EP4666461A1

EP4666461A1 - Method and apparatus for channel state information reporting using an autoencoder

Info

Publication number: EP4666461A1
Application number: EP24707985.8A
Authority: EP
Inventors: Chandan PRADHAN; Xinlin ZHANG; Ilmiawan SHUBHI; Emil RINGH
Original assignee: Telefonaktiebolaget LM Ericsson AB
Current assignee: Telefonaktiebolaget LM Ericsson AB
Priority date: 2023-02-16
Filing date: 2024-02-16
Publication date: 2025-12-24
Also published as: CN120937280A; WO2024172744A1

Abstract

A method for channel state information, CSI, reporting, performed by a wireless device. The wireless device has one or more encoders of one or more autoencoders available. The method comprises generating a CSI report using an autoencoder. The method further comprises segmenting the output of the autoencoder into a first CSI report and a second CSI report, wherein the first CSI report and the second CSI report are transmitted on different parts of the Uplink Control Information, UCI. Further disclosed are related methods for a radio access node, and related wireless devices, radio access nodes, computer programs, and computer program products.

Description

METHOD AND APPARATUS FOR CSI REPORTING TECHNICAL FIELD The present disclosure relates broadly to wireless communications and more particularly to methods for compression of channel state information. Further disclosed are related apparatuses. BACKGROUND The Third Generation Partnership Project (3GPP) has developed and is developing standards for Fourth Generation (4G) (also referred to as Long Term Evolution (LTE)) and Fifth Generation (5G) (also referred to as New Radio (NR)) wireless communication systems. Such systems provide, among other features, broadband communication between network nodes, such as base stations, and mobile wireless devices, as well as communication between network nodes and between wireless devices. The 3GPP is also developing standards for Sixth Generation (6G) wireless communication networks. NR uses Orthogonal Frequency Division Multiplexing (OFDM) with configurable bandwidths and subcarrier spacing to efficiently support a diverse set of use cases and deployment scenarios. With respect to LTE, NR improves in deployment flexibility, user throughputs, latency and reliability. With NR comes also enhanced support for spatial multiplexing in which time-frequency resources are spatially shared across users, commonly referred to as Multi-User MIMO (MU-MIMO). MU-MIMO operations are illustrated in Fig.1, where a multi-antenna network node with ^^_்^ antenna ports is spatially transmitting information to several wireless devices, in which sequence ^^^{^^^} is intended for wireless device UE(1), sequence ^^^{^ଶ^} is intended for wireless UE(2), and so on. Before each transmission is modulated and transmitted, a precoding ^^^{^^^} ^ us applied to each sequence to spatially separate the transmissions, i.e., to mitigate multiplexing interference. At the receiver side, each wireless device demodulates its received signal and _{combines received antenna signals to obtain an estimate ^} ^{^} _^ ^{^^^} _{of the transmitted} _{sequence. The estimate ^} ^{^} _^ ^{^^^} _{can be described mathematically as:} 1 _^ ^{^} _^ ^{^^^} _{ൌ ^^} ^{^^^} ^_^^ ^{^^^} _^^ ^{^^^} ^_^^ ^{^^^} _{^ ^^} ^{^^^} ^_^^ ^{^^^} _{^ ^^} ^{^^^} ^_^^ ^{^^^} , ^ ^^ ^^.1^. In Eq.1, the matrix and the second term multiplexing interference seen by UE(i). The goal for the network node for MU-MIMO is to construct the set of precoders ^ ^^^{^^^} ^ ^ such that the norm ^ ^^^{^^^} ^^^{^^^} ^ ^ is large whereas the norm ^ ^^^{^^^} ^^^{^^^} ^ ^, ^^ ് ^^ is _^ correlates well with the channel ^^^{^^^} observed by UE^ ^^^ whereas it poorly with other channels. construct precoders for efficient MU-MIMO transmissions, the network node may need to acquire detailed knowledge of the channels ^^^ ^^^. In deployments where channel reciprocity holds, channel knowledge can be acquired from sounding reference signals (SRS) that are transmitted periodically, or on demand, by active wireless devices. Based on these SRS, the network node estimates ^^^{^^^}. However, when channel reciprocity does not hold or when SRS coverage is active wireless devices may need to feedback channel details to the network node. In NR (as well as in LTE), this is done by having the network node periodically transmit Channel State Information Reference Signals (CSI-RS) from which a wireless device can estimate its channel. The wireless device then reports CSI, from which the network node can determine suitable precoders for MU-MIMO. However, transmitting CSI from the wireless device to the network node is costly in terms of use bandwidth and computational power. It is therefore of interest to develop methods to compress and transmit CSI efficiently. SUMMARY Some embodiments advantageously provide methods, systems, and apparatuses for RRC signaling and CSI reporting for AI-based CSI compression and feedback. Disclosed herein are methods for RRC signaling along with reporting AI-based CSI on UCI, which include segmentation of the CSI report and mapping order of CSI report to UCI bit sequences. 2 The methodology for the feedback of the CSI report on the UCI is described herein, where the wireless devices process the eigenvectors per transmission layer with an AI model to generate the CSI report. According to a first aspect, there is a method for channel state information, CSI, reporting performed by a wireless device. The wireless device has one or more encoders of one or more autoencoders available. The method comprises generating a CSI report using an autoencoder. The method comprises segmenting the output of the autoencoder into a first CSI report and a second CSI report. The first CSI report and the second CSI report are transmitted on different parts of an Uplink Control Information, UCI. According to an embodiment of the first aspect, the method further comprises receiving, through Radio Resource Control, RRC, signaling, an indication to obtain a CSI report using an autoencoder. According to an embodiment of the first aspect, generating a CSI report using an autoencoder comprises preprocessing an estimated channel. According to an embodiment of the first aspect, preprocessing an estimated channel comprises extracting eigenvectors per transmission layer from an estimated channel; and reducing the dimension of the extracted eigenvectors per transmission layer by applying a pre-processing in a space, frequency, and/or time domain. According to an embodiment of the first aspect, generating a CSI report using an autoencoder comprises compressing and quantizing the pre-processed eigenvectors per transmission layer with an autoencoder. According to an embodiment of the first aspect. the information of the space, frequency, and/or time domain forms part of the CSI report on UCI. According to an embodiment of the first aspect, the quantized bits form part of the CSI report on UCI. According to an embodiment of the first aspect, the method further comprises receiving through RRC signaling, parameters associated to the content and/or size of the encoded CSI. The parameters include one or more of: a restriction on the rank of the encoded CSI; an indication of the space, frequency, and/or time domain basis 3 used to pre-process the eigenvectors per transmission layer; an indication of the quantization bits which can be used by the wireless device to quantize the pre- processed eigenvectors per transmission layer; an indication of the number of latent space coefficients of the encoded CSI per transmission layer; an indication of whether to use a transmission layer-common processing or a transmission layer- specific processing to generate the CSI report at the wireless device. According to an embodiment of the first aspect, one or more of the parameters is explicitly associated to an autoencoder deployed at the wireless device. According to an embodiment of the first aspect, a model identification uniquely associated to one of the autoencoders available to the wireless device is signaled to the wireless device. According to an embodiment of the first aspect, one or more of the parameters is dynamically configured on the Downlink Control Information, DCI, and/or the Medium Access Control, MAC, Control Element, CE. According to an embodiment of the first aspect, one or more of the parameters is configured by the wireless device and reported as part of the CSI report on UCI. According to an embodiment of the first aspect, the first CSI report includes one or more of: a legacy CSI report quantity; information about the autoencoder used to generate the CSI report; auxiliary information for the AI-based quantization bits; a basis used for pre-processing. According to an embodiment of the first aspect, the second CSI report includes one or more of: an index for the space, frequency, and/or time dimension pre-processing basis for the eigenvectors per transmission layer; transmission layer-specific information required by the network node to decode the quantized bits per transmission layer; quantized bits per transmission layer from the output of the autoencoder. According to a second aspect, there is a method for channel state information, CSI, reporting performed by a network node. The network node has one or more decoders of one or more autoencoders available. The method comprises indicating, through Radio Resource Control, RRC, signaling, that a wireless device is to transmit a CSI report compressed by means of an autoencoder. The method 4 comprises receiving, from a wireless device, a first CSI report and a second CSI report on different parts of Uplink Control Information, UCI. The method comprises using a decoder from the one or more decoders to decode the first CSI report and the second CSI report. According to an embodiment of the second aspect, the method further comprises transmitting, through RRC signaling, parameters associated to the content and/or size of the encoded CSI. The parameters include one or more of: a restriction on the rank of the encoded CSI; an indication of the space, frequency, and/or time domain basis used to pre-process the eigenvectors per transmission layer; an indication of the quantization bits which can be used by the wireless device to quantize the pre- processed eigenvectors per transmission layer; an indication of the number of latent space coefficients of the encoded CSI per transmission layer; an indication of whether to use a transmission layer-common processing or a transmission layer- specific processing to generate the CSI report at the wireless device. According to an embodiment of the second aspect, one or more of the parameters is explicitly associated to an autoencoder deployed at the network node. According to an embodiment of the second aspect, a model identification uniquely associated to one of the autoencoders available to the wireless device is signaled to the wireless device. According to an embodiment of the second aspect. one or more of the parameters is dynamically configured on the Downlink Control Information, DCI, and/or the Medium Access Control, MAC, Control Element, CE. According to an embodiment of the second aspect, one or more of the parameters is configured by the wireless device and received by the network node as part of the CSI report on UCI. According to an embodiment of the second aspect, the first CSI report includes one or more of: a legacy CSI report quantity; information about the autoencoder used to generate the CSI report; auxiliary information for the AI-based quantization bits; a basis used for pre-processing. According to an embodiment of the second aspect, the second CSI report includes one or more of: an index for the space, frequency, and/or time dimension pre- 5 processing basis for the eigenvectors per transmission layer; transmission layer- specific information required by the network node to decode the quantized bits per transmission layer; quantized bits per transmission layer from the output of the autoencoder. According to a third aspect, there is a wireless device configured to perform a method for channel state information, CSI, reporting. The wireless device has one or more encoders of one or more autoencoders available. The method comprises generating a CSI report using an autoencoder. The method comprises segmenting the output of the autoencoder into a first CSI report and a second CSI report, wherein the first CSI report and the second CSI report are transmitted on different parts of the Uplink Control Information, UCI. According to an embodiment of the third aspect, the wireless device is further configured to perform a method according to any embodiment of the first aspect. According to a fourth aspect, there is a wireless device configured to perform a method for channel state information, CSI, reporting. The wireless device comprises processing circuitry and a memory. The wireless device has one or more encoders of one or more autoencoders available. The method comprises generating a CSI report using an autoencoder. The method comprises segmenting the output of the autoencoder into a first CSI report and a second CSI report, wherein the first CSI report and the second CSI report are transmitted on different parts of the Uplink Control Information, UCI. According to an embodiment of the fourth aspect, the wireless device is further configured to perform a method according to any embodiment of the first aspect. According to a fifth aspect, there is a radio access node in a communication network configured to perform a method for channel state information, CSI, reporting. The network node has one or more decoders of one or more autoencoders available. The method comprises indicating, through Radio Resource Control, RRC, signaling, that a wireless device is to transmit a CSI report compressed by means of an autoencoder. The method comprises receiving, from a wireless device, a first CSI report and a second CSI report on different parts of Uplink Control Information, UCI. The method comprises using a decoder from the one or more decoders to decode the first CSI report and the second CSI report. 6 According to an embodiment of the fifth aspect, the radio access node is further configured to perform a method according to any embodiment of the second aspect. According to a sixth aspect, there is a radio access node in a communication network configured to perform a method for channel state information, CSI, reporting. The network node comprises processing circuitry and a memory. The network node has one or more decoders of one or more autoencoders available. The method comprises indicating, through Radio Resource Control, RRC, signaling, that a wireless device is to transmit a CSI report compressed by means of an autoencoder. The method comprises receiving, from a wireless device, a first CSI report and a second CSI report on different parts of Uplink Control Information, UCI. The method comprises using a decoder from the one or more decoders to decode the first CSI report and the second CSI report. According to an embodiment of the sixth aspect, the radio access node is further configured to perform a method according to any embodiment of the second aspect. According to a seventh aspect, there is a computer program comprising machine- readable instructions which, when executed by the processor of a wireless device, cause the wireless device to perform a method according to any embodiment of the first aspect. According to an eighth aspect, there is a computer program product comprising a non-transient computer readable storage medium on which a computer program according to the seventh aspect is stored. According to a ninth aspect, there is a computer program comprising machine- readable instructions which, when executed by the processor of a radio access node, cause the radio access node to perform a method according to any embodiment of the second aspect. According to a tenth aspect, there is a computer program product comprising a non- transient computer readable storage medium on which a computer program according to the ninth aspect is stored. BRIEF DESCRIPTION OF THE DRAWINGS 7 A more complete understanding of the present embodiments, and the attendant advantages and features thereof, will be more readily understood by reference to the following detailed description when considered in conjunction with the accompanying drawings wherein: FIG.1 is a flowchart illustrating a multiple MU-MIMO operation; FIG.2 is an illustration of CSI Type II feedback; FIG.3 is an illustration of a fully connected autoencoder; FIG.4 is an illustration of use of an autoencoder for CSI compression; FIG.5 is flowchart of a quantization operation at the output of the encoder to fit the CSI payload over the air interface; FIG.6 is an illustration of pre-processing for implicit feedback of the eigenvector based on an estimated transmission rank; FIG.7 is an illustration of a transmission layer common model; FIG.8 is an illustration of a transmission layer specific model; FIG.9 is a schematic diagram of an example network architecture illustrating a communication system connected via an intermediate network to a host computer according to the principles in the present disclosure; FIG.10 is a block diagram of a host computer communicating via a network node with a wireless device over an at least partially wireless connection according to some embodiments of the present disclosure; FIG.11 is a flowchart illustrating example methods implemented in a communication system including a host computer, a network node and a wireless device for executing a client application at a wireless device according to some embodiments of the present disclosure; FIG.12 is a flowchart illustrating example methods implemented in a communication system including a host computer, a network node and a wireless device for receiving user data at a wireless device according to some embodiments of the present disclosure; 8 FIG.13 is a flowchart illustrating example methods implemented in a communication system including a host computer, a network node and a wireless device for receiving user data from the wireless device at a host computer according to some embodiments of the present disclosure; FIG.14 is a flowchart illustrating example methods implemented in a communication system including a host computer, a network node and a wireless device for receiving user data at a host computer according to some embodiments of the present disclosure; FIG.15A is a flowchart of an example process in a network node according to some embodiments of the present disclosure; FIG.15B is a flowchart of an example process in a network node according to some embodiments of the present disclosure; FIG.16A is a flowchart of an example process in a wireless device according to some embodiments of the present disclosure; FIG.16B is a flowchart of an example process in a wireless device according to some embodiments of the present disclosure; and FIG.17 is a schematic diagram of an example architecture for a channel eigenvector feedback approach according to some embodiments of the present disclosure. DETAILED DESCRPTION OF THE DRAWINGS Hardware components have been represented where appropriate by conventional symbols in the drawings, showing only those specific details that are pertinent to understanding the embodiments so as not to obscure the disclosure with details that will be readily apparent to those of ordinary skill in the art having the benefit of the description herein. Like numbers refer to like elements throughout the description. As used herein, relational terms, such as “first” and “second,” “top” and “bottom,” and the like, may be used solely to distinguish one entity or element from another entity or element without necessarily requiring or implying any physical or logical relationship or order between such entities or elements. The terminology used herein is for the purpose of describing particular embodiments only and is not 9 intended to be limiting of the concepts described herein. 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,” “includes” and/or “including” when used herein, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. In embodiments described herein, the joining term, “in communication with” and the like, may be used to indicate electrical or data communication, which may be accomplished by physical contact, induction, electromagnetic radiation, radio signaling, infrared signaling or optical signaling, for example. One having ordinary skill in the art will appreciate that multiple components may interoperate and modifications and variations are possible of achieving the electrical and data communication. In some embodiments described herein, the term “coupled,” “connected,” and the like, may be used herein to indicate a connection, although not necessarily directly, and may include wired and/or wireless connections. The term “network node” used herein can be any kind of network node comprised in a radio network which may further comprise any of base station (BS), radio base station, base transceiver station (BTS), base station controller (BSC), radio network controller (RNC), g Node B (gNB), evolved Node B (eNB or eNodeB), Node B, multi- standard radio (MSR) radio node such as MSR BS, multi-cell/multicast coordination entity (MCE), integrated access and backhaul (IAB) node, relay node, donor node controlling relay, radio access point (AP), transmission points, transmission nodes, Remote Radio Unit (RRU) Remote Radio Head (RRH), a core network node (e.g., mobile management entity (MME), self-organizing network (SON) node, a coordinating node, positioning node, MDT node, etc.), an external node (e.g., 3rd party node, a node external to the current network), nodes in distributed antenna system (DAS), a spectrum access system (SAS) node, an element management system (EMS), etc. The network node may also comprise test equipment. The term “radio node” used herein may be used to also denote a wireless device (WD) such as a wireless device or a radio network node. 10 In some embodiments, the non-limiting terms wireless device or a user equipment (UE) are used interchangeably. The wireless device herein can be any type of wireless device capable of communicating with a network node or another wireless device over radio signals, such as wireless device wireless device. The wireless device may also be a radio communication device, target device, device to device (D2D) wireless device, machine type wireless device or wireless device capable of machine to machine communication (M2M), low-cost and/or low-complexity wireless device, a sensor equipped with wireless device, Tablet, mobile terminals, smart phone, laptop embedded equipped (LEE), laptop mounted equipment (LME), USB dongles, Customer Premises Equipment (CPE), an Internet of Things (IoT) device, or a Narrowband IoT (NB-IOT) device, etc. Also, in some embodiments the generic term “radio network node” is used. It can be any kind of a radio network node which may comprise any of base station, radio base station, base transceiver station, base station controller, network controller, RNC, evolved Node B (eNB), Node B, gNB, Multi-cell/multicast Coordination Entity (MCE), IAB node, relay node, access point, radio access point, Remote Radio Unit (RRU) Remote Radio Head (RRH). Note that the concept of ‘network’ and/or a network node can be understood as a generic network node, gNB, base station, unit within the base station to handle at least some ML operation, relay node, core network node, a core network node that handle at least some ML operations, or a device supporting D2D communication. The node may be deployed in a 5G network, or a 6G network. Note that although terminology from one particular wireless system, such as, for example, 3GPP LTE and/or New Radio (NR), may be used in this disclosure, this should not be seen as limiting the scope of the disclosure to only the aforementioned system. Other wireless systems, including without limitation Wide Band Code Division Multiple Access (WCDMA), Worldwide Interoperability for Microwave Access (WiMax), Ultra Mobile Broadband (UMB) and Global System for Mobile Communications (GSM), may also benefit from exploiting the ideas covered within this disclosure. In some embodiments, the general description elements in the form of “one of A and B” corresponds to A or B. In some embodiments, at least one of A and B 11 corresponds to A, B or AB, or to one or more of A and B, or one or both of A and B. In some embodiments, at least one of A, B and C corresponds to one or more of A, B and C, and/or A, B, C or a combination thereof. Note further, that functions described herein as being performed by a wireless device or a network node may be distributed over a plurality of wireless devices and/or network nodes. In other words, it is contemplated that the functions of the network node and wireless device described herein are not limited to performance by a single physical device and, in fact, can be distributed among several physical devices. Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs. It will be further understood that terms used herein should be interpreted as having a meaning that is consistent with their meaning in the context of this specification and the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein. Some embodiments provide for RRC signaling and CSI reporting for AI-based CSI compression and feedback. The CSI feedback mechanism targeting multi-user multiple-input multiple-output (MU-MIMO) operations in NR is referred to as CSI type II, in which a wireless device reports CSI feedback with high CSI resolution, as specified, e.g., in 3GPP TS 38.214. It is based on specifying sets of Direct Fourier Transform (DFT) base functions (grid of beams) from which the wireless device selects those that best match its channel conditions (like classical codebook Pre-coding Matrix Indicator (PMI)). The number of beams the wireless device reports is configurable via Radio Resource Control (RRC) signaling, and may be 2 or 4, as specified in, e.g., 3GPP Technical Specification (T.S.) Rel-15 Type II or 2, 4 or 6, as specified, e.g., in T.S. Rel-16 Type II. In Rel-16 Type II, the CSI report can be further compressed in the frequency domain (FD), where a set of FD DFT basis vectors are selected by the wireless device. The number of selected FD basis vectors is a function of the number of CQI sub-bands, the number of PMI sub-bands per Channel Quality Indicator (CQI) sub-band and a ratio that determines the FD compression (which may be termed as ^^_௩, as specified e.g., in 3GPP TS 38.214, where ^^ is the layer index), which is configured by network node via RRC signaling. In addition, the 12 wireless device also reports non-zero coefficients (NZCs) associated with the selected beams for Rel-15 Type II, which informs the network node how these beams should be combined in terms of relative amplitude scaling and co-phasing for each sub-band. In Rel-16, the reported NZCs are then associated with selected beams and FD basis vectors. In Rel-16, to further compress the CSI report, network node also configures a ratio, termed as ^^, to the wireless device via RRC signaling, that determines the maximum number of NZCs to be reported. For example, for a single layer transmission where 2 ^^ beams and ^^ FD basis vectors are configured by network node, there are in total 2 ^^ ^^ linear combination coefficients. Then, only ^⌈ _{2 ^^ ^^ ^^} ^⌉ _{NZCs may be reported at most, the remaining 2 ^^ ^^ െ} ^⌈ _{2 ^^ ^^ ^^} ^⌉ _{are treated as} zeros and are not reported. The selected beams are commonly used for all subbands and all transmission layers, whereas the NZCs (for both Rel-15 and Rel- 16 Type II) and FD basis vectors (for Rel-16 Type II) are layer-specific. To further explain the structure of the Type II CSI, and example of the Rel-15 CSI type II is illustrated in FIG.2 (illustrating a CSI Type II feedback), which illustrates the selection of DFT beam vectors ^^_^, and their relative amplitudes ^^_^, are determined from a wideband perspective whereas the co-phasing is per subband. Here, wideband means that the selected DFT beam vectors are the same for all subcarriers used in the OFDM transmission, whereas subband means that co- phasing parameters are determined over subsets of contiguous subcarriers. The co- phasing parameters are quantized such that ^^^{^ఏ^} is taken from either a QPSK or 8PSK signal constellation. With ^^ denoting a sub-band index, the precoder reported by the wireless device can be expressed as ^^{^} _^ ^{^ ^^^ ൌ} _^ ^{^^} _^ ^{^^} _^ ^{^^^ఏ^^^^} . Note that the reporting large, especially when comparing to the Type I CSI. A dominant part of the reporting overhead is from sub- band reporting, e.g., the layer-specific NZCs. For instance, it requires about 7 bits (the actual number depends on the release version and parameter configuration) to report the phase and amplitude for one coefficient. CSI reporting in NR 13 In NR, a wireless device can be configured with one or multiple CSI Report Settings, each configured by a higher layer parameter CSI-ReportConfig. Each CSI- ReportConfig is associated with a Bandwidth Part (BWP) and contains one or more of the following: ^ a CSI resource configuration for channel measurement ^ a CSI Interference Measurement (CSI-IM) resource configuration for interference measurement ^ reporting configuration type, i.e., aperiodic CSI (on Physical Uplink Shared Channel (PUSCH)), periodic CSI (on Physical Uplink Control Channel (PUCCH)), or semi-persistent CSI on PUCCH or PUSCH ^ report quantity specifying what to be reported, such as Rank Indication (RI), PMI, CQI ^ codebook configuration such as type I or type II CSI ^ frequency domain configuration, i.e., sub-band vs. wideband CQI or PMI, and sub-band size ^ CQI table to be used A wireless device can be configured with one or multiple CSI resource configurations for channel measurement and one or more CSI-IM resources for interference measurement. Each CSI resource configuration for channel measurement can contain one or more NZP CSI-RS resource sets. For each NZP CSI-RS resource set, it can further contain one or more NZP CSI-RS resources. A NZP CSI-RS resource can be periodic, semi-persistent, or aperiodic. Similarly, each CSI-IM resource configuration for interference measurement can contain one or more CSI-IM resource sets. For each CSI-IM resource set, it can further contain one or more CSI-IM resources. A CSI-IM resource can be periodic, semi-persistent, or aperiodic. Type II CSI report on PUSCH 14 A wireless device performs aperiodic CSI reporting using PUSCH upon successful decoding of a Downlink Control Information (DCI) format 0_1 or DCI format 0_2, which triggers an aperiodic CSI trigger state. When a DCI format 0_1 schedules two PUSCH allocations, the aperiodic CSI report is carried on the second scheduled PUSCH. When a DCI format 0_1 schedules more than two PUSCH allocations, the aperiodic CSI report is carried on the penultimate scheduled PUSCH. A wireless device performs semi-persistent CSI reporting on the PUSCH upon successful decoding of a DCI format 0_1 or DCI format 0_2 which activates a semi- persistent CSI trigger state. DCI format 0_1 and DCI format 0_2 contains a CSI request field which indicates the semi-persistent CSI trigger state to activate or deactivate. The PUSCH resources and MCS are allocated semi-persistently by an uplink DCI. CSI reporting on PUSCH can be multiplexed with uplink data on PUSCH. CSI reporting on PUSCH can also be performed without any multiplexing with uplink data from the wireless device. Part 1 and Part 2 for Type II CSI report For the Rel-15 Type II and the Rel-16 Type II (aka Enhanced Type II, or eType II) CSI feedback on PUSCH, a CSI report includes two parts: Part 1 and Part 2. One reason for dividing a CSI report into Part 1 and Part 2 is to deal with the dynamically varying CSI payload. For example, based on the time-varying channel, wireless device may report different ranks over the whole period of connection, which has significant impact on the actual required CSI payload size. In order for the network node to know the actual payload size, Part 1, which has a fixed payload size that carries the information to calculate the payload size of Part 2, will be decoded first by network node. ^ For the Rel-15 Type II CSI feedback, Part 1 contains RI (if reported), CQI, and an indication of the number of non-zero wideband amplitude coefficients per layer for the Type II CSI (see, e.g., Clause 5.2.2.2.3 in 3GPP T.S.38.214). The fields of Part 1 – RI (if reported), CQI, and the indication of the number of non-zero wideband amplitude coefficients 15 for each layer – are separately encoded. Part 2 contains the PMI of the Type II CSI. Part 1 and 2 are separately encoded. ^ For the Rel-16 Type II CSI feedback, Part 1 contains RI, CQI, and an indication of the overall number of non-zero amplitude coefficients across layers for the Rel-16 Type II CSI (see, e.g., Clause 5.2.2.2.5 in 3GPP T.S.38.214). The fields of Part 1 – RI, CQI, and the indication of the overall number of non-zero amplitude coefficients across layers – are separately encoded. Part 2 contains the PMI of the Enhanced Type II CSI. Part 1 and 2 are separately encoded. Autoencoders for AI/ML-enhanced CSI reporting Recently neural network (NN) based autoencoders (AEs) have been used for compressing downlink MIMO channel estimates for uplink feedback. Furthermore, 3GPP decided to start a study item for Rel.18 that includes the use case of AI-based CSI reporting, in which AEs are part of the study. Specifically, an AE is a type of artificial neural network that can be used to compress and decompress data, in an unsupervised manner, often with high fidelity. FIG.3 illustrates a low complexity fully connected (dense) AE. The AE is divided into two parts: - an encoder (used to compress the input data ^^), and - a decoder (used to de-compress the input data). AEs can have different architectures. For example, AEs can be based on dense NNs, multi-dimensional convolution NNs, variational, recurrent NNs, transformer networks, or any combination thereof. However, all AE architectures possess an encoder-bottleneck-decoder structure illustrated in FIG.3. The size of the codeword (denoted by ^^ in FIG.3) of an AE is typically a lot smaller than the size of the input data ( ^^ in FIG.3). The AE encoder thus reduces the dimensionality of the input features ^^ down to ^^. The decoder part of the AE tries to invert the encoder and reconstruct ^^ with minimal error, according to some predefined loss function. 16 FIG.4 illustrates how an AE might be used for AI/ML-enhanced CSI reporting in NR. The wireless device measures the channel in the downlink using CSI-RS. The wireless device estimates that channel for each subcarrier (SC) from each network node transmission (TX) antenna and at each wireless device receiving (RX) antenna. The estimate can be viewed as a three-dimensional (3D) channel matrix. The 3D channel matrix represents the MIMO channel estimated over several SCs and is input to the encoder. The AE encoder is implemented in the wireless device, and the AE decoder is implemented in the network, e.g., in a network node. The output of the AE encoder is signaled from the wireless device to the network node over the uplink. The codeword can be viewed a learned latent representation of the channel. The architecture of an AE (e.g., number of layers, nodes per layer, activation function) typically needs to be numerically optimized for CSI reporting via a process called hyperparameter tuning. Properties of the data (e.g., CSI-RS channel estimates), the channel size, uplink feedback rate, and hardware limitations of the encoder and decoder may all need to be considered when optimizing the AE’s architecture. The weights and biases of an AE (with a fixed architecture) are trained to minimize the reconstruction error (the error between the input ^^ and output ^^{^}^) on some training dataset. For example, the weights and biases can be trained to minimize the mean squared error (MSE) ^ ^^ െ ^^{^}^^^ଶ. Model training is typically done using some variant of the gradient descent algorithm on a large training data set. To achieve good performance during live operation, the training data set should be representative of the actual data the AE will encounter during live operation. In the two-sided CSI compression, the output of the wireless device-side encoder needs to be communicated over the air interface to the network node decoder with the assigned CSI reporting payload and, therefore, needs to be quantized to a finite number of bits (e.g., 1-4 bits per sample for the UCI) to obtain an efficient transmission, as shown in FIG.5 (illustrating quantization operation at the output of the encoder to fit the CSI payload over the air interface). Accordingly, a quantization layer is usually connected at the output of the encoder or directly included in the encoder. In an example, the quantization layer may implement scalar quantization which quantizes the output of each neuron of the encoder output layer (the 17 bottleneck layer of AE) to generate bits to fit the CSI reporting payload in the UCI. Other quantization methods, e.g., vector quantization, may also be used. Pre-processing for input data to the AE A proper pre-processing on the input to the encoder can greatly reduce the size and complexity for designing and/or training an AI/ML model, and in the meantime, improving the scalability and transferability of the model. In the CSI compression, a pre-processing method could be a transformation of the channel from antenna- frequency domain to beam-delay domain, or from the antenna-frequency-time domain to the beam-delay-doppler domain. In addition, the pre-processing is used to reduce the need for multiple models depending on bandwidth variation and variation in the number of antenna ports at the network node. To further explain this, the channel representation in the antenna-frequency domain is usually rich and hard to compress, however, its equivalent form in the beam-delay domain is sparse and easier to compress. Such sparsity, to some extent, reflects the physical interpretation of a propagation channel. That is, it reflects how the numerous sinusoidal signals traverse from the transmitting end, along different paths, to the receiving end. Essentially, each beam can be associated with a certain direction of a propagation path, and each delay can reflect the relative difference in distance if a signal propagates along different paths. Each pair of beam and delay may be associated with a single propagation path, if there is infinite spatial resolution and delay resolution. In real propagation environment, dominant paths that contribute to conveying a signal are usually sparse if looking at the whole 3D space, since the signal cannot reach to the receiver end from any direction. Among other reasons, this is limited by the antenna directivity and the number of antenna elements deployed at both the transmitter and the receiver, as well as the number of objects in the propagation environment that can reflect a signal without introducing significant loss. The above sparsity can be exploited to assist an AI/ML model. For example, the beam-delay domain transformation could help the AI/ML model with an initial feature extraction. Another advantage of this pre-processing is that the beam-delay transformation can be achieved using Fast Fourier Transforms (FFTs), for which there are already fast implementations with hardware support. The sparsity can be further exploited by 18 removing a number of insignificant beams and delays, so that the input dimensions could also be reduced with a marginal loss, likely resulting in smaller AI/ML models. The beam-delay transformation and feature extraction can be applied both cases of explicit channel feedback and eigenvector-based feedback. A brief example in described next for pre-processing of the eigenvector-based feedback, which has received immediate attention in 3GPP. The first step is that the wireless device measures the channel on CSI-RS. For example, let the wireless device have 4 Rx-ports, the configured CSI-format has 32 virtual Tx-ports, and the bandwidth are 52 Resource Blocks (RBs) corresponding to 10 MHz at 15 kHz subcarrier spacing. The feature extraction for eigenvector-based feedback is illustrated in FIG.6. The steps are as follows: 1. The wireless device does a spatial domain DFT on the 32x4 matrix per RB and selects the ^^ strongest beams out of 16 (for one polarization) (Block S10). This is done in a wideband manner, including the spatial oversampling of the spatial-domain (SD) basis, and the same beams are used for both polarizations. The covariance of the beam-space channel is summed over, e.g., 4 RBs to produce a covariance matrix for each sub-band. 2. For each covariance matrix (per sub-band) the device extracts a number of eigenvectors and may select the rank, i.e., number of layers (Block S12). 3. The device performs a frequency domain DFT per layer, transforming to delay domain, whereafter it selects the ^^ strongest taps (Block S12). The resulting tensor of dimensions 2 ^^ x number of layers x ^^ is called the linear combination coefficients and can be used to reconstruct, by the wireless device suggested, precoding matrices. 4. The tensor of linear combination coefficients is used as input in the AI/ML model (Block S14). The input could be further enhanced with information about the selected beams and taps, noise levels, etc. AE models implicit CSI feedback for ^^ ^^ ^ 1 19 ^{There can be several schemes for AI/ML models for implicit CSI feedback when RI ^} _{1, i.e., the indicated rank is greater than one. In this disclosure, the focus is on two} main categorizations of the AI/ML models when RI ^ 1: 1. Transmission layer common scheme: This scheme includes one AI/ML model, which is trained and deployed for all the transmission layers based on the estimated RI, as illustrated in FIG.7, where ^^ and ^^ _^^ are the estimated transmission channel and interference channel, respectively. 2. Transmission layer specific scheme: This scheme includes multiple AI/ML models, which are trained and deployed for each of the transmission layers based on the estimated RI as illustrated in FIG.8, where ^^ and ^^ _^^ are the estimated transmission channel and interference channel, respectively. Note that in the above examples, a model is trained for a transmission layer irrespective of the RI. Additionally, the model for a transmission layer can further depend on RI. However, the CSI report mechanism in the UCI may apply to both the cases where the model is either independent or dependent on RI. Some embodiments of the disclosure provide for RRC signaling and CSI reporting for AI-based CSI compression and feedback. Referring now to the drawing figures, in which like elements are referred to by like reference numerals, there is shown in FIG.9 a schematic diagram of a communication system 10, according to an embodiment, such as a 3GPP-type cellular network that may support standards such as LTE and/or NR (5G), which comprises an access network 12, such as a radio access network, and a core network 14. The access network 12 comprises a plurality of network nodes 16a, 16b, 16c (referred to collectively as network nodes 16), such as NBs, eNBs, gNBs or other types of wireless access points, each defining a corresponding coverage area 18a, 18b, 18c (referred to collectively as coverage areas 18). Each network node 16a, 16b, 16c is connectable to the core network 14 over a wired or wireless connection 20. A first wireless device 22a located in coverage area 18a is configured 20 to wirelessly connect to, or be paged by, the corresponding network node 16a. A second wireless device 22b in coverage area 18b is wirelessly connectable to the corresponding network node 16b. While a plurality of wireless devices 22a, 22b (collectively referred to as wireless devices 22) are illustrated in this example, the disclosed embodiments are equally applicable to a situation where a sole wireless device is in the coverage area or where a sole wireless device is connecting to the corresponding network node 16. Note that although only two wireless devices 22 and three network nodes 16 are shown for convenience, the communication system may include many more wireless devices 22 and network nodes 16. Also, it is contemplated that a wireless device 22 can be in simultaneous communication and/or configured to separately communicate with more than one network node 16 and more than one type of network node 16. For example, a wireless device 22 can have dual connectivity with a network node 16 that supports LTE and the same or a different network node 16 that supports NR. As an example, wireless device 22 can be in communication with an eNB for LTE/E-UTRAN and a gNB for NR/NG-RAN. The communication system 10 may itself be connected to a host computer 24, which may be embodied in the hardware and/or software of a standalone server, a cloud- implemented server, a distributed server or as processing resources in a server farm. The host computer 24 may be under the ownership or control of a service provider, or may be operated by the service provider or on behalf of the service provider. The connections 26, 28 between the communication system 10 and the host computer 24 may extend directly from the core network 14 to the host computer 24 or may extend via an optional intermediate network 30. The intermediate network 30 may be one of, or a combination of more than one of, a public, private or hosted network. The intermediate network 30, if any, may be a backbone network or the Internet. In some embodiments, the intermediate network 30 may comprise two or more sub-networks (not shown). The communication system of FIG.9 as a whole enables connectivity between one of the connected wireless devices 22a, 22b and the host computer 24. The connectivity may be described as an over-the-top (OTT) connection. The host computer 24 and the connected wireless devices 22a, 22b are configured to 21 communicate data and/or signaling via the OTT connection, using the access network 12, the core network 14, any intermediate network 30 and possible further infrastructure (not shown) as intermediaries. The OTT connection may be transparent in the sense that at least some of the participating communication devices through which the OTT connection passes are unaware of routing of uplink and downlink communications. For example, a network node 16 may not or need not be informed about the past routing of an incoming downlink communication with data originating from a host computer 24 to be forwarded (e.g., handed over) to a connected wireless device 22a. Similarly, the network node 16 need not be aware of the future routing of an outgoing uplink communication originating from the wireless device 22a towards the host computer 24. A network node 16 is configured to include a configuration unit 32 which is configured to perform one or more network node 16 functions described herein, including functions related to RRC signaling and CSI reporting for AI-based CSI compression and feedback. A wireless device 22 is configured to include an implementation unit 34 which is configured to perform one or more wireless device 22 functions described herein, including functions related to RRC signaling and CSI reporting for AI-based CSI compression and feedback. Example implementations, in accordance with an embodiment, of the wireless device 22, network node 16 and host computer 24 discussed in the preceding paragraphs will now be described with reference to FIG.10. In a communication system 10, a host computer 24 comprises hardware (HW) 38 including a communication interface 40 configured to set up and maintain a wired or wireless connection with an interface of a different communication device of the communication system 10. The host computer 24 further comprises processing circuitry 42, which may have storage and/or processing capabilities. The processing circuitry 42 may include a processor 44 and memory 46. In particular, in addition to or instead of a processor, such as a central processing unit, and memory, the processing circuitry 42 may comprise integrated circuitry for processing and/or control, e.g., one or more processors and/or processor cores and/or FPGAs (Field Programmable Gate Array) and/or ASICs (Application Specific Integrated Circuitry) adapted to execute instructions. The processor 44 may be configured to access (e.g., write to and/or read from) memory 46, which may comprise any kind of volatile and/or nonvolatile memory, e.g., cache 22 and/or buffer memory and/or RAM (Random Access Memory) and/or ROM (Read- Only Memory) and/or optical memory and/or EPROM (Erasable Programmable Read-Only Memory). Processing circuitry 42 may be configured to control any of the methods and/or processes described herein and/or to cause such methods, and/or processes to be performed, e.g., by host computer 24. Processor 44 corresponds to one or more processors 44 for performing host computer 24 functions described herein. The host computer 24 includes memory 46 that is configured to store data, programmatic software code and/or other information described herein. In some embodiments, the software 48 and/or the host application 50 may include instructions that, when executed by the processor 44 and/or processing circuitry 42, causes the processor 44 and/or processing circuitry 42 to perform the processes described herein with respect to host computer 24. The instructions may be software associated with the host computer 24. The software 48 may be executable by the processing circuitry 42. The software 48 includes a host application 50. The host application 50 may be operable to provide a service to a remote user, such as a wireless device 22 connecting via an OTT connection 52 terminating at the wireless device 22 and the host computer 24. In providing the service to the remote user, the host application 50 may provide user data which is transmitted using the OTT connection 52. The “user data” may be data and information described herein as implementing the described functionality. In one embodiment, the host computer 24 may be configured for providing control and functionality to a service provider and may be operated by the service provider or on behalf of the service provider. The processing circuitry 42 of the host computer 24 may enable the host computer 24 to observe, monitor, control, transmit to and/or receive from the network node 16 and or the wireless device 22. The processing circuitry 42 of the host computer 24 may include a control unit 54 configured to enable the service provider to observe/monitor/ control/transmit to/receive from the network node 16 and or the wireless device 22. The communication system 10 further includes a network node 16 provided in a communication system 10 and including hardware 58 enabling it to communicate with the host computer 24 and with the wireless device 22. The hardware 58 may 23 include a communication interface 60 for setting up and maintaining a wired or wireless connection with an interface of a different communication device of the communication system 10, as well as a radio interface 62 for setting up and maintaining at least a wireless connection 64 with a wireless device 22 located in a coverage area 18 served by the network node 16. The radio interface 62 may be formed as or may include, for example, one or more RF transmitters, one or more RF receivers, and/or one or more RF transceivers. The communication interface 60 may be configured to facilitate a connection 66 to the host computer 24. The connection 66 may be direct or it may pass through a core network 14 of the communication system 10 and/or through one or more intermediate networks 30 outside the communication system 10. In the embodiment shown, the hardware 58 of the network node 16 further includes processing circuitry 68. The processing circuitry 68 may include a processor 70 and a memory 72. In particular, in addition to or instead of a processor, such as a central processing unit, and memory, the processing circuitry 68 may comprise integrated circuitry for processing and/or control, e.g., one or more processors and/or processor cores and/or FPGAs (Field Programmable Gate Array) and/or ASICs (Application Specific Integrated Circuitry) adapted to execute instructions. The processor 70 may be configured to access (e.g., write to and/or read from) the memory 72, which may comprise any kind of volatile and/or nonvolatile memory, e.g., cache and/or buffer memory and/or RAM (Random Access Memory) and/or ROM (Read-Only Memory) and/or optical memory and/or EPROM (Erasable Programmable Read-Only Memory). Thus, the network node 16 further has software 74 stored internally in, for example, memory 72, or stored in external memory (e.g., database, storage array, network storage device, etc.) accessible by the network node 16 via an external connection. The software 74 may be executable by the processing circuitry 68. The processing circuitry 68 may be configured to control any of the methods and/or processes described herein and/or to cause such methods, and/or processes to be performed, e.g., by network node 16. Processor 70 corresponds to one or more processors 70 for performing network node 16 functions described herein. The memory 72 is configured to store data, programmatic software code and/or other information described herein. In some embodiments, the software 74 may include instructions 24 that, when executed by the processor 70 and/or processing circuitry 68, causes the processor 70 and/or processing circuitry 68 to perform the processes described herein with respect to network node 16. For example, processing circuitry 68 of the network node 16 may include configuration unit 32 configured to perform one or more network node 16 functions described herein, including functions related to RRC signaling and CSI reporting for AI-based CSI compression and feedback. The communication system 10 further includes the wireless device 22 already referred to. The wireless device 22 may have hardware 80 that may include a radio interface 82 configured to set up and maintain a wireless connection 64 with a network node 16 serving a coverage area 18 in which the wireless device 22 is currently located. The radio interface 82 may be formed as or may include, for example, one or more RF transmitters, one or more RF receivers, and/or one or more RF transceivers. The hardware 80 of the wireless device 22 further includes processing circuitry 84. The processing circuitry 84 may include a processor 86 and memory 88. In particular, in addition to or instead of a processor, such as a central processing unit, and memory, the processing circuitry 84 may comprise integrated circuitry for processing and/or control, e.g., one or more processors and/or processor cores and/or FPGAs (Field Programmable Gate Array) and/or ASICs (Application Specific Integrated Circuitry) adapted to execute instructions. The processor 86 may be configured to access (e.g., write to and/or read from) memory 88, which may comprise any kind of volatile and/or nonvolatile memory, e.g., cache and/or buffer memory and/or RAM (Random Access Memory) and/or ROM (Read-Only Memory) and/or optical memory and/or EPROM (Erasable Programmable Read-Only Memory). Thus, the wireless device 22 may further comprise software 90, which is stored in, for example, memory 88 at the wireless device 22, or stored in external memory (e.g., database, storage array, network storage device, etc.) accessible by the wireless device 22. The software 90 may be executable by the processing circuitry 84. The software 90 may include a client application 92. The client application 92 may be operable to provide a service to a human or non-human user via the wireless device 22, with the support of the host computer 24. In the host computer 24, an 25 executing host application 50 may communicate with the executing client application 92 via the OTT connection 52 terminating at the wireless device 22 and the host computer 24. In providing the service to the user, the client application 92 may receive request data from the host application 50 and provide user data in response to the request data. The OTT connection 52 may transfer both the request data and the user data. The client application 92 may interact with the user to generate the user data that it provides. The processing circuitry 84 may be configured to control any of the methods and/or processes described herein and/or to cause such methods, and/or processes to be performed, e.g., by wireless device 22. The processor 86 corresponds to one or more processors 86 for performing wireless device 22 functions described herein. The wireless device 22 includes memory 88 that is configured to store data, programmatic software code and/or other information described herein. In some embodiments, the software 90 and/or the client application 92 may include instructions that, when executed by the processor 86 and/or processing circuitry 84, causes the processor 86 and/or processing circuitry 84 to perform the processes described herein with respect to wireless device 22. For example, the processing circuitry 84 of the wireless device 22 may include an implementation unit 34 configured to perform one or more wireless device 22 functions described herein, including functions related to RRC signaling and CSI reporting for AI-based CSI compression and feedback. In some embodiments, the inner workings of the network node 16, wireless device 22, and host computer 24 may be as shown in FIG.10 and independently, the surrounding network topology may be that of FIG.9. In FIG.10, the OTT connection 52 has been drawn abstractly to illustrate the communication between the host computer 24 and the wireless device 22 via the network node 16, without explicit reference to any intermediary devices and the precise routing of messages via these devices. Network infrastructure may determine the routing, which it may be configured to hide from the wireless device 22 or from the service provider operating the host computer 24, or both. While the OTT connection 52 is active, the network infrastructure may further take decisions by 26 which it dynamically changes the routing (e.g., on the basis of load balancing consideration or reconfiguration of the network). The wireless connection 64 between the wireless device 22 and the network node 16 is in accordance with the teachings of the embodiments described throughout this disclosure. One or more of the various embodiments improve the performance of OTT services provided to the wireless device 22 using the OTT connection 52, in which the wireless connection 64 may form the last segment. More precisely, the teachings of some of these embodiments may improve the data rate, latency, and/or power consumption and thereby provide benefits such as reduced user waiting time, relaxed restriction on file size, better responsiveness, extended battery lifetime, etc. In some embodiments, a measurement procedure may be provided for the purpose of monitoring data rate, latency and other factors on which the one or more embodiments improve. There may further be an optional network functionality for reconfiguring the OTT connection 52 between the host computer 24 and wireless device 22, in response to variations in the measurement results. The measurement procedure and/or the network functionality for reconfiguring the OTT connection 52 may be implemented in the software 48 of the host computer 24 or in the software 90 of the wireless device 22, or both. In embodiments, sensors (not shown) may be deployed in or in association with communication devices through which the OTT connection 52 passes; the sensors may participate in the measurement procedure by supplying values of the monitored quantities exemplified above, or supplying values of other physical quantities from which software 48, 90 may compute or estimate the monitored quantities. The reconfiguring of the OTT connection 52 may include message format, retransmission settings, preferred routing etc.; the reconfiguring need not affect the network node 16, and it may be unknown or imperceptible to the network node 16. Some such procedures and functionalities may be known and practiced in the art. In certain embodiments, measurements may involve proprietary wireless device signaling facilitating the host computer’s 24 measurements of throughput, propagation times, latency and the like. In some embodiments, the measurements may be implemented in that the software 48, 90 causes messages to be transmitted, in particular empty or ‘dummy’ messages, using the OTT connection 52 while it monitors propagation times, errors, etc. 27 Thus, in some embodiments, the host computer 24 includes processing circuitry 42 configured to provide user data and a communication interface 40 that is configured to forward the user data to a cellular network for transmission to the wireless device 22. In some embodiments, the cellular network also includes the network node 16 with a radio interface 62. In some embodiments, the network node 16 is configured to, and/or the network node’s 16 processing circuitry 68 is configured to perform the functions and/or methods described herein for preparing/initiating/maintaining/supporting/ending a transmission to the wireless device 22, and/or preparing/terminating/maintaining/supporting/ending in receipt of a transmission from the wireless device 22. In some embodiments, the host computer 24 includes processing circuitry 42 and a communication interface 40 that is configured to a communication interface 40 configured to receive user data originating from a transmission from a wireless device 22 to a network node 16. In some embodiments, the wireless device 22 is configured to, and/or comprises a radio interface 82 and/or processing circuitry 84 configured to perform the functions and/or methods described herein for preparing/initiating/maintaining/supporting/ending a transmission to the network node 16, and/or preparing/terminating/maintaining/supporting/ending in receipt of a transmission from the network node 16. Although FIG.9 and 10 show various “units” such as configuration unit 32, and implementation unit 34 as being within a respective processor, it is contemplated that these units may be implemented such that a portion of the unit is stored in a corresponding memory within the processing circuitry. In other words, the units may be implemented in hardware or in a combination of hardware and software within the processing circuitry. FIG.11 is a flowchart illustrating an example method implemented in a communication system, such as, for example, the communication system of FIG.9 and 10, in accordance with one embodiment. The communication system may include a host computer 24, a network node 16 and a wireless device 22, which may be those described with reference to FIG.10. In a first step of the method, the host computer 24 provides user data (Block S100). In an optional substep of the first step, the host computer 24 provides the user data by executing a host application, such 28 as, for example, the host application 50 (Block S102). In a second step, the host computer 24 initiates a transmission carrying the user data to the wireless device 22 (Block S104). In an optional third step, the network node 16 transmits to the wireless device 22 the user data which was carried in the transmission that the host computer 24 initiated, in accordance with the teachings of the embodiments described throughout this disclosure (Block S106). In an optional fourth step, the wireless device 22 executes a client application, such as, for example, the client application 92, associated with the host application 50 executed by the host computer 24 (Block S108). FIG.12 is a flowchart illustrating an example method implemented in a communication system, such as, for example, the communication system of FIG.9, in accordance with one embodiment. The communication system may include a host computer 24, a network node 16 and a wireless device 22, which may be those described with reference to FIG.9 and 10. In a first step of the method, the host computer 24 provides user data (Block S110). In an optional substep (not shown) the host computer 24 provides the user data by executing a host application, such as, for example, the host application 50. In a second step, the host computer 24 initiates a transmission carrying the user data to the wireless device 22 (Block S112). The transmission may pass via the network node 16, in accordance with the teachings of the embodiments described throughout this disclosure. In an optional third step, the wireless device 22 receives the user data carried in the transmission (Block S114). FIG.13 is a flowchart illustrating an example method implemented in a communication system, such as, for example, the communication system of FIG.9, in accordance with one embodiment. The communication system may include a host computer 24, a network node 16 and a wireless device 22, which may be those described with reference to FIG.9 and 10. In an optional first step of the method, the wireless device 22 receives input data provided by the host computer 24 (Block S116). In an optional substep of the first step, the wireless device 22 executes the client application 92, which provides the user data in reaction to the received input data provided by the host computer 24 (Block S118). Additionally or alternatively, in an optional second step, the wireless device 22 provides user data (Block S120). In an optional substep of the second step, the wireless device provides the user data by executing a client application, such as, for example, client application 92 (Block 29 S122). In providing the user data, the executed client application 92 may further consider user input received from the user. Regardless of the specific manner in which the user data was provided, the wireless device 22 may initiate, in an optional third substep, transmission of the user data to the host computer 24 (Block S124). In a fourth step of the method, the host computer 24 receives the user data transmitted from the wireless device 22, in accordance with the teachings of the embodiments described throughout this disclosure (Block S126). FIG.14 is a flowchart illustrating an example method implemented in a communication system, such as, for example, the communication system of FIG.9, in accordance with one embodiment. The communication system may include a host computer 24, a network node 16 and a wireless device 22, which may be those described with reference to FIG.9 and 10. In an optional first step of the method, in accordance with the teachings of the embodiments described throughout this disclosure, the network node 16 receives user data from the wireless device 22 (Block S128). In an optional second step, the network node 16 initiates transmission of the received user data to the host computer 24 (Block S130). In a third step, the host computer 24 receives the user data carried in the transmission initiated by the network node 16 (Block S132). FIG.15A is a flowchart of an example process in a network node 16. One or more blocks described herein may be performed by one or more elements of network node 16 such as by one or more of processing circuitry 68 (including the configuration unit 32), processor 70, radio interface 62 and/or communication interface 60. Network node 16 is configured to transmit an indication to the wireless device to cause a generation of an artificial intelligence based, AI-based, channel state information, CSI, report (Block S134). Network node 16 is configured to receive the AI-based CSI report (Block S136). Network node 16 is configured to perform at least one action based on the received CSI report (Block S138). In at least one embodiment, the indication includes at least one of: a restriction on a rank that can be reported by the wireless device; at least one configuration for at least one of a spatial domain, SD, a frequency domain, FD, and a time domain, TD, the configuration being used to pre-process eigenvectors; at least one configuration for the quantization bits that can be used by the wireless device to quantize the 30 eigenvectors; at least one configuration for a number of latent-space coefficients at an output of the AI model; and an indication for the wireless device to use at least one of transmission layer-common processing and transmission layer-specific processing to generate the CSI report. In at least one embodiment, the network node 16 is configured to indicate at least one parameter to be used in the generation of the CSI report, the at least one parameter being indicated on at least one of a downlink control information, DCI, and a medium access control-control element, MAC-CE. FIG.15B is a flowchart of an example process in a network node 16. One or more blocks described herein may be performed by one or more elements of network node 16 such as by one or more of processing circuitry 68 (including the configuration unit 32), processor 70, radio interface 62 and/or communication interface 60. The network node 16 has one or more decoders of one or more autoencoders available. Network node 16 is configured to indicate (Block S135), through RRC signaling, that a wireless device is to transmit a CSI report compressed by means of an autoencoder. Network node 16 is configured to receive a first CSI report and a second CSI report on different parts of UCI (Block S137). Network node 16 is configured to use (Block S139) a decoder from the one or more decoders to decode the first CSI report and the second CSI report. In at least one embodiment, the indication includes at least one of: a restriction on a rank that can be reported by the wireless device; at least one configuration for at least one of a spatial domain, SD, a frequency domain, FD, and a time domain, TD, the configuration being used to pre-process eigenvectors; at least one configuration for the quantization bits that can be used by the wireless device to quantize the eigenvectors; at least one configuration for a number of latent-space coefficients at an output of the AI model; and an indication for the wireless device to use at least one of transmission layer-common processing and transmission layer-specific processing to generate the CSI report. In at least one embodiment, the network node 16 is configured to indicate at least one parameter to be used in the generation of the CSI report, the at least one parameter being indicated on at least one of a downlink control information, DCI, and a medium access control-control element, MAC-CE. 31 FIG.16A is a flowchart of an example process in a wireless device 22 according to some embodiments of the present disclosure. One or more blocks described herein may be performed by one or more elements of wireless device 22 such as by one or more of processing circuitry 84 (including the implementation unit 34), processor 86, radio interface 82 and/or communication interface 60. Wireless device 22 is configured to receive an indication from the network node to generate an artificial intelligence based, AI-based, channel state information, CSI, report (Block S140). Wireless device 22 is configured to generate the CSI report (Block S142). Wireless device 22 is configured to transmit the CSI report to the network node (Block S144). In at least one embodiment, the indication includes at least one of: a restriction on a rank that can be reported by the wireless device; at least one configuration for at least one of a spatial domain, SD, a frequency domain, FD, and a time domain, TD, the configuration being used to pre-process eigenvectors; at least one configuration for the quantization bits that can be used by the wireless device to quantize the eigenvectors; at least one configuration for a number of latent-space coefficients at an output of the AI model; and an indication for the wireless device to use at least one of transmission layer-common processing and transmission layer-specific processing to generate the CSI report. In at least one embodiment, the wireless device 22 is configured to receive at least one parameter to be used in the generation of the CSI report, the at least one parameter being received on at least one of a downlink control information, DCI, and a medium access control-control element, MAC-CE. FIG.16B is a flowchart of an example process in a wireless device 22 according to some embodiments of the present disclosure. One or more blocks described herein may be performed by one or more elements of wireless device 22 such as by one or more of processing circuitry 84 (including the implementation unit 34), processor 86, radio interface 82 and/or communication interface 60. Wireless device 22 has one or more encoders of one or more autoencoders available. Wireless device 22 may be configured to receive, through RRC signaling, an indication from the network node to generate a channel state information, CSI, report (Block S141) using an encoder. Wireless device 22 is configured to generate the CSI report (Block S143) using an encoder. Wireless device 22 is configured to segment the output of the encoder into 32 a first CSI report and a second CSI report, wherein the first CSI report and the second CSI report are transmitted on different parts of an UCI (Block S145). In at least one embodiment, the indication includes at least one of: a restriction on a rank that can be reported by the wireless device; at least one configuration for at least one of a spatial domain, SD, a frequency domain, FD, and a time domain, TD, the configuration being used to pre-process eigenvectors; at least one configuration for the quantization bits that can be used by the wireless device to quantize the eigenvectors; at least one configuration for a number of latent-space coefficients at an output of the AI model; and an indication for the wireless device to use at least one of transmission layer-common processing and transmission layer-specific processing to generate the CSI report. In at least one embodiment, the wireless device 22 is configured to receive at least one parameter to be used in the generation of the CSI report, the at least one parameter being received on at least one of a downlink control information, DCI, and a medium access control-control element, MAC-CE. Having described the general process flow of arrangements of the disclosure and having provided examples of hardware and software arrangements for implementing the processes and functions of the disclosure, the sections below provide details and examples of arrangements for RRC signaling and CSI reporting for AI-based CSI compression and feedback. One or more wireless device 22 functions described below may be performed by one or more of processing circuitry 84, processor 86, implementation unit 34, etc. One or more network node 16 functions described below may be performed by one or more of processing circuitry 68, processor 70, configuration unit 32, etc. In at least one embodiment, the wireless device 22 estimates the DL channel based on the configured DL reference signals (e.g., CSI-RS, Demodulation Reference Signal (DMRS)), and produces a channel estimate ^^, for example, in the antenna- frequency domain. The raw channel ^^ can be expressed per CSI-RS port (TX side), per receive antenna (RX side), per frequency sub-band, and measured at one or more points in time. Hence, in the most general cases, the channel ^^ is a four- dimensional matrix or tensor. 33 The raw channel estimate ^^ (possibly together with interference channel) is leveraged to estimate the appropriate rank for the downlink transmission and further processed to extract the eigenvector corresponding to each layer according to the estimated rank ^^. The eigenvectors per transmission layer based on ^^ is denoted by ^^_^,^^, where ^^_^ ൌ 1, 2, … , ^^. For measurements with CSI-RS at a single time instance, ^{^^^,^^ is a tensor with dimension equal to number of CSI െ RS ports x number of layers x} _{number of frequency subbands. The extracted ^^^,^^ are compressed and quantized at} the encoder into bits, such that ^^_^^ represents the bits for quantizing the ^^_^-th transmission layer. Subsequently, the concatenated bits across all the transmission layers, denoted by ^^_AE ൌ ^ ^^_^, ^^_ଶ, … , ^^_^^^, along with the rank indication (RI) and channel quality indicator (CQI), is reported back to the network node 16 as part of the uplink CSI report. The CSI report comprising of ^^_AE and the legacy parameters computed from the estimated channel ^^, i.e., RI and CQI, are fed to the decoder deployed at the network node 16 to reconstruct the eigenvectors per layer, denoted by ^^^_^,^^ , ^^_^ ൌ 1, 2, … , ^^. The network node 16 can further process the eigenvectors to obtain the for each layer, denoted by ^^_^,^^ , ^^_^ ൌ 1, 2, … , ^^, for the transmission of the PDSCH. The dimension of raw ^^_^,^^ can be very large depending on the number of CSI-RS ports and the number of subbands, which can make the AE model and training complex. Accordingly, ^^ may be further pre-processed to have reduced dimension compared to the raw eigenvectors per layer based on feature extraction of the eigenvectors. As described herein, the pre-processing of the channel to extract features of eigenvectors per layer in the beam-delay domain with ^^ SD basis and ^^ delay-taps (through ^^ FD basis), results in a linear combination coefficient tensor of dimensions 2 ^^ x number of layers x ^^, denoted by ^^_ଶ. In a more specific implementation, values for ^^ and ^^ can be chosen from Rel-16 Type-II pre- processing, as specified, e.g., in 3GPP TS 38.214 ( ^^ depends on the value of ^^_௩ in, e.g., 3GPP TS 38.214, where ^^ is the layer index). With the above pre-processing, the encoder at the wireless device 22 compresses and quantizes ^^_ଶ, where the reduced dimension of ^^_ଶ can lead to reduced AE model size and lower training complexity. As described herein, the feedback of NZCs of ^^_ଶ contributes to major overhead for Type-II, which can be reduced leveraging feedback through an AE. The 34 above pre-processing for eigenvectors may require the wireless device 22 to explicitly feedback ^^ SD and ^^ FD basis back to the network node 16 as part of the uplink CSI report in the UCI. Accordingly, the indices of the ^^ SD and ^^ FD basis are encoded into bits, denoted as ^^_model, and reported to the network node 16, as part of the uplink CSI report. Based on the above, the processing of the channel to produce the CSI report in the UCI to generate the precoders for each transmitted layer through the autoencoder (AE) is shown in FIG.17, which depicts architecture of the channel eigenvector feedback approach for the CSI report in the UCI to generate the precoders for each transmitted layer at the network node 16. The per-layer input of the encoder is called eigenvectors. However, the term eigenvector may also be used in a wide sense that incorporates different ways for the wireless device 22 to extract precoding information for different layers. Based on the above, a standardized format of CSI report (e.g., what to report, and how to report) enables the wireless device 22 to efficiently compress and report the CSI, and then the network node 16 can correctly retrieve the CSI according to the reported CSI. Accordingly, detailed reporting mechanisms (what quantities to report and how to report them) for the AI-based implicit CSI feedback based on the eigenvector decomposition of the estimated channel per transmission layer based on RI. RRC configuration for AI-based CSI reporting The network node 16 can RRC configure various parameters, which can determine the size of the CSI report, and help the network node 16 to correctly decode the CSI report received from the wireless device 22. Specifically, the network node 16 can explicitly configure parameters like the model ID to identify the AI model to use, the number of quantization bits to use per transmission layer at the wireless device 22, any restrictions on the reported rank and/or the precoder, etc. TABLE 1 shows an AI-based field in the CodebookConfig Information Element (IE) for RRC configuration of CSI report CodebookConfig-ai ::= SEQUENCE { codebookType CHOICE { 35 typeAI SEQUENCE { n1-n2-codebookSubsetRestriction-ai CHOICE { two-one BIT STRING (SIZE (8)), two-two BIT STRING (SIZE (64)), four-one BIT STRING (SIZE (16)), three-two BIT STRING (SIZE (96)), six-one BIT STRING (SIZE (24)), four-two BIT STRING (SIZE (128)), eight-one BIT STRING (SIZE (192)), four-three BIT STRING (SIZE (192)), six-two BIT STRING (SIZE (48)), twelve-one BIT STRING (SIZE (96)), four-four BIT STRING (SIZE (256)), eight-two BIT STRING (SIZE (256)), sixteen-one BIT STRING (SIZE (64)) }, typeAI-RI-Restriction BIT STRING (SIZE(4)), typeAI-QB-Restriction BIT STRING (SIZE(5)), paramCombination-ai INTEGER (1..P), typeAI-Layer-Common BOOLEAN } } } 36 • In at least one embodiment, the network node 16 can configure the use of the AI-model based PMI reporting by setting the reportQuantity in the CSI-reportConfig IE to a new value ‘cri-RI-aiPMI-CQI’. • In at least one embodiment, a new codebook type ‘typeAI’ can be included under ‘codebookType’ in the CodebookConfig IE as shown in TABLE 1. The network node 16 can RRC configure parameters for CSI reporting similar to the legacy ‘type1’ or ‘type2’ codebook type, which can include additional parameters according to the AI model deployed by the network node 16 and/or the wireless device 22. Note that the ‘typeAI’ can also include the legacy parameters like ‘n1-n2- codebookSubsetRestriction-ai’ and ‘typeAI-RI-Restriction’, which restrict the beams that can be used for precoding and the rank that can be reported by the wireless device 22, respectively, as specified in, e.g., 3GPP TS 38.331 and 38.212. The following additional parameters specific to AI-based CSI reporting can be configured under codebook type ‘typeAI’: i. Quantization bit restriction ‘typeAI-QB-Restriction’ field, which can restrict the quantization bits that can be used to quantize the latent-space coefficient. o In at least one embodiment, if the maximum quantization bits that can be processed by the AI model at the network node 16 is Q, then a bit string of length 4 is specified, say ^^_ொି^, … , ^^_^, ^^_^, where ^^_^ is the LSB (corresponds to quantization bit of 1) and ^^_ொି^ (corresponds to quantization bit of Q ) is the MSB. The network node 16 can further restrict the value of quantization bit that can be used at the wireless device 22 to a subset of allowed number of quantization bits through the bit string. For example, if Q=4 and the allowed number of quantization bits are restricted to 2 and 4, then the bit string can be set to ^ ^^_ଷ, ^^_ଶ, ^^_^, ^^_^^ ൌ ^1,0,1,0^. o^{In at least one embodiment, this restriction can be configured by} l_{ogଶ ^^ொ^ bits, where ^^ொ^ denotes the number of possibilities for} quantization bits. For example, if 2 and 4 bits quantization of the latent space variable is supported by the encoder/decoder, then 37 ^^_ொ^ = 2, and ‘0’ can be used to configure 2-bit quantization while ‘1’ can be used to configure 4-bit quantization. o In at least one embodiment, the number of quantization bits is rank dependent. In this case, one way to configure the number of quantization bits is to pre-define a number of configurations. For example, two configurations are pre-defined, one being ^ _{^^^, ^^ଶ, ^^ଷ, ^^ସ} ^{^} _{ൌ ^4,4,2,2^ and the other being} ^{^} _{^^^, ^^ଶ, ^^ଷ, ^^ସ} ^{^} _ൌ ^2,2,2,2^, where ^^_^ for ^^ ൌ 1,2,3,4 is the number of bits for quantizing layer r. Then, the network node 16 can configure one of the pre- defined configurations to the wireless device 22 for AI CSI reporting. o In at least one embodiment, there are a number of pre-defined hypotheses for the number of quantization bits. In a dependent claim, the wireless device 22 is free to determine the hypothesis and the wireless device 22 will report the hypothesis used as part of the CSI report. In at least one embodiment, the network node 16 configures a subset of hypotheses and the wireless device 22 reports with hypotheses within the configured subset is used for CSI reporting as part of the CSI report. o Note that in the above, scalar quantization is used as the example. Other embodiments may use other type of quantization methods, e.g., vector quantization, in which, instead of the number of quantization bits, the wireless device 22 may be configured with optionally, or additionally, the quantization codebook size. Further, the wireless device 22 may also be additionally configured with a more detailed aspect of the quantization method. In scalar quantization, for example, the mechanism to determine the quantization points may also be configured, e.g., tanh-based quantization, uniform (or non-uniform) quantization, etc. ii. Similar to the legacy Rel-16 Type-II codebook, the parameters ^ ^^, ^^_జ^ may need to be configured at the wireless device 22, where ^^ and ^^_జ give the number of spatial domain basis and number of frequency domain basis to pre-process the eigenvectors per transmission layer. o In at least one embodiment, the values of ^ ^^, ^^_జ^ are implicitly determined by the AI model deployed at the network node 16 and signaled to the wireless device 22 through the model ID. In a related embodiment, when a network node 16 and a wireless device 22 is configured with multiple models to handle a range of ^ ^^, ^^_జ^, a ‘paramCombination-ai’ field can be included in the CSI- reportConfig IE, to explicitly signal the ^ ^^, ^^_జ^ to the wireless device 22 for pre-processing the eigenvector per transmission layer, where: o In at least one embodiment, the value of ^ ^^, ^^_జ^ is comparable to legacy Rel-16 Type-II preprocessing, with value of ^^_జ being constant across the reported rank, as shown in TABLE 2. Further, additional high-resolution parameters, i.e., higher value for ^^ and ^^_జ can be included, for example, paramCombination index 8 in TABLE 2. Additionally, network node 16 can configure only spatial domain pre-preprocessing or only frequency domain pre-preprocessing by including paramCombination index 9 and 10, respectively. Finally, in another embodiment, network node 16 can configure the wireless device 22 to compress and feedback the raw eigenvector per layer by including the paramCombination index 11. - TABLE 2 - Codebook parameter configurations for ^^, β, ^^_జ and ^^_^,జ paramCombination‐ai ^^ ^^_జ, ^^ ൌ ^1,2,3,4^ 1 ₂ ¹ _{ൗ 4} 2 ₂ ¹ _{ൗ 4} 3^{4 1} _{ൗ 4} 4 ₄ ¹ _{ൗ 4} 5 ₄ ¹ _{ൗ 4} ⁶ ₄ ¹ _{ൗ 2} 7^{6 1} _{ൗ 2} 8^{8 1} _{ൗ 2} 9 8 ‐ 1^{0 ‐ 1} _{ൗ 2} 11 ‐ ‐ o In at least one embodiment, where the values of ^ ^^, ^^_జ^ are implicitly signaled by the network node 16, the ‘paramCombination-ai’ field can include configuration for the number of active latent-space coefficient per transmission layer, given by ratio of latent-space coefficients active and denoted by ^^_^,జ, and/or number of quantization bits for active latent-space coefficient per layer, ^^_^,జ, as shown in TABLE 3. Note that the values of ^^_^,జ, can either be explicitly specified or can be the function of allowed quantization bits configured by ‘typeAI-QB-Restriction’ field. TABLE 3 - Codebook parameter configurations for ^^_^,జ and ^^_^,జ ^^_^,జ ^^_^,జ paramCombination- ai ^^ ^{^^ ∈ ^1,2^ ^^} ∈_^1,2^ ^{^^ ∈ ^3,4^} ∈_^3,4^ 1 ₁ ¹ _{ൗ 4} ^{4 2} ² ₁ ¹ _{ൗ 2} ^{4 2} ³ ₁ ¹ _{ൗ 2} ^{4 2} ⋮ 40 ^^ 1 1 4 4 o In at least one embodiment, the ‘paramCombination-ai’ field can contain the subset of parameter configuration mentioned above, where the inclusion of additional parameter configuration is not precluded. o In at least one embodiment, the ‘paramCombination-ai’ field can be excluded from the RRC configuration and the parameters ^ ^^, ^^_జ^ can be either implicitly signaled to the wireless device 22 by the network node 16 through the model ID or can be configured and signaled (explicitly or implicitly) to the network node 16 by the wireless device 22 (further discussed below). iii. Use of layer-common or layer-specific processing can be configured by network node 16 by the Boolean field ‘typeAI-Layer-Common’. The network node 16 can signal the wireless device 22 to use the layer-common model for compressing all the transmission layers with same parameters by activating the above parameter. • In at least one embodiment, a subset of the above parameters is included for codebook type ‘typeAI’. The inclusion of additional parameters to further enhance the configuration by the network node 16 for AI-based CSI reporting is not precluded. In another embodiment, the codebook type ‘typeAI’ can further divided into subtypes based on if network node 16 configure the CSI reporting to layer- common or layer specific, where any other criteria for defining the subtypes for codebook type ‘typeAI’ is not precluded. • In at least one embodiment, a subset of parameters related to the AI model, described in this section, can either be dynamically signaled by the network node 16 to the wireless device 22 through DCI and/or MAC-CE or by the wireless device 22 to the network node 16 in the UCI. • In at least one embodiment, as shown in TABLE 1, the network node 16 can also RRC configure multiple AI-based CSI report settings. Network node 16 can then indicate which of the CSI report setting the wireless device 22 shall use for calculating a CSI report. This can be performed in multiple ways. For example, each of the multiple configured CSI reporting settings can be associated with one CSI- 41 SemiPersistentOnPUSCH-TriggerState or CSI-AperiodicTriggerState, then the network node 16 indicates which CSI report setting the wireless device 22 may use for calculating a CSI report via MAC-CE and/or DCI. • For each of the configurations above, one value (or one combination or one index) may serve as the default value. For example, the default value may be the first value from multiple values of parameter configured for the wireless device 22. The default value may be used, e.g., for the case of the wireless device 22 is not able to receive DCI. UCI configuration for AI-based CSI reporting framework Further CSI reporting methodology for the AI-based implicit CSI feedback described herein, which mainly focus on enhancing the legacy UCI framework. Following the legacy structure, the CSI report is segmented into Part 1 CSI and Part 2 CSI, where Part 2 CSI can be further divided into sub-segments to cater the requirements of the feedback mechanism of the deployed AI model. The CSI reports carried on Part 1 and Part 2 are part of the uplink control information (UCI), which can either be carried on PUCCH or PUSCH. The common report quantities for the AI-based CSI and the legacy CSI (e.g., Type I/II), such as CSI-RS resource indicator (CRI), rank indicator (RI) and channel quality indicator (CQI) are reported in Part 1 CSI. As discussed herein, the pre-processing is carried out to extract the features of eigenvectors per transmission layer in the beam-delay domain with ^^ SD basis and ^^ FD basis, which are feedback back with ^^_model bits. The bit sequence ^^_model has a predefined order so that network node 16 knows how to map the corresponding part of ^^_model to a certain extracted feature. Accordingly, the number of bits to signal ^ ^^, ^^^ SD and FD basis can be carried on Part 1 CSI report of the UCI. In at least one embodiment, the number of bits to signal ^ ^^, ^^^ SD and FD basis, is implicitly mapped to the AI model. Such model could configured by the network node 16 or the wireless device 22. When the model is configured by the wireless device 22, then the selected model needs to be reported to the network node 16. The selected model, identified by model ID, can be reported to the network node 16 in CSI Part 1. The ^^_model and ^^_AE bits to extract the transmission layer information form the Part 2 of the CSI report in the UCI. To prioritize the wideband large-scale features of the 42 eigenvectors per transmission layer, the ^^_model bits to signal the pre-processing information is reported with higher priority compared to the ^^_AE bits extracted from the AE-based processing step. Note that since the pre-processing step is optional, as described herein, then, in that case, reporting of ^^_model (and the number of bits in to signal ^^_model) is not needed, i.e., when raw eigenvector per transmission layer is compressed and reported by the wireless device 22. Example Embodiments related to the bitwidth and the segmentation of ^^_AE The number of bits generated from the AI model at the wireless device 22, i.e., the bitwidth of ^^_AE, can be implicitly mapped to the AI model. Such model could be either configured by the network node 16 or the wireless device 22. When the model is configured by the wireless device 22, then the selected model (through model ID) is reported to the network node 16 in CSI Part 1. In various embodiments, further methodologies for signaling the bitwidth of the ^^_AE for AI-based implicit CSI feedback, to the network node 16 is discussed. Specifically, the bitwidth of ^^_AE can implicitly depend on the auxiliary information for each eigenvector per transmission layer, for example, the number of active latent-space coefficient and/or the number of quantization bit used per latent-space coefficient to process a specific transmission layer. • In at least one embodiment, the auxiliary information can be tied to the model ID, which can be configured by network node 16 or signaled to the network node 16 by the wireless device 22. With the auxiliary information, along with the bitwidth, the network node 16 can determine the proper sequence of ^^_AE bits to decode all the transmission layers. In at least one embodiment, the auxiliary information can be dynamically configured by the wireless device 22 based on the CSI reporting payload, and explicitly signaled to the network node 16 in the UCI. With the auxiliary information, along with the bitwidth, the network node 16 can determine the proper sequence of ^^_AE bits to decode all the transmission layers. o In at least one embodiment, the above auxiliary information per transmission layer is reported to the network node 16 in CSI Part 1, 43 since they can be used by the network node 16 to determine the bitwidth of ^^_AE. o In at least one embodiment, the wireless device 22 can explicitly signal the bitwidth of ^^_AE in the Part 1 of CSI report, while moving the signaling of the auxiliary information per transmission layer to Part 2 of CSI report. o In at least one embodiment, the wireless device 22 can signal both the size of ^^_AE and the auxiliary information per transmission layer in Part 1 of the CSI report. Furthermore, since the bit sequence ^^_AE can have large payload, the bit sequence ^^_AE can be divided into multiple segments, which are transmitted in CSI Part 2. It allows dropping of some part(s) of ^^_AE when the allocated UCI resource, e.g., PUSCH allocation, for carrying such CSI report is not sufficient, as in the legacy CSI reporting framework. Accordingly, for AI-based implicit CSI feedback, ^^_AE can be segmented into multiple non-overlapping parts, where each segment corresponds to a transmission layer. • In at least one embodiment, where AI-based CSI report follows implicit-CSI feedback, each segment of ^^_AE can be further divided into sub-segments to carry transmission layer-specific auxiliary information in CSI Part 2, for example, the number of active latent-space coefficient and/or the number of quantization bits per latent-space coefficient along with the output of the AI model at the wireless device 22 per transmission layer. Embodiments related to mapping order of CSI report to UCI bit sequence In the CSI report, the parameters like CQI, RI, CQI, the model ID, the number of SD and FD basis (determining size of ^^_model), the number of active latent-space coefficient and/or the number of quantization bits per latent-space coefficient (to process the transmission layer information) provide auxiliary information for the network node 16 to estimate the payload of the received CSI report and decode the transmission layer information. The following embodiments describe the mapping strategy for bits corresponding to these parameters, denoted by ^^_AUX, to a CSI report for AI-based implicit CSI feedback. 44 ^ In at least one embodiment, the bits related to the auxiliary information common across all the transmission layers, such as CQI, RI, CQI, model ID, the number of SD and FD basis, and the layer-specific auxiliary information, such as the number of active latent-space coefficient and/or the number of quantization bits per latent-space coefficient, are included in ^^_AUX and transmitted in CSI Part 1. Subsequently, the bits in CSI Part 1 across multiple CSI reports are mapped on the UCI bit sequence. ^ In at least one embodiment, the bits related to the auxiliary information common across all the transmission layers, such as CQI, RI, CQI, model ID and the number of SD and FD basis are included in ^^_AUX and transmitted in CSI Part 1, while the bits related to the layer-specific auxiliary information, such as the number of active latent-space coefficient and/or the number of quantization bits per latent-space coefficient, are transmitted in CSI Part 2. Subsequently, the bits in CSI Part 1 across multiple CSI reports are mapped on the UCI bit sequence. As discussed herein, the bits for ^^_model and ^^_AE are allocated to CSI Part 2 of a CSI report, followed by mapping of the multiple CSI reports onto UCI bit sequence. Next, in the following embodiments, the mapping of the ^ ^^_^^ଡ଼, ^^_୫୭^^୪, ^^_^^^ bits onto a CSI report corresponding to different AI-based CSI feedback models. Example embodiment for a layer-common CSI reporting: ^ In at least one embodiment, where a layer-common CSI reporting is adopted, the parameters corresponding to ^^_AUX, ^^_model and ^^_AE can be defined as: - ^^_AUX: bits corresponding to the legacy quantities like CRI, RI and CQI, the AI model ID, the number of selected SD and FD basis, and the total number of bits in ^^_AE, if reported - ^^_model: bits corresponding to the selected SD and FD basis. - ^^_AE: bits corresponding to the compressed transmission layer information generated at the output of the AI model at the wireless device 22. The bits are segmented with each segment associated with a transmission layer, such that equal number of bits are allocated to each transmission layer. 45 Accordingly, let ^^_^ா^ be the bits generated by compressing and quantizing the ^^^th transmission layer at the output of the AI model in the wireless device 22, then ^^_^ாభ ൌ ^^_^ாమ ൌ ⋯ ൌ ^^_^ாೡ ൌ ^^_AE/ ^^, where ^^ is the number of transmission layers based on the reported RI. With the above description of ^^_AUX, ^^_model and ^^_AE, ^^_AUX are transmitted in CSI Part 1 and ^ ^^_model, ^^_AE^ bits are transmitted in CSI Part 2. Further, ^ ^^_model, ^^_AE^ are segmented into two different groups, similar to the legacy CSI report framework, where they are further segmented within each group as shown in TABLE 4. The segmentation of ^ ^^_AUX, ^^_model, ^^_AE^ as shown in TABLE 4 can be specified by the 3GPP. Accordingly, the network node 16 can be aware of the segmentation of ^ ^^_AUX, ^^_model, ^^_AE^, and decode them sequentially, starting from CSI Part 1, which always have a fixed payload size and carries information to calculate the payload size of CSI Part 2. Note that network node 16 can deduce or determine the number of bits in ^^_AE corresponding to each transmission layer through the RI and the total number of bits in ^^_AE reported in CSI Part 1. o In at least one embodiment, ^^_AUX bits can include the number of active latent-space coefficient at the output of the AI model at the wireless device 22 and the number of quantization bits for each latent-space coefficient, instead of the total number of bits in ^^_AE. Accordingly, network node 16 can deduce (e.g., determine) the number of bits in ^^_AE through the RI, the number of active latent-space coefficient and the number of quantization bits for each latent-space coefficient reported in CSI Part 1. TABLE 4 - Mapping order of ^^_AUX, ^^_model and ^^_AE to different parts/sub-parts of one CSI report for layer-common processing: CSI report CSI fields number Bits in ^^_AUX that correspond to CRI, if reported CSI report Bits in ^^_AUX that correspond to RI, if reported #n, Bits in ^^_AUX that correspond to Wideband CQI, if reported CSI Part 1 Bits in ^^_AUX that correspond to Subband CQI, if reported Bits in ^^_AUX that correspond to number of selected SD basis in 46 ^^_model, if reported Bits in ^^_AUX that correspond to number of selected FD basis in ^^_model, if reported Bits in ^^_AUX that correspond to the selected AI model ID, if reported Bits in ^^_AUX that correspond to the total number of bits in ^^_AE, if reported CSI report Bits in ^^_model that correspond to the selected SD basis, if #n reported CSI Part 2, Bits in ^^_model that correspond to the selected FD basis, if reported Group 0 CSI report ^^_^ாభ bits that correspond to the 1^st transmission layer, if reported #n ⋮ CSI Part 2, ^^_^ாೡ bits that correspond to the ^^^th transmission layer, if reported Group 1 In at least one embodiment, the CSI report can further include time-domain (TD) basis if the eigenvectors per layer are reported over a time interval and pre- processed in doppler domain. Example embodiment for a layer-specific CSI reporting: ^ In at least one embodiment, where a layer-specific CSI reporting is adopted, the parameters corresponding to ^^_AUX, ^^_model and ^^_AE can be defined as: - ^^_AUX: bits corresponding to the legacy quantities like CRI, RI and CQI, the AI model ID, the number of selected SD and FD basis, and the number of bits in ^^_AE. - ^^_model: bits corresponding to the selected SD basis and FD basis. - ^^_AE: bits corresponding to the compressed transmission layer information generated at the output of the AI model at the wireless device 22. The bits are segmented so that each segment is associated with a transmission 47 layer, where each segment consists of ^^_^ா^ bits, which can be a function of the transmission layer. Accordingly, ^{∑ ௩} ^_ୀ^ ^^_^ா^ ൌ ^^_AE , where ^^ is the number of transmission layers based on the reported RI. The total number of bits corresponding to a layer, ^^_^ா^, can further be segmented into three parts, also known as sub-groups (per layer): • ^^_^ா^,^^ bits corresponding to the number of active latent-space coefficient at the output of the encoder for the ^^^th transmission layer. • ^^_^ா^,^^ bits corresponding to the number of quantization bits used for each latent-space coefficient for the ^^^th transmission layer. • ^^ത^_^ா^ bits corresponding to the quantized latent-space coefficients at the output of the AI model at the wireless device 22 for the ^^^th transmission layer. With the above description of ^^_AUX, ^^_model and ^^_AE, ^^_AUX are transmitted in CSI Part 1 and ^ ^^_model, ^^_AE^ are transmitted in CSI Part 2. Further, ^ ^^_model, ^^_AE^ are segmented into two different groups, where they are further segmented within each group as shown in TABLE 5. Additionally, a segment for ^^_^ா^ can be divided into _{three sub-segments, corresponding to ^ ^^^ா^,^^ , ^^^ா^,^^ , ^} ^ത _{^^ா^ ^. The segmentation of} ^ ^^_AUX, ^^_model, ^^_AE^ as shown in TABLE 5 can be specified by the 3GPP. Accordingly, the network node 16 can be aware of the segmentation of ^ ^^_AUX, ^^_model, ^^_AE^, and decode them sequentially, starting from CSI Part 1, which always have a fixed payload size and carries information to calculate the payload size of CSI Part 2. TABLE 5 - Mapping order of ^^_AUX, ^^_model and ^^_AE to different parts/sub-parts of one CSI report for layer-specific processing: CSI report CSI fields number Bits in ^^_AUX that correspond to CRI, if reported 48 Bits in ^^_AUX that correspond to RI, if reported Bits in ^^_AUX that correspond to Wideband CQI, if reported Bits in ^^_AUX that correspond to Subband CQI, if reported CSI Bits in ^^_AUX that correspond to number of selected SD basis in report #n, ^^_model, if reported CSI Part Bits in ^^_AUX that correspond to number of selected FD basis in 1 ^^_model, if reported Bits in ^^_AUX that correspond to the selected AI model ID, if reported Bits in ^^_AUX that correspond to the total number of bits in ^^_AE, if reported CSI Bits in ^^_model that correspond to the selected SD basis, if report #n reported CSI Part 2, Bits in ^^_model that correspond to the selected FD basis, if reported Group 0 ^^_^ாభ,^భ bits that correspond to the number of active Group latent-space coefficient for the 1^st transmission 1.1.1 layer, if reported CSI report #n ^^_^ாభ,^భ ^^its that correspond to the number Group quantization bits for the 1^{^^} transmission layer, if CSI Part 1.1.2 reported 2, Group 1 _^ ^ത _{^^ா bits that correspond to the 1} ^st _transmission Group _భ layer, if reported 1.1.3 ⋮ ^^_^ாೡ,^ೡ bits that correspond to the number of active Group latent-space coefficient for the ^^ ^th transmission 1. ^^.1 layer, if reported ^^_^ாೡ,^ೡ bits that correspond to the number Group quantization bits for the ^^^th transmission layer, if 1. ^^.2 reported ^^ത _{^^ bits that correspond to the ^^} ^th _transmission Group _ாೡ layer, if reported 1. ^^.3 ^ In at least one embodiment, where the number of latent-space coefficients are not explicitly signaled, then ^^_^ா^, is only segmented into two parts, also known as sub-groups (per layer): ^ ^^_^ா^,^^ bits corresponding to a description of the total n_{umber of bits in ^} ^ത _^^ா ^th ^_{, for the ^^ transmission layer.} Group 1.i.1. ^_^ ^ത _{^^ா^ bits corresponding to the quantized latent-space} coefficients at the output of the AI model at the wireless device 22 for the ^^^th transmission layer. Group 1.i.2 ^ In at least one embodiment, where a layer-specific CSI reporting is adopted, the parameters corresponding to ^^_AUX, ^^_model and ^^_AE can be defined as: - ^^_AUX: bits corresponding to the legacy quantities like CRI, RI and CQI, the AI model ID, the number of selected SD and FD basis, the number of bits reported in ^^_AE, the number of active latent-space coefficient at the encoder per transmission layer, and the number of quantization bits used per latent-space coefficient. - ^^_model: bits corresponding to the selected SD basis and FD basis. - ^^_AE: bits corresponding to the compressed transmission layer information generated at the output of the AI model at the wireless device 22. The bits are segmented so that each segment is associated with a transmission layer, where each segment consists of ^^_^ா^ bits, which can be a function of the transmission layer. Accordingly, ∑ ^௩ ^_ୀ^ ^^_^ா^ ൌ ^^_AE , where ^^ is the number of transmission layers based on the reported RI. ^^_^ா^ can further be segmented into two parts: • ^^_^ா^,^^ bits corresponding to the step size ^^_^ to sample the active latent-space coefficient at the output of the AI model at the wireless device 22, such that the output from every ^^_^ ^{௧^} coefficient is signaled to the network node 16 for transmission layer. • ^^ത^_^ா^ bits corresponding to the quantized latent-space coefficients at the output of the AI model at the wireless device 22 for the i^th transmission layer. With the above description of ^^_^^^, ^^_^^ௗ^^ and ^^_^ா, bits are segmented similar to the method described in the above embodiment, as shown in TABLE 6. However, different from the above embodiment, a layer-common number of active latent-space coefficient and number of quantization bits for each latent-space coefficient is reported in CSI Part 1, where the layer specific CSI processing is enforced by including the output from only the specific latent-space coefficient taken at regular intervals, with the interval being a function of transmission layer and explicitly signaled to the network node 16 through ^^_^ா^,^^. Accordingly, for each segment (or sub-segment) corresponding to a transmission layer, ^^_^ா^,^^. bits are firstly decoded by the network node 16 to process ^^ത^_^ா^ with the information regarding the number of active latent-space coefficient and the number of quantization bits for each latent- space coefficient reported in CSI Part 1. TABLE 6 - Mapping order of ^^_^^^, ^^_^^ௗ^^ and ^^_^ா to different parts/sub-parts of one CSI report for PMI-based reporting: 51 CSI report CSI fields number Bits in ^^_AUX that correspond to CRI, if reported Bits in ^^_AUX that correspond to RI, if reported Bits in ^^_AUX that correspond to Wideband CQI, if reported Bits in ^^_AUX that correspond to Subband CQI, if reported Bits in ^^_AUX that correspond to number of selected SD basis in ^^_model, if reported CSI Bits in ^^_AUX that correspond to number of selected FD basis in report #n, ^^_model, if reported CSI Part Bits in ^^_AU that correspond to the selected AI model ID, if 1 _X reported Bits in ^^_AUX that correspond to the total number of bits in ^^_AE, if reported Bits in ^^_AUX that correspond to the number of active latent- space coefficients, if reported Bits in ^^_AUX that correspond to the number of quantization bits per active latent-space coefficient, if reported CSI Bits in ^^_model that correspond to the selected SD basis, if report #n reported CSI Part 2, Bits in ^^_model that correspond to the selected FD basis, if reported Group 0 ^^ bits that correspond to the step size to CSI _^ாభ,^భ Group sample the latent-space coefficient for the 1^{^^} report #n 1.1.1 transmission layer, if reported CSI Part ^^ത _{^^ bits that correspond to the 1} ^st _transmission 2, Group _ாభ layer, if reported 1.1.2 Group 1 ⋮ ^^_^ாೡ,^ೡ bits that correspond to the step size to Group sample the latent-space coefficient for the ^^ ^th 1. ^^.1 transmission layer, if reported ^^{ത^} _{bits that correspond to the ^^} ^th _t up _{^ ransmission} Gro _ாೡ _{1. ^^.2} layer, if reported ^ In at least one embodiment, when the AI model does not pre-process the eigenvectors per layer, i.e., the AI model compress and quantizes the raw eigenvectors per layer, the CSI report generated by the wireless device 22, does not include any bits corresponding to the SD and FD basis. ^ In at least one embodiment, inclusion of subset of parameters or addition of other parameters in each of ^ ^^_^^ଡ଼, ^^_୫୭^^୪, ^^_^^^ bits and their corresponding segmentation is not precluded. Embodiment for explicit CSI reporting: ^ In at least one embodiment, where the explicit CSI estimated by the wireless device 22, illustrated in FIG.10, can be directly pre-processed compressed and quantized by the AI model for such CSI reporting, the parameters corresponding to ^^_AUX, ^^_model and ^^_AE can be defined as: - ^^_AUX: bits corresponding to the legacy quantities like CRI, RI and CQI, the AI model ID, the number of selected SD and FD basis, and the total number of bits in ^^_AE . - ^^_model: bits corresponding to the selected SD and FD basis. - ^^_AE: bits corresponding to the compressed pre-processed CSI generated at the output of the AI model at the wireless device 22, illustrated in FIG. 53 10. The bits are divided into segments such that each segment contains bits generated by each latent-space coefficient. Accordingly, the bits in each segment can be represented by ^^_^ாೕ , ^^ ൌ 1, … , ^^, such that ∑ ^ே ^_ୀ^ ^^_^ாೕ ൌ ^^_AE , where ^^ is the number of latent- coefficients. With the above description of ^^_AUX, ^^_model and ^^_AE, the ^^_AUX are transmitted in CSI Part 1 and ^ ^^_model, ^^_AE^ are transmitted in CSI Part 2. Further, ^ ^^_model, ^^_AE^ are segmented into two different groups, similar to the legacy CSI report framework, where they are further segmented within each group as shown in TABLE 7. The segmentation of ^ ^^_AUX, ^^_model, ^^_AE^ as shown in TABLE 7 can be specified by the 3GPP. Accordingly, the network node 16 illustrated in FIG.10 can be aware of the segmentation of ^ ^^_AUX, ^^_model, ^^_AE^, and decode them sequentially, starting from CSI Part 1, which always have a fixed payload size and carries information to calculate the payload size of CSI Part 2. o In at least one embodiment, ^^_AUX bits can include the number of active latent-space coefficient at the output of the AI model at the wireless device 22 illustrated in FIG.10 and the number of quantization bits for each latent-space coefficient, instead of the total number of bits in ^^_AE. Accordingly, network node 16 illustrated in FIG.10 can determine the number of bits in ^^_AE through the number of active latent-space coefficient and the number of quantization bits for each latent-space coefficient reported in CSI Part 1. TABLE 7 - Mapping order of ^^_AUX, ^^_model and ^^_AE to different parts/sub-parts of one CSI report for explicit CSI processing: CSI report CSI fields number Bits in ^^_AUX that correspond to CRI, if reported 54 Bits in ^^_AUX that correspond to RI, if reported Bits in ^^_AUX that correspond to Wideband CQI, if reported Bits in ^^_AUX that correspond to Subband CQI, if reported Bits in ^^_AUX that correspond to number of selected SD basis in CSI report ^^_model, if reported #n, Bits in ^^_AUX that correspond to number of selected FD basis in CSI Part 1 ^^_model, if reported Bits in ^^_AUX that correspond to the selected AI model ID, if reported Bits in ^^_AUX that correspond to the total number of bits in ^^_AE, if reported CSI report Bits in ^^_model that correspond to the selected SD basis, if #n reported CSI Part 2, Bits in ^^_model that correspond to the selected FD basis, if reported Group 0 CSI report ^^_^ாభ bits that correspond to the 1^st ^^_AE segment, if reported #n ⋮ CSI Part 2, ^^_^ாಿ bits that correspond to the ^^^th ^^_AE segment, if reported Group 1 ^ In at least one embodiment, the quantities like RI and/or CQI are computed at the network node 16 illustrated in FIG.10 after decoding the explicit CSI, which can then be signaled to the wireless device 22 illustrated in FIG.10 in the DCI. In such case, the RI and/or the CQI can be excluded from the CSI report in the UCI. Embodiments related to mapping of CSI reports to UCI bit sequence Once the ^ ^^_^^ଡ଼, ^^_୫୭^^୪, ^^_^^^ bits are mapped onto a CSI report, one or multiple CSI reports are mapped on to the UCI bit sequence, which are the signaled to the network node 16, as specified, e.g., in 3GPP TS 38.214. The UCI bit sequences may be transmitted on PUCCH or PUSCH. As per NR 3GPP Rel-17, two bit sequences can be created, i.e., ^^^{^^^}, ^^^{^^^}, ^^^{^^^ ^^^ ^^^} ^_{^ ଶ} , ^^_ଷ , … , ^^_^^భ^ି^ for CSI Part 1, and ^^^{^ଶ^} ^ , ^^^{^ଶ^} ^ , ^^^{^ଶ^} ଶ , ^{^ଶ^} and ^^ are the number of bits Part 1 to the UCI bit sequence ^^^{^^^} ^ , ^^^{^^^} ^ , ^^^{^^^} ଶ , ^^^{^^^} ଷ , … , ^^^{^^^} ^_^భ^ି^ can be performed in the same way as defined in, for 3GPP NR Rel-17 TS 38.212 V17.2.0, thus is omitted here. In the following embodiment, the mapping order of multiple CSI reports to the corresponding bit sequences for CSI Part 2 is discussed, which is enhanced for AI-based CSI reporting as described herein. ^ In at least one embodiment, the mapping order of Part 2 CSI for the UCI bit sequence ^^^{^ଶ^} ^ , ^^^{^ଶ^} ^ , ^^^{^ଶ^} ଶ , ^^^{^ଶ^} ଷ , … , ^^^{^ଶ^} ^_^మ^ି^ can be done either by prioritizing the report number, the group the subgroup number within a group. When the group number is prioritized, as in the following tables, which further requires prioritizing the sub-group number per report, if configured, it ensures that all the transmission layers for a report is transmitted first. If there is any remaining resource, it can be used to transmit additional reports. When the report number is prioritized, as in TABLES 8 and 9, it ensures that all the lower transmission layers for a report is transmitted first, which further requires prioritizing the sub-group number for report, if configured. TABLE 8 - Mapping order of CSI reports to UCI bit sequence ^^^{^ଶ^}, ^^^{^ଶ^}, ^^^{^ଶ^}, ^{^ଶ^ ^ଶ^} ^_{^ ଶ} ^^_ଷ , … , ^^_^^మ^ି^ , where group has higher priority: CSI report number C^{SI report #1, CSI Part 2, Group 0} ^_{^^ଶ^ ^ଶ^ ^ଶ^ ^ଶ^} _{^ , ^^^ , ^^ଶ , ^^ଷ , … , ^^} CSI report #1, CSI Part 2, Group 1.1 56 ⋮ CSI report #1, CSI Part 2, Group 1. ^^ ⋮ CSI report #n, CSI Part 2, Group 0 CSI report #n, CSI Part 2, Group 1.1 ⋮ CSI report #n, CSI Part 2, Group 1. ^^ TABLE 9 - Mapping order of CSI reports to UCI bit sequence ^^^{^ଶ^}, ^^^{^ଶ^}, ^^^{^ଶ^}, ^^^{^ଶ^}, ^{^ଶ^} ^_{^ ଶ ଷ} … , ^^_^^మ^ି^ , where report number has higher priority report number CSI report #1, CSI Part 2, Group 0 ⋮ CSI report #n, CSI Part 2, Group 0 CSI report #1, CSI Part 2, Group 1.1 ⋮ ^^^{^ଶ^ ^ଶ^ ^ଶ^ ^ଶ^} ^ , ^^_^ , ^^_ଶ , ^^_ଷ , … , ^^ CSI report #n, CSI Part 2, Group 1.1 ⋮ CSI report #1, CSI Part 2, Group 0. ^^ ⋮ CSI report #n, CSI Part 2, Group 1. ^^ Some Examples in accordance with the present disclosure: 57 Example 1: A method, wherein the network node 16 can signal the use of AI-based CSI report through RRC signaling. Example 2: Example 1, wherein the wireless device 22 reports an AI-based CSI report on UCI, where the generation of the CSI report comprises one or multiple of the following: extracting the eigenvectors per transmission layer from the estimated channel; reducing the dimension of eigenvectors per transmission layer by applying the pre- processing in space, frequency and/or time domain, where the information of the corresponding SD, FD and/or TD basis form part of the CSI report on UCI; compressing and quantizing the pre-processed (or raw) eigenvectors per transmission layer with the deployed AI model, where the quantized bits form part of the CSI report on UCI; segmenting the AI-based CSI report into Part 1 CSI and Part 2 CSI, which are transmitted on different parts of the UCI; each of Part 1 and Part 2 CSI can be further segmented into multiple sub-segments. Example 3: Any one of Examples 1 and 2, wherein the network node 16 RRC configures the parameters to determine the content and size of the AI-based CSI report on the UCI, which can include one or multiple of the following: restriction on the rank that can be reported back by the wireless device 22; configuration for the SD, FD and/or TD basis used to pre-process the eigenvectors per transmission layer; configuration for the quantization bits that can be used by the wireless device 22 to quantize the pre-preprocessed (or raw) eigenvectors per transmission layer; configuration for the number of latent-space coefficients at the output of the AI model at the wireless device 22 per transmission layer; configuration to either use a transmission layer-common or a transmission layer- specific processing to generate the CSI report at the wireless device 22. 58 Example 4: Example 3, wherein one or multiple of the parameters are implicitly associated (and configured) with the model deployed at the network node 16 and the wireless device 22. The model ID, representing the model, can be either be explicitly signaled by the network node 16 or wireless device 22. Example 5: Example 3, wherein one or multiple of the parameters are dynamically configured by the network node 16 on the DCI and/or MAC-CE. Example 6: Example 3, wherein one or multiple of the parameters are explicitly configured by the wireless device 22 and reported as a part of CSI report on UCI. Example 7: Any one of Examples 2-6, wherein one or multiple of the followings are included in Part 1 CSI, if reported: any of the legacy CSI report quantities, e.g., CRI, RI, CQI, etc.; any information of the AI-model that is used to generate the CSI report, e.g., the AI model ID as described in, for example, the Chair’s Notes for RAN1110bis-e, version 17; any of the auxiliary information for the AI-based quantization bits along with the corresponding pre-processing basis to decode the eigenvector per transmission layer transmitted in Part 2 CSI. Example 8: Any of Examples 2-6, wherein one or multiple of the followings are included in Part 2 CSI, if reported: the index for the SD, FD and/or TD pre-processing basis for the eigenvectors per transmission layer; transmission layer-specific information required by network node 16 to decode the quantized bits per transmission layer; quantized bits per transmission layer from the output of the AI model at the wireless device 22. As will be appreciated by one of skill in the art, the concepts described herein may be embodied as a method, data processing system, computer program product and/or computer storage media storing an executable computer program. 59 Accordingly, the concepts described herein may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects all generally referred to herein as a “circuit” or “module.” Any process, step, action and/or functionality described herein may be performed by, and/or associated to, a corresponding module, which may be implemented in software and/or firmware and/or hardware. Furthermore, the disclosure may take the form of a computer program product on a tangible computer usable storage medium having computer program code embodied in the medium that can be executed by a computer. Any suitable tangible computer readable medium may be utilized including hard disks, CD-ROMs, electronic storage devices, optical storage devices, or magnetic storage devices. Some embodiments are described herein with reference to flowchart illustrations and/or block diagrams of methods, systems and computer program products. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer (to thereby create a special purpose computer), special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer program instructions may also be stored in a computer readable memory or storage medium that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instruction means which implement the function/act specified in the flowchart and/or block diagram block or blocks. The computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the 60 computer or other programmable apparatus provide steps for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. It is to be understood that the functions/acts noted in the blocks may occur out of the order noted in the operational illustrations. For example, two blocks shown in succession may in fact be executed substantially concurrently or the blocks may sometimes be executed in the reverse order, depending upon the functionality/acts involved. Although some of the diagrams include arrows on communication paths to show a primary direction of communication, it is to be understood that communication may occur in the opposite direction to the depicted arrows. Computer program code for carrying out operations of the concepts described herein may be written in an object oriented programming language such as Python, Java® or C++. However, the computer program code for carrying out operations of the disclosure may also be written in conventional procedural programming languages, such as the "C" programming language. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer. In the latter scenario, the remote computer may be connected to the user's computer through a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider). Many different embodiments have been disclosed herein, in connection with the above description and the drawings. It will be understood that it would be unduly repetitious and obfuscating to literally describe and illustrate every combination and subcombination of these embodiments. Accordingly, all embodiments can be combined in any way and/or combination, and the present specification, including the drawings, shall be construed to constitute a complete written description of all combinations and subcombinations of the embodiments described herein, and of the manner and process of making and using them, and shall support claims to any such combination or subcombination. Abbreviations that may be used in the preceding description include: Abbreviation Explanation 61 3GPP 3rd Generation Partnership Project AE Auto Encoder AI Artificial Intelligence CQI Channel Quality Indicator CSI Channel State Information CSI-RS Channel State Information Reference Signal DCI Downlink Control Information FD Frequency Domain gNB A radio base station in NR LSB Least significant bit ML Machine Learning MSB Most significant bit MU-MIMO Multi User-Multiple Input, Multiple Output NR New Radio PMI Precoder Matrix Indicator PUCCH Physical Uplink Control Channel PUSCH Physical Uplink Shared Channel RI Rank Indicator RRC Radio Resource Control SD Spatial Domain SRS Sounding Reference Signal TD Time Domain UCI Uplink Control Information UE User Equipment 62 It will be appreciated by persons skilled in the art that the embodiments described herein are not limited to what has been particularly shown and described herein above. In addition, unless mention was made above to the contrary, it should be noted that all of the accompanying drawings are not to scale. A variety of modifications and variations are possible in light of the above teachings. Some enumerated Embodiments as presented herein: Embodiment A1. A network node configured to communicate with a wireless device the network node configured to, and/or comprising a radio interface and/or comprising processing circuitry configured to: transmit an indication to the wireless device to cause a generation of an artificial intelligence based, AI-based, channel state information, CSI, report; receive the AI-based CSI report; and perform at least one action based on the received CSI report. Embodiment A2. The network node of Embodiment A1, wherein the indication comprises at least one of: a restriction on a rank that can be reported by the wireless device; at least one configuration for at least one of a spatial domain, SD, a frequency domain, FD, and a time domain, TD, the configuration being used to pre-process eigenvectors; at least one configuration for the quantization bits that can be used by the wireless device to quantize the eigenvectors; at least one configuration for a number of latent-space coefficients at an output of the AI model; and an indication for the wireless device to use at least one of transmission layer- common processing and transmission layer-specific processing to generate the CSI report. 63 Embodiment A3. The network node of Embodiment A1, wherein the processing circuitry is further configured to indicate at least one parameter to be used in the generation of the CSI report, the at least one parameter being indicated on at least one of a downlink control information, DCI, and a medium access control-control element, MAC-CE. Embodiment B1. A method implemented in a network node, the method comprising: transmitting an indication to the wireless device to cause a generation of an artificial intelligence based, AI-based, channel state information, CSI, report; receiving the AI-based CSI report; and performing at least one action based on the received CSI report. Embodiment B2. The method of Embodiment B1, wherein the indication comprises at least one of: a restriction on a rank that can be reported by the wireless device; at least one configuration for at least one of a spatial domain, SD, a frequency domain, FD, and a time domain, TD, the configuration being used to pre-process eigenvectors; at least one configuration for the quantization bits that can be used by the wireless device to quantize the eigenvectors; at least one configuration for a number of latent-space coefficients at an output of the AI model; and an indication for the wireless device to use at least one of transmission layer- common processing and transmission layer-specific processing to generate the CSI report. Embodiment B3. The method of Embodiment B1, further comprising indicating at least one parameter to be used in the generation of the CSI report, the at least one 64 parameter being indicated on at least one of a downlink control information, DCI, and a medium access control-control element, MAC-CE. Embodiment C1. A wireless device configured to communicate with a network node, the wireless device configured to, and/or comprising a radio interface and/or processing circuitry configured to: receive an indication from the network node to generate an artificial intelligence based, AI-based, channel state information, CSI, report; generate the CSI report; and transmit the CSI report to the network node. Embodiment C2. The wireless device of Embodiment C1, wherein the indication comprises at least one of: a restriction on a rank that can be reported by the wireless device; at least one configuration for at least one of a spatial domain, SD, a frequency domain, FD, and a time domain, TD, the configuration being used to pre-process eigenvectors; at least one configuration for the quantization bits that can be used by the wireless device to quantize the eigenvectors; at least one configuration for a number of latent-space coefficients at an output of the AI model; and an indication for the wireless device to use at least one of transmission layer- common processing and transmission layer-specific processing to generate the CSI report. Embodiment C3. The wireless device of Embodiment C1, wherein the processing circuitry is further configured to receive at least one parameter to be used in the generation of the CSI report, the at least one parameter being received on at least one of a downlink control information, DCI, and a medium access control-control element, MAC-CE. 65 Embodiment D1. A method implemented in a wireless device, the method comprising: receiving an indication from the network node to generate an artificial intelligence based, AI-based, channel state information, CSI, report; generating the CSI report; and transmitting the CSI report to the network node. Embodiment D2. The method of Embodiment D1, wherein the indication comprises at least one of: a restriction on a rank that can be reported by the wireless device; at least one configuration for at least one of a spatial domain, SD, a frequency domain, FD, and a time domain, TD, the configuration being used to pre-process eigenvectors; at least one configuration for the quantization bits that can be used by the wireless device to quantize the eigenvectors; at least one configuration for a number of latent-space coefficients at an output of the AI model; and an indication for the wireless device to use at least one of transmission layer- common processing and transmission layer-specific processing to generate the CSI report. Embodiment D3. The method of Embodiment D1, further comprising receiving at least one parameter to be used in the generation of the CSI report, the at least one parameter being received on at least one of a downlink control information, DCI, and a medium access control-control element, MAC-CE. 66

Claims

CLAIMS 1. A method for channel state information, CSI, reporting, performed by a wireless device, the wireless device having one or more encoders of one or more autoencoders available, the method comprising: generating (S143) a CSI report using an encoder; and segmenting (S145) the output of the encoder into a first CSI report and a second CSI report, wherein the first CSI report and the second CSI report are transmitted on different parts of an Uplink Control Information, UCI.

2. The method according to claim 1, further comprising receiving (S141), through Radio Resource Control, RRC, signaling, an indication to generate a CSI report using an encoder.

3. The method according to any one of claims 1-2, wherein generating (S142) a CSI report using an encoder comprises preprocessing an estimated channel.

4. The method according to claim 3, wherein preprocessing an estimated channel comprises: extracting eigenvectors per transmission layer from an estimated channel; and reducing the dimension of the extracted eigenvectors per transmission layer by applying a pre-processing in a space, frequency, and/or time domain.

5. The method according to claim 4, wherein generating (S14) a CSI report using an encoder comprises compressing and quantizing the pre-processed eigenvectors per transmission layer with an autoencoder.

6. The method according to any one of claims 1-5, wherein the information of the space, frequency, and/or time domain forms part of the CSI report on UCI.

7. The method according to any one of claims 1-6, wherein the quantized bits form part of the CSI report on UCI.

8. The method according to any one of claims 1-7, further comprising receiving (S141) through RRC signaling, parameters associated to the content and/or size of the encoded CSI, wherein the parameters include one or more of; 67 a restriction on the rank of the encoded CSI; an indication of the space, frequency, and/or time domain basis used to pre-process the eigenvectors per transmission layer; an indication of the quantization bits which can be used by the wireless device to quantize the pre-processed eigenvectors per transmission layer; an indication of the number of latent space coefficients of the encoded CSI per transmission layer; an indication of whether to use a transmission layer-common processing or a transmission layer-specific processing to generate the CSI report at the wireless device.

9. The method according to claim 8, wherein one or more of the parameters is explicitly associated to an encoder deployed at the wireless device.

10. The method according to claim 9, wherein a model identification uniquely associated to one of the encoders available to the wireless device is signaled to the wireless device.

11. The method according to claim 8, wherein one or more of the parameters is dynamically configured on the Downlink Control Information, DCI, and/or the Medium Access Control, MAC, Control Element, CE.

12. The method according to claim 8, wherein one or more of the parameters is configured by the wireless device and reported as part of the CSI report on UCI.

13. The method according to any one of claims 1-12, wherein the first CSI report includes one or more of: a legacy CSI report quantity; information about the autoencoder used to generate the CSI report; auxiliary information for the AI-based quantization bits; a basis used for pre-processing.

14. The method according to any one of claims 1-13, wherein the second CSI report includes one or more of: 68 an index for the space, frequency, and/or time dimension pre-processing basis for the eigenvectors per transmission layer; transmission layer-specific information required by the network node to decode the quantized bits per transmission layer; quantized bits per transmission layer from the output of the autoencoder.

15. A method for channel state information, CSI, reporting, performed by a network node (16), the network node having one or more decoders of one or more autoencoders available, the method comprising: indicating (S135), through Radio Resource Control, RRC, signaling, that a wireless device is to transmit a CSI report compressed by means of an autoencoder; receiving (S137), from a wireless device (22), a first CSI report and a second CSI report on different parts of Uplink Control Information, UCI; using (S139) a decoder from the one or more decoders to decode the first CSI report and the second CSI report.

16. The method according to claim 15, further comprising transmitting, through RRC signaling, parameters associated to the content and/or size of the encoded CSI, wherein the parameters include one or more of; a restriction on the rank of the encoded CSI; an indication of the space, frequency, and/or time domain basis used to pre-process the eigenvectors per transmission layer; an indication of the quantization bits which can be used by the wireless device to quantize the pre-processed eigenvectors per transmission layer; an indication of the number of latent space coefficients of the encoded CSI per transmission layer; an indication of whether to use a transmission layer-common processing or a transmission layer-specific processing to generate the CSI report at the wireless device.

17. The method according to claim 16, wherein one or more of the parameters is explicitly associated to an autoencoder deployed at the network node. 69

18. The method according to claim 17, wherein a model identification uniquely associated to one of the encoders available to the wireless device is signaled to the wireless device.

19. The method according to claim 16, wherein one or more of the parameters is dynamically configured on the Downlink Control Information, DCI, and/or the Medium Access Control, MAC, Control Element, CE.

20. The method according to claim 16, wherein one or more of the parameters is configured by the wireless device and received by the network node as part of the CSI report on UCI.

21. The method according to any one of claims 15-20, wherein the first CSI report includes one or more of: a legacy CSI report quantity; information about the autoencoder used to generate the CSI report; auxiliary information for the AI-based quantization bits; a basis used for pre-processing.

22. The method according to any one of claims 15-21, wherein the second CSI report includes one or more of: an index for the space, frequency, and/or time dimension pre-processing basis for the eigenvectors per transmission layer; transmission layer-specific information required by the network node to decode the quantized bits per transmission layer; quantized bits per transmission layer from the output of the autoencoder.

23. A wireless device (22) configured to perform a method for channel state information, CSI, reporting, the wireless device having one or more encoders of one or more autoencoders available, the method comprising: generating (S143) a CSI report using an encoder; and 70 segmenting (S145) the output of the encoder into a first CSI report and a second CSI report, wherein the first CSI report and the second CSI report are transmitted on different parts of the Uplink Control Information, UCI.

24. The wireless device (22) according to claim 23, further configured to perform a method according to any one of claims 2-14.

25. A wireless device (22) configured to perform a method for channel state information, CSI, reporting, the wireless device comprising processing circuitry and a memory, the wireless device having one or more encoders of one or more autoencoders available, the method comprising: generating (S143) a CSI report using an encoder; and segmenting (S145) the output of the encoder into a first CSI report and a second CSI report, wherein the first CSI report and the second CSI report are transmitted on different parts of the Uplink Control Information, UCI.

26. The wireless device (22) according to claim 25, further configured to perform a method according to any one of claims 2-14.

27. A radio access node (16) in a communication network configured to perform a method for channel state information, CSI, reporting, the network node having one or more decoders of one or more autoencoders available, the method comprising: indicating (S135), through Radio Resource Control, RRC, signaling, that a wireless device is to transmit a CSI report compressed by means of an autoencoder; receiving (S137), from a wireless device (22), a first CSI report and a second CSI report on different parts of Uplink Control Information, UCI; using (S139) a decoder from the one or more decoders to decode the first CSI report and the second CSI report.

28. The radio access node (16) according to claim 27, further configured to perform a method according to any one of claims 15-22.

29. A radio access node (16) in a communication network configured to perform a method for channel state information, CSI, reporting, the network node 71 comprising processing circuitry and a memory, the network node having one or more decoders of one or more autoencoders available, the method comprising: indicating (S135), through Radio Resource Control, RRC, signaling, that a wireless device is to transmit a CSI report compressed by means of an autoencoder; receiving (S137), from a wireless device (22), a first CSI report and a second CSI report on different parts of Uplink Control Information, UCI; using (S139) a decoder from the one or more decoders to decode the first CSI report and the second CSI report.

30. The radio access node (16) according to claim 29, further configured to perform a method according to any one of claims 15-22.

31. A computer program comprising machine-readable instructions which, when executed by the processor of a wireless device, cause the wireless device to perform a method according to any one of claims 1-14.

32. A computer program product comprising a non-transient computer readable storage medium on which a computer program according to claim 31 is stored.

33. A computer program comprising machine-readable instructions which, when executed by the processor of a radio access node, cause the radio access node to perform a method according to any one of claims 15-22.

34. A computer program product comprising a non-transient computer readable storage medium on which a computer program according to claim 33 is stored. 72