combination is configured for the data set.
[0163] In some implementations, the combination indicates the first configuration of the first CSI, the number of the one or more second configurations of the second CSI, and the one or more second configurations of the second CSI.
[0164] Hereinafter, some implementations of a performance target of the model will be described.
[0165] In some implementations, the UE 104 may determine an SGCS or NMSE as a performance target of the model.
[0166] In some implementations, the SGCS or NMSE is configured for the data set. For example, the SGCS or NMSE may be transmitted together with the data set transmitted from the base station 102 to the UE 104.
[0167] In some implementations, SGCS is more suitable when the UE 104 is able to train both the encoder and the decoder as the target CSI and the recovered CSI are compared to calculate SGCS.
[0168] In some implementations, NMSE is more suitable when the UE 104 is only able to train the encoder as the model output by the UE 104 trained encoder and the base station 102 transmitted the CSI feedback is compared to determine the NMSE.
[0169] In some implementations, all the CSI feedback types can apply to both SGCS and NMSE.
[0170] In some implementations, when SGCS is applied, the scalar quantized or vector quantized CSI feedback is used to determine the model output of the decoder at the base station 102.
[0171] In some implementations, when NMSE is applied, the quantized CSI feedback needs to be reverted to floating point and then to be compared with the model output of the encoder at the UE 104.
[0172] In some implementations, different configurations of the target CSI and CSI feedback may correspond to different values of SGCS or NMSE. The reason is that more bits for CSI feedback correspond to more accurate recovery ability.
[0173] In some implementations, after reception of SCGS or NMSE, the UE 104 may transmit, to the base station 102, an indication indicating whether the SGCS or NMSE is supported.
[0174] In some implementations, the SGCS or NMSE can be configured separately with the target CSI type configuration and the CSI feedback configuration. In such implementations, the value of the SGCS or NMSE can be indicated in each subset of the data set or each data set. Alternatively, in some implementations, the base station 102 may transmit, to the UE 104, an indication indicating whether the value of the SGCS or NMSE for a data set / subset is the same as or different from that for the previous data set / subset. If the indication indicates that the value of the SGCS or NMSE for a data set / subset is different from that for the previous data set / subset, the base station 102 may transmit, to the UE 104, a new value of SGCS or NMSE for the data set / subset. Such implementations can be similar to that of target CSI configuration and CSI feedback configuration.
[0175] Alternatively, in some implementations, the SGCS or NMSE is based on at least one of the configurations of the target CSI and the CSI feedback. The mapping between at least one of target CSI and CSI feedback configurations as well as SGCS or NMSE can be per-configured, so that the UE 104 can determine the SGCS or NMSE based on the determined target CSI configuration and / or CSI feedback configuration.
[0176] Alternatively, in some implementations, the SGCS or NMSE is implicitly determined based on pair ID. As one alternative, different pair IDs correspond to different values of SGCS or NMSE. The mapping between pair IDs and values of SGCS or NMSE can be pre-configured.
[0177] Alternatively, in some implementations, the data set may comprise multiple subsets, and a group of the subsets is associated with a group index. Multiple subsets or different groups can be preconfigured to be associated with a value of SGCS or NMSE. The SGCS or NMSE is implicitly determined based on a subset of the data set or the group index. Hereinafter, a subset of the data set is also referred to as a segment of the data set.
[0178] In some implementations, when the rank value is larger than 1, SGCS or NMSE can be separately configured or determined per layer.
[0179] In some implementations, there may be a case that base station 102 wishes the UE 104 to refine model at the UE 104 with some new data sets, for a better SGCS / NMSE or for better generalization. Before this, the UE 104 has already received one or more data sets with target CSI configuration and CSI feedback configuration. In such implementations, after receiving a data set, the UE 104 may receive a further data set from the base station 102. The UE 104 may determine whether the further data set can be combined with the data set for training of the model. Hereinafter, the further data set is also referred to as a new data set.
[0180] It shall be understood that although some implementations of the present disclosure will be described by taking a data set for example, some implementations may be applied to one or more data subsets of the data set. In this regard, a data set may be replaced with a data subset.
[0181] In some implementations, the UE 104 may determine whether the new data set can be combined with the data set for training of the model based on whether the new data set and the data set are associated with a same pair ID or not. If the new data set and the data set are associated with a same pair ID, they can be used together to train the new model. In such implementations, same or different values of SCGS or NMSE can be indicated together as previous implementations for determination of SGCS or NMSE. Different target CSI configuration or CSI feedback configuration can be used to train the model with better generalization performance.
[0182] In some implementations, the data set may be associated with a first configuration of the target CSI and a second configuration of the CSI feedback, and the new data set may be associated with a third configuration of the target CSI and a fourth configuration of the CSI feedback. The UE 104 may determine whether the new data set can be combined with the data set for training of the model based on whether the first configuration is the same as the third configuration or not and / or whether the second configuration is the same as the fourth configuration or not. In such implementations, the SGCS or NMSE may be configured with the new data set. And the received new SGCS or NMSE replaces the old SGCS or NMSE. This corresponds to the scenario that without more data, the UE 104 trains a model with better SGCS or NMSE.
[0183] Fig. 7 illustrates an example of combination of the data set and the new data set for training of the model in accordance with aspects of the present disclosure. In the example of Fig. 7, after receiving data subsets#1, #2 and #3 from the base station 102, the UE 104 may receive data subsets#4 to #12 from the base station 102.
[0184] Because each of the data subsets#1, #2 and #3 are associated with a target CSI configuration#1 (i.e., target CSI config#1) and a CSI feedback configuration#1 (i.e., CSI feedback config#1) while each of the data subsets#4 to #9 are associated with the target CSI config#1 and a CSI feedback configuration#2 (i.e., CSI feedback config#2) , the UE 104 may determine that the data subsets#4 to #9 cannot be combined with the data subsets#1, #2 and #3 for training of the model.
[0185] Because each of the data subsets#1, #2 and #3 are associated with the target CSI config#1) and the CSI feedback config#1 while each of the data subsets#10 to #12 are associated with the target CSI config#1 and the CSI feedback config#1, the UE 104 may determine that the data subsets#10 to #12 can be combined with the data subsets#1, #2 and #3 for training of the model.
[0186] Alternatively, in some implementations, each data set is configured with a group index, and data sets with same group index can be combined for training. In other words, different data subsets will be preconfigured to be associated with a group index, and a data subset belonging to a same group can be jointly used for model training. This will be described with reference to Fig. 8.
[0187] Fig. 8 illustrates an example of combination of the data set and the new data set for training of the model in accordance with aspects of the present disclosure. In the example of Fig. 8, a first group may comprise data subsets#1, #2, #3, #4 and #10, and a second group may comprise data subsets#5, #6, #7, #8, #9, #11 and #12. The data subsets#1, #2, #3, #4 and #10 may be preconfigured to be associated with a first group index of the first group. The data subsets#5, #6, #7, #8, #9, #11 and #12 may be preconfigured to be associated with a second group index of the second group.
[0188] After receiving the data subsets#1, #2 and #3 from the base station 102, the UE 104 may receive the data subsets#10 to #12 from the base station 102. Because the data subsets#4 and #10 are preconfigured to be associated with the first group index, the data subsets#4 and #10 are added to the first group. Because the data subsets#5, #6, #7, #8, #9, #11 and #12 are preconfigured to be associated with the second group index, the data subsets#5, #6, #7, #8, #9, #11 and #12 are added to the second group.
[0189] In turn, the data subsets#1, #2, #3, #4 and #10 can be jointly used for model training, and the data subsets#5, #6, #7, #8, #9, #11 and #12 can be jointly used for model training.
[0190] Alternatively, in some implementations, the UE 104 may receive, from the base station 102, an indicator indicating whether the new data set can be combined with the data set. If the indicator indicates the new data set cannot be combined with the data set, the data set needs to be trained separately. If the UE 104 does not receive the indicator, the UE 104 may determine whether the new data set can be combined with the data set based on judgement of other subsets. If no other data set refers to this data set, this data set can be trained separately, otherwise this data set can be combined with other data set for training.
[0191] If the indicator indicates the new data set can be combined with the data set, the new data set can be combined with the previous data set to make the model has a better generalization performance. The UE 104 may also receive a first index of a first data set and receive a second index of a second data set. Alternatively, the UE 104 may also receive a first index of a first data set and determine the second index of the second data set based on the first index of the first data set and a first number of data sets. In turn, the UE 104 may determine that the new data set can be combined with one or more data sets between the first data set and the second data set.
[0192] For example, if a data subset N indicates to be combined with subsets K to M, and a subset N+1 indicates to be combined with N, it can be considered that the subsets K to M, N and N+1 can be combined.
[0193] Fig. 9 illustrates an example of combination of the data set and the new data set for training of the model in accordance with aspects of the present disclosure. In the example of Fig. 9, after receiving data subsets#1 to #7 from the base station 102, the UE 104 may receive a data subset#8 from the base station 102.
[0194] Together with the data subset#8, the UE 104 also receives “indicator ‘yes’ with subsets from 4 to 7” . In other words, the UE 104 also receives an indicator indicating the new data subset can be combined with the previous data set (i.e., the indicator indicates “yes” ) . The UE 104 also receives a first index of the data subset#4 (i.e., the first data subset) and a second index of the data subset#7 (i.e., the second data subset) . Therefore, based on the indicator as well as the first index and the second index, the UE 104 may determine that the data subset#8 can be combined with the data subsets#4 to #7 for training of the model.
[0195] Then, the UE 104 may receive a data subset#9 from the base station 102. Together with the data subset#9, the UE 104 also receives “indicator ‘yes’ with subset 8” . In other words, the UE 104 also receives an indicator indicating the new data subset can be combined with the previous data subset (i.e., the indicator indicates “yes” ) . The data subset#9 indicates that it can be combined with the data subset#8. Therefore, the UE 104 may determine that the data subset#9 can be combined with the data subsets#4 to #7 and #8 for training of the model.
[0196] In some implementations, different data subsets can be associated with a subset group. The subset group can be configured when the indicator indicates the new data set can be combined with the data set. In such implementations, the group is enlarged by absorb the new data subset. The UE 104 may receive an indication of a group index, and the new data set (or a new data subset) can be combined with a data set (or a data subset) associated with the group index. In such implementations, the new data set is added to the group and associated with the group index.
[0197] Fig. 10 illustrates an example of combination of the data set and the new data set for training of the model in accordance with aspects of the present disclosure. In the example of Fig. 10, after receiving data subsets#1 to #3 from the base station 102, the UE 104 may receive data subsets#4 to #12 from the base station 102.
[0198] Together with each of the data subsets#1 to #12, the UE 104 also receives “indicator ‘yes’ . In other words, the UE 104 also receives an indicator indicating the new data subset can be combined with the previous data subset.
[0199] Together with each of the data subsets#1 to #3 and #10 to #12, the UE 104 also receives a “subset group #1 (i.e., a group index #1) . Therefore, the UE 104 may determine that the data subsets#1 to #3 and #10 to #12 can be combined for training of the model.
[0200] Together with each of the data subsets#4 to #9, the UE 104 also receives a “subset group #2 (i.e., a group index #2) . Therefore, the UE 104 may determine that the data subsets#4 to #9 can be combined for training of the model.
[0201] Alternatively, in some implementations, if the new data set and the previous data set are associated with the same target CSI configuration or CSI feedback configuration, the new data set can be combined with the previous data set.
[0202] Alternatively, in some implementations, if the new data set and the previous data set are associated with the same pair ID, the new data set can be combined with the previous data set.
[0203] In some implementations, the data set is associated with a first SGCS or a first NMSE as a first performance target of the model and the new data set is associated with a second SGCS or a second NMSE as a second performance target of the model. The second SGCS / NMSE overrides the first SGCS / NMSE. The UE 104 is to train the model with the second SGCS / NMSE as performance target. Upon receiving the second SGCS or the second NMSE, the UE 104 may transmit, to the base station 102, an indication indicating whether the second SGCS or second NMSE is supported. Only after UE 104 indicates the second SGCS or second NMSE is supported, the base station 102 will configure corresponding CSI report configuration.
[0204] After combining the new data set with the data set for training of the model, the UE 104 may release the data set, or update the first SGCS with the second SGCS or update the first NMSE with the second NMSE.
[0205] In some implementations, the UE 104 may release the data set after finishing the training of the model.
[0206] Fig. 11 illustrates an example of a device 1100 that supports data set transmission in accordance with aspects of the present disclosure. The device 1100 may be an example of the UE 104 or the base station 102 as described herein. The device 1100 may support wireless communication with one or more network entities 102, UEs 104, or any combination thereof. The device 1100 may include components for bi-directional communications including components for transmitting and receiving communications, such as a processor 1102, a memory 1104, a transceiver 1106, and, optionally, an I / O controller 1108. 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) .
[0207] The processor 1102, the memory 1104, the transceiver 1106, 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 1102, the memory 1104, the transceiver 1106, or various combinations or components thereof may support a method for performing one or more of the operations described herein.
[0208] In some implementations, the processor 1102, the memory 1104, the transceiver 1106, 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 1102 and the memory 1104 coupled with the processor 1102 may be configured to perform one or more of the functions described herein (e.g., executing, by the processor 1102, instructions stored in the memory 1104) .
[0209] For example, the processor 1102 may support wireless communication at the device 1100 in accordance with examples as disclosed herein. The processor 1102 may be configured to operable to support a means for performing the following: receiving, from a base station, a data set comprising data samples related to first CSI and second CSI, wherein the first CSI is an input of a model, and the second CSI is an output of the model; receiving configurations of the first CSI and the second CSI from the base station; and performing training of the model based on the data set and the configurations of the first CSI and the second CSI.
[0210] Alternatively, in some implementations, the processor 1102 may be configured to operable to support a means for performing the following: transmitting, to a UE, a data set comprising data samples related to first CSI and second CSI, wherein the first CSI is an input of a model, and the second CSI is an output of the model; and transmitting configurations of the first CSI and the second CSI to the UE.
[0211] The processor 1102 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 1102 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 1102. The processor 1102 may be configured to execute computer-readable instructions stored in a memory (e.g., the memory 1104) to cause the device 1100 to perform various functions of the present disclosure.
[0212] The memory 1104 may include random access memory (RAM) and read-only memory (ROM) . The memory 1104 may store computer-readable, computer-executable code including instructions that, when executed by the processor 1102 cause the device 1100 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 1102 but may cause a computer (e.g., when compiled and executed) to perform functions described herein. In some implementations, the memory 1104 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.
[0213] The I / O controller 1108 may manage input and output signals for the device 1100. The I / O controller 1108 may also manage peripherals not integrated into the device M02. In some implementations, the I / O controller 1108 may represent a physical connection or port to an external peripheral. In some implementations, the I / O controller 1108 may utilize an operating system such as or another known operating system. In some implementations, the I / O controller 1108 may be implemented as part of a processor, such as the processor 1102. In some implementations, a user may interact with the device 1100 via the I / O controller 1108 or via hardware components controlled by the I / O controller 1108.
[0214] In some implementations, the device 1100 may include a single antenna 1110. However, in some other implementations, the device 1100 may have more than one antenna 1110 (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 1106 may communicate bi-directionally, via the one or more antennas 1110, wired, or wireless links as described herein. For example, the transceiver 1106 may represent a wireless transceiver and may communicate bi-directionally with another wireless transceiver. The transceiver 1106 may also include a modem to modulate the packets, to provide the modulated packets to one or more antennas 1110 for transmission, and to demodulate packets received from the one or more antennas 1110. The transceiver 1106 may include one or more transmit chains, one or more receive chains, or a combination thereof.
[0215] 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 1110 for transmitting the amplified signal into the air or wireless medium.
[0216] 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 1110 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.
[0217] Fig. 12 illustrates an example of a processor 1200 that supports data set transmission in accordance with aspects of the present disclosure. The processor 1200 may be an example of a processor configured to perform various operations in accordance with examples as described herein. The processor 1200 may include a controller 1202 configured to perform various operations in accordance with examples as described herein. The processor 1200 may optionally include at least one memory 1204, such as L1 / L2 / L3 cache. Additionally, or alternatively, the processor 1200 may optionally include one or more arithmetic-logic units (ALUs) 1206. 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) .
[0218] The processor 1200 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 1200) 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) .
[0219] The controller 1202 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 1200 to cause the processor 1200 to support various operations in accordance with examples as described herein. For example, the controller 1202 may operate as a control unit of the processor 1200, generating control signals that manage the operation of various components of the processor 1200. These control signals include enabling or disabling functional units, selecting data paths, initiating memory access, and coordinating timing of operations.
[0220] The controller 1202 may be configured to fetch (e.g., obtain, retrieve, receive) instructions from the memory 1204 and determine subsequent instruction (s) to be executed to cause the processor 1200 to support various operations in accordance with examples as described herein. The controller 1202 may be configured to track memory address of instructions associated with the memory 1204. The controller 1202 may be configured to decode instructions to determine the operation to be performed and the operands involved. For example, the controller 1202 may be configured to interpret the instruction and determine control signals to be output to other components of the processor 1200 to cause the processor 1200 to support various operations in accordance with examples as described herein. Additionally, or alternatively, the controller 1202 may be configured to manage flow of data within the processor 1200. The controller 1202 may be configured to control transfer of data between registers, arithmetic logic units (ALUs) , and other functional units of the processor 1200.
[0221] The memory 1204 may include one or more caches (e.g., memory local to or included in the processor 1200 or other memory, such RAM, ROM, DRAM, SDRAM, SRAM, MRAM, flash memory, etc. In some implementation, the memory 1204 may reside within or on a processor chipset (e.g., local to the processor 1200) . In some other implementations, the memory 1204 may reside external to the processor chipset (e.g., remote to the processor 1200) .
[0222] The memory 1204 may store computer-readable, computer-executable code including instructions that, when executed by the processor 1200, cause the processor 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. The controller 1202 and / or the processor 1200 may be configured to execute computer-readable instructions stored in the memory 1204 to cause the processor 1200 to perform various functions. For example, the processor 1200 and / or the controller 1202 may be coupled with or to the memory 1204, the processor 1200, the controller 1202, and the memory 1204 may be configured to perform various functions described herein. In some examples, the processor 1200 may include multiple processors and the memory 1204 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.
[0223] The one or more ALUs 1206 may be configured to support various operations in accordance with examples as described herein. In some implementation, the one or more ALUs 1206 may reside within or on a processor chipset (e.g., the processor 1200) . In some other implementations, the one or more ALUs 1206 may reside external to the processor chipset (e.g., the processor 1200) . One or more ALUs 1206 may perform one or more computations such as addition, subtraction, multiplication, and division on data. For example, one or more ALUs 1206 may receive input operands and an operation code, which determines an operation to be executed. One or more ALUs 1206 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 1206 may support logical operations such as AND, OR, exclusive-OR (XOR) , not-OR (NOR) , and not-AND (NAND) , enabling the one or more ALUs 1206 to handle conditional operations, comparisons, and bitwise operations.
[0224] The processor 1200 may support wireless communication in accordance with examples as disclosed herein. The processor 1200 may be configured to operable to support a means for performing the following: receiving, from a base station, a data set comprising data samples related to first CSI and second CSI, wherein the first CSI is an input of a model, and the second CSI is an output of the model; receiving configurations of the first CSI and the second CSI from the base station; and performing training of the model based on the data set and the configurations of the first CSI and the second CSI.
[0225] Alternatively, in some implementations, the processor 1200 may be configured to operable to support a means for performing the following: transmitting, to a UE, a data set comprising data samples related to first CSI and second CSI, wherein the first CSI is an input of a model, and the second CSI is an output of the model; and transmitting configurations of the first CSI and the second CSI to the UE.
[0226] Fig. 13 illustrates a flowchart of a method 1300 that supports data set transmission in accordance with aspects of the present disclosure. The operations of the method 1300 may be implemented by a device or its components as described herein. For example, the operations of the method 1300 may be performed by the first communication device 130 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.
[0227] At 1310, the method may include receiving, from a base station, a data set comprising data samples related to first CSI and second CSI, wherein the first CSI is an input of a model, and the second CSI is an output of the model. The operations of 1310 may be performed in accordance with examples as described herein. In some implementations, aspects of the operations of 1310 may be performed by the UE 104 as described with reference to Fig. 1.
[0228] At 1320, the method may include receiving configurations of the first CSI and the second CSI from the base station. The operations of 1320 may be performed in accordance with examples as described herein. In some implementations, aspects of the operations of 1320 may be performed by the UE 104 as described with reference to Fig. 1.
[0229] At 1330, the method may include performing training of the model based on the data set and the configurations of the first CSI and the second CSI. The operations of 1330 may be performed in accordance with examples as described herein. In some implementations, aspects of the operations of 1330 may be performed by the UE 104 as described with reference to Fig. 1.
[0230] Fig. 14 illustrates a flowchart of a method 1400 that supports data set transmission 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 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.
[0231] At 1410, the method may include transmitting, to a UE, a data set comprising data samples related to first CSI and second CSI, wherein the first CSI is an input of a model, and the second CSI is an output of the model. 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 the base station 102 as described with reference to Fig. 1.
[0232] At 1420, the method may include transmitting configurations of the first CSI and the second CSI to the UE. 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 the base station 102 as described with reference to Fig. 1.
[0233] It shall be noted that implementations of the present disclosure which have been described with reference to Figs. 1 to 10 are also applicable to the device 1100, the processor 1200 as well as the methods1300 and 1400.
[0234] It should 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.
[0235] 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.
[0236] 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.
[0237] 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.
[0238] 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.
[0239] 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, via the transceiver from a base station, a data set comprising data samples related to first channel state information (CSI) and second CSI, wherein the first CSI is an input of a model, and the second CSI is an output of the model;receive configurations of the first CSI and the second CSI via the transceiver from the base station; andperform training of the model based on the data set and the configurations of the first CSI and the second CSI.2.The UE of claim 1, wherein the configurations of the first CSI and the second CSI comprise a first configuration of the first CSI, wherein the first configuration of the first CSI comprises at least one of the following:the number of antenna ports;the number of subbands;a rank value;an indication indicating whether format of the first CSI is a precoding matrix or a channel matrix; ora quantization type of the first CSI.3.The UE of claim 1, wherein the configurations of the first CSI and the second CSI comprise a second configuration of the second CSI, wherein the second configuration of the second CSI comprises at least one of the following:the number of an output value of the model,a size of segments of the second CSI, wherein each of the segments of the second CSI comprises one or more output values of the model,a quantization type of each of the segments of the second CSI, orthe number of bits for quantization of each of the segments of the second CSI.4.The UE of claim 2 or 3, wherein the quantization type of the first CSI or quantization type of each of the segments of the second CSI comprises one of the following:scalar quantization, andvector quantization.5.The UE of claim 4, wherein the quantization type of the first CSI or quantization type of each of the segments of the second CSI comprises the vector quantization; andwherein the processor is further configured to:receive, via the transceiver from the base station, a configuration of a codebook or parameters related to the codebook.6.The UE of claim 1, wherein the configurations of the first CSI and the second CSI comprise at least one of the following:a first configuration of the first CSI,one or more second configurations of the second CSI associated with the first configuration of the first CSI, orthe number of second configurations of the second CSI associated with the first configuration of the first CSI.7.The UE of claim 6, wherein the processor is further configured to:receive, via the transceiver from the base station, a configuration of the following:mapping between a first index of the first configuration of the first CSI and the first configuration; andmapping between one or more second indexes of the one or more second configurations of the second CSI and the one or more second configurations; andwherein the first index and the one or more second indexes are configured for the data set.8.The UE of claim 6, wherein the data set comprises a first subset and a second subset;wherein the processor is configured to receive the data set by:receiving the second subset after receiving the first subset; andwherein the second subset is configured with an indicator, wherein the indicator indicates whether at least one of the first configuration, the number of the one or more second configurations associated with the first configuration, and the one or more second configurations are same as or different from those in the first subset.9.The UE of claim 1, wherein the processor is further configured to:determine a squared generalized cosine similarity (SGCS) or a normalized mean squared error (NMSE) as a performance target of the model.10.The UE of claim 9, wherein the SGCS or NMSE is configured for the data set; and / orwherein the SGCS or NMSE is based on at least one of the configurations of the first CSI and the second CSI.11.The UE of claim 9, wherein the SGCS or NMSE is configured or determined per layer; and / or wherein the processor is further configured to:transmit an indication indicating whether the SGCS or NMSE is supported.12.The UE of claim 1, wherein the processor is further configured to:transmit an indication indicating whether the configurations of the first CSI or the second CSI are supported; and / orwherein the processor is further configured to:after finishing the training of the model, release the data set.13.The UE of claim 1, wherein the processor is further configured to:after receiving the data set, receive a further data set via the transceiver from the base station; anddetermine whether the further data set can be combined with the data set for training.14.The UE of claim 13, wherein the processor is configured to determine whether the further data set can be combined with the data set for training of the model:based on whether the further data set and the data set are associated with a same pair identity (ID) or not; and / orwherein the processor is configured to determine whether the further data set can be combined with the data set for training of the model:based on whether the first configuration is the same as the third configuration or not, and / or whether the second configuration is the same as the fourth configuration or not.15.The UE of claim 13, wherein the processor is further configured to:receive, via the transceiver from the base station, an indicator indicating whether the further data set can be combined with the data set.16.The UE of claim 15, wherein the processor is further configured to:receive an indication of a group index, and the further data set can be combined with a data set associated with the group index; and / orwherein the processor is further configured to:receive a first index of a first data set;receive a second index of a second data set or determine the second index of the second data set based on the first index of the first data set and a first number of data sets; anddetermine that the further data set can be combined with one or more data sets between the first data set and the second data set.17.The UE of claim 13, wherein the data set is associated with a first squared generalized cosine similarity (SGCS) or a first normalized mean squared error (NMSE) as a first performance target of the model;wherein the further data set is associated with a second SGCS or a second NMSE as a second performance target of the model; andwherein the processor is further configured to:after combining the further data set with the data set for training of the model, release the data set; orupdate the first SGCS with the second SGCS or update the first NMSE with the second NMSE.18.A base station, comprising:a processor; anda transceiver coupled to the processor,wherein the processor is configured to:transmit, via the transceiver to a user equipment (UE) , a data set comprising data samples related to first channel state information (CSI) and second CSI, wherein the first CSI is an input of a model, and the second CSI is an output of the model; andtransmit configurations of the first CSI and the second CSI via the transceiver to the UE.19.A processor for wireless communication, comprising:at least one memory; anda controller coupled with the at least one memory and configured to cause the controller to:receive, via a transceiver from a base station, a data set comprising data samples related to first channel state information (CSI) and second CSI, wherein the first CSI is an input of a model, and the second CSI is an output of the model;receive configurations of the first CSI and the second CSI via the transceiver from the base station; andperform training of the model based on the data set and the configurations of the first CSI and the second CSI.20.A processor for wireless communication, comprising:at least one memory; anda controller coupled with the at least one memory and configured to cause the controller to:transmit, via a transceiver to a user equipment (UE) , a data set comprising data samples related to first channel state information (CSI) and second CSI, wherein the first CSI is an input of a model, and the second CSI is an output of the model; andtransmit configurations of the first CSI and the second CSI via the transceiver to the UE.