Channel state information feedback compression in new radio transmissions

Abstract

The present disclosure includes a method for managing channel state information feedback compression in a user equipment. The method comprises the following steps: receiving at least one reference signal; estimating a channel condition based on the at least one reference signal; reducing a dimension of the estimated channel condition using an encoder of a first auto-encoder in a user equipment (UE) such that a low-dimension channel condition is generated; calculating an error measurement based on a difference between the UE channel state input and the UE channel state output using an encoder of a second auto-encoder and a decoder of the auto-encoder in the UE; and transmitting at least one of the channel state information bitstream or the at least one message, where the at least one of the channel state information bitstream or the at least one message is based on at least one of the low dimension channel condition or the value indicating the error measurement value.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of priority to U.S. Provisional Patent Application No. 63 / 373,985, filed on August 30, 2022, entitled “CHANNEL STATE INFORMATION FEEDBACK COMPRESSION IN NEW RADIO TRANSMISSIONS,” the entire contents of which are incorporated herein by reference. Technical Field

[0002] Apparatus and methods consistent with the present disclosure relate generally to communications, and more particularly, to methods, systems, and devices for reporting channel state information (CSI) in wireless networks. Background Art

[0003] CSI provides information about the state of the radio channel between the base station (gNB) and the user equipment (UE). This information is essential for various tasks such as beamforming, power control, and scheduling to optimize the performance of the wireless network. CSI may include one or more parameters such as channel quality, signal strength, interference level, and other relevant indicators to provide efficient communication between the base station and the UE.

[0004] In 3GPP New Radio (NR), there may be challenges in accurately describing channel conditions. For example, when the dimension of the channel condition is high, conveying the channel condition may require a high number of bits. Summary of the invention

[0005] In view of the above, embodiments of the present disclosure address the shortcomings of existing systems by providing an apparatus, system, and method for CSI reporting in a wireless network.

[0006] According to some embodiments of the present disclosure, a method for managing CSI feedback compression in a UE is provided. The method includes: receiving at least one reference signal; estimating a channel condition based on the at least one reference signal; using an encoder of a first autoencoder in the UE, reducing the dimension of the estimated channel condition so as to generate a low-dimensional channel condition; using an encoder of a second autoencoder in the UE and a decoder of the autoencoder, calculating an error measurement value based on the difference between a UE channel state input and a UE channel state output; and transmitting at least one of a CSI bitstream or at least one message, wherein at least one of the CSI bitstream or at least one message is based on at least one of the low-dimensional channel condition or a value indicating the error measurement value.

[0007] According to some embodiments of the present disclosure, a device for CSI feedback compression is provided. The device includes: a memory storing instructions; and a processor configured to execute the instructions stored in the memory to: receive at least one reference signal; estimate channel conditions based on at least one reference signal; use an encoder of a first autoencoder in a UE to reduce the dimension of the estimated channel conditions so as to generate a low-dimensional channel condition; use an encoder of a second autoencoder in the UE and a decoder of the autoencoder to calculate an error measurement value based on the difference between a UE channel state input and a UE channel state output; and transmit at least one of a CSI bit stream or at least one message, wherein at least one of the CSI bit stream or at least one message is based on at least one of the low-dimensional channel conditions or a value indicating the error measurement value.

[0008] According to some embodiments of the present disclosure, a non-transitory computer-readable medium is provided, which stores instructions that can be executed by one or more processors of an apparatus for managing CSI feedback compression to perform a method. The method includes: receiving at least one reference signal; estimating a channel condition based on at least one reference signal; using an encoder of a first autoencoder in a UE to reduce the dimension of the estimated channel condition so as to generate a low-dimensional channel condition; using an encoder of a second autoencoder in the UE and a decoder of the autoencoder to calculate an error measurement value based on the difference between a UE channel state input and a UE channel state output; and transmitting a CSI bitstream or at least one message, wherein at least one of the CSI bitstream or at least one message is based on at least one of the low-dimensional channel condition or a value indicating an error measurement value.

[0009] According to some embodiments of the present disclosure, a method for managing CSI feedback compression is provided. The method includes: transmitting at least one reference signal; receiving at least one of a CSI bit stream or at least one message in response to the transmitted at least one reference signal; determining at least one of a low-dimensional channel condition or a value indicating an error measurement value based on the CSI bit stream or at least one of the messages; calculating at least one of the updated weights or updated parameters of a decoder of an autoencoder in a base station based on the estimated channel condition or at least one of the values ​​indicating the error measurement value, wherein the estimated channel condition is based on the low-dimensional channel condition; and updating at least one of the weights or parameters of the decoder of the autoencoder in the base station based on at least one of the calculated updated weights or updated parameters.

[0010] According to some embodiments of the present disclosure, a device for managing CSI feedback is provided. The device includes: a memory storing instructions; and a processor configured to execute the instructions stored in the memory to: transmit at least one reference signal; receive at least one of a CSI bit stream or at least one message in response to the transmitted at least one reference signal; determine at least one of a low-dimensional channel condition or a value indicating an error measurement value based on the CSI bit stream or at least one message; calculate at least one of the updated weights or updated parameters of a decoder of an autoencoder in a base station based on at least one of the estimated channel or the value indicating the error measurement value, wherein the estimated channel condition is based on the low-dimensional channel condition; and update at least one of the weights or parameters of the decoder of the autoencoder in the base station based on at least one of the calculated updated weights or updated parameters.

[0011] According to some embodiments of the present disclosure, a non-transitory computer-readable medium is provided, which stores instructions that can be executed by one or more processors of an apparatus for managing CSI feedback compression to perform a method. The method includes: transmitting at least one reference signal; receiving at least one of a CSI bit stream or at least one message in response to the transmitted at least one reference signal; determining at least one of a low-dimensional channel condition or a value indicating an error measurement value based on the CSI bit stream or at least one message; calculating at least one of the updated weights or updated parameters of a decoder of an autoencoder in a base station based on at least one of the estimated channel conditions or the values ​​indicating the error measurement value, wherein the estimated channel conditions are based on the low-dimensional channel conditions; and updating at least one of the weights or parameters of a decoder of an autoencoder in a base station based on at least one of the calculated updated weights or updated parameters.

[0012] According to some embodiments of the present disclosure, a method for managing CSI feedback compression in a UE is provided. The method includes: receiving at least one reference signal; estimating a channel condition based on at least one reference signal; using an encoder of a first autoencoder in the UE to reduce the dimension of the estimated channel condition so as to generate a low-dimensional channel condition; using an encoder of a second autoencoder in the UE and a decoder of the autoencoder to calculate an error measurement value based on the difference between a UE channel state input and a UE channel state output; calculating at least one of the updated weights or updated parameters of the encoder of the first autoencoder in the UE based on a value indicating the error measurement value; updating at least one of the weights or parameters of the encoder of the first autoencoder in the UE based on the calculated updated weights or at least one of the updated parameters; generating at least one of a CSI bitstream or at least one of at least one of the low-dimensional channel condition or at least one of the updated weights or updated parameters of the encoder of the first autoencoder in the UE; and transmitting at least one of the CSI bitstream or at least one of the at least one of the at least one of the messages.

[0013] According to some embodiments of the present disclosure, a method for managing CSI feedback compression in a base station is provided. The method includes: transmitting at least one reference signal; receiving at least one of a CSI bitstream or at least one message in response to the transmitted at least one reference signal; determining at least one of an updated weight, updated parameter or structure for at least one of a first autoencoder in a UE, a second autoencoder in a UE or a decoder of an autoencoder in a UE; based on at least one of the updated weights or updated parameters of the encoder of the autoencoder in the UE, calculating at least one of the updated weights or updated parameters of the decoder of the autoencoder in the base station; and updating at least one of the weights, parameters or structures of the decoder of the autoencoder in the base station based on at least one of the calculated updated weights or updated parameters or structures.

[0014] Some embodiments of the present disclosure relate to receiving at least weights, parameters or structures of an encoder of at least a first autoencoder, an encoder of a second autoencoder and / or a decoder of an autoencoder in a UE from a base station; and updating at least weights, parameters or (one or more) structures of an encoder of at least a first autoencoder, an encoder of a second autoencoder and / or a decoder of an autoencoder in the UE using the weights, parameters or structures received from the base station.

[0015] Some embodiments of the present disclosure relate to calculating and / or updating at least one of the weights, parameters or (one or more) structures of at least one of the decoder of a first autoencoder, the decoder of a second autoencoder and / or the encoder of an autoencoder in a base station; and transmitting the weights, parameters or structures to a UE. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] [ Figure 1 ] Figure 1 is a diagram showing an example of CSI feedback for downlink (DL) and uplink (UL) transmission.

[0017] [ Figure 2 ] Figure 2 is a diagram showing an example of a basic architecture of an autoencoder.

[0018] [ Figure 3 ] Figure 3 is a diagram showing the basic architecture of an autoencoder and associated examples of training on a dataset to determine the weights / parameters of the encoder and decoder.

[0019] [ Figure 4 ] Figure 4 is a schematic diagram showing an example of a sandwich structure of an autoencoder, in which the structure of the encoder-decoder is symmetrical.

[0020] [ Figure 5 ] Figure 5 is a schematic diagram showing an example of a general architecture for CSI compression using an autoencoder.

[0021] [ Figure 6 ] Figure 6 is a schematic diagram showing an example of structures of an encoder of an autoencoder and a decoder of the autoencoder, updating weights / parameters for the encoder of the autoencoder and the decoder of the autoencoder, and a CSI bitstream generation and decoding scheme.

[0022] [ Figure 7 ] Figure 7 is a flow chart illustrating an exemplary method for managing CSI feedback compression, for example in a UE, according to an embodiment of the present disclosure.

[0023] [ Figure 8 ] Figure 8 is a flow chart illustrating another exemplary method for managing CSI feedback compression, for example in a base station, according to an embodiment of the present disclosure.

[0024] [ Fig. 9 ] Fig. 9 is a schematic diagram showing an example of structures of an encoder of an autoencoder and a decoder of the autoencoder, updating weights / parameters for the encoder of the autoencoder and the decoder of the autoencoder, and a CSI bitstream generation and decoding scheme.

[0025] [ Fig.10 ] Fig.10is a flow chart illustrating a further exemplary method for managing CSI feedback compression, for example in a UE, according to an embodiment of the present disclosure.

[0026] [ Fig.11 ] Fig.11 is a flow chart illustrating an additional exemplary method for managing CSI feedback compression in a gNB according to an embodiment of the present disclosure.

[0027] [ Fig.12 ] Fig.12 A block diagram of a device consistent with some embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0028] Reference will now be made in detail to the exemplary embodiments, examples of which are shown in the accompanying drawings. The following description refers to the accompanying drawings, in which the same reference numerals in different figures represent the same or similar elements unless otherwise specified. The embodiments set forth in the following description of the exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of systems, devices, and methods consistent with various aspects related to the present disclosure as described in the appended claims.

[0029] In 3GPP NR, the radio resource allocation, configuration, and transmission scheme on the Uu interface (the interface between the UE and the gNB) are determined by the base station. In order to optimize the performance of different aspects on the Uu interface, the base station can dynamically adjust the radio resource allocation, configuration, and transmission scheme based on the current channel conditions on the Uu interface. To this end, the base station may need to understand the current channel conditions on the Uu interface in both DL transmission and UL transmission. The dimensionality of the channel conditions reported in the CSI report may be high, and it may be challenging to accurately describe the channel conditions. For example, when the dimensionality of the channel conditions is high, conveying the channel conditions may involve a high number of bits. In order to solve the problem of using CSI feedback to accurately describe the channel conditions, the method may include using a smaller number of bits (e.g., using CSI compression) to report the channel conditions through CSI.

[0030] A second problem in reporting channel conditions via CSI feedback may include presenting channel conditions more accurately (i.e., with greater accuracy), in other words, increased accuracy of CSI reporting. At least some embodiments disclosed herein provide methods for addressing problems in CSI compression. In addition, at least some disclosed embodiments provide methods for presenting channel conditions more accurately.

[0031] Figure 1is a schematic diagram showing CSI feedback for DL ​​transmission and UL transmission. In 3GPP NR, CSI feedback may be used only for DL ​​transmission. In DL transmission, base station 120 is a transmitter (Transmitter, TX) of downlink traffic, and UE 110 is a receiver (Receiver, RX) of downlink traffic. In order to estimate the channel condition in downlink transmission, in 3GPP NR, base station 120 transmits reference signal RS130 on physical downlink shared channel (Physical Downlink Shared Channel, PDSCH) and physical downlink control channel (Physical Downlink Control Channel, PDCCH), as shown in FIG. Figure 1 As shown, the sequence and format of the reference signal RS130 are provided in the 3GPP NR standard. In response to receiving the reference signal RS130, the UE 110 can estimate the downlink channel condition based on the distortion of the received reference signal RS130. Subsequently, the UE 110 can provide the base station 120 with information associated with the downlink channel condition by sending CSI 140 to the base station 120 (i.e., providing CSI feedback). The content of the channel state information 140 in 3GPP NR may include a channel quality indicator (Channel Quality Indicator, CQI), a precoding matrix indicator (Precoding Matrix Indicator, PMI), a CSI-RS resource indicator (CSI-RS Resource Indicator, CRI), a synchronization signal and a physical broadcast channel resource block indicator (Synchronization Signal and Physical Broadcast Channel Resource Block Indicator, SSBRI), a layer indicator (Layer Indicator, LI) and a rank indicator (Rank Indicator, RI). In uplink transmission, UE 110 is the transmitter (TX) of uplink traffic, and base station 120 is the receiver (RX) of uplink traffic. In order to estimate the channel condition in uplink transmission, UE 110 may also transmit reference signal RS150 (reference signal on physical uplink shared channel (Physical Uplink Shared Channel, PUSCH)). Upon receiving reference signal RS150, base station 120 may estimate the uplink channel condition based on the distortion of reference signal RS150. In the uplink case, base station 120 may not send CSI to UE 110.

[0032] Figure 2is a schematic diagram showing an example of the basic architecture of an autoencoder. An autoencoder can be a type of artificial neural network that can be used for dimensionality reduction. An autoencoder can include an encoder and a decoder. The encoder can receive input data and compress it into a low-dimensional representation. The decoder can then receive the compressed representation and reconstruct the original input data. Consistent with some disclosed embodiments, the autoencoder can be used to reduce the dimensionality of CSI. The autoencoder composed of the encoder and the decoder can involve an unsupervised learning method based on a deep neural network. Figure 2 As shown, both the encoder 210 and the decoder 220 may be composed of a plurality of neural nodes 230 belonging to different stages of operation, wherein the neural nodes 230 belonging to the same stage of operation form a layer. There may be several links connecting the neural nodes in different layers, each of which may be multiplied by a weight / parameter 240. At each neural node 230, a nonlinear operation (e.g., a Sigmoid function, a step function, etc.) may be applied to all inputs (multiplied by the corresponding weight / parameter 240) to generate an output. Before the output of a neural node 230 becomes the input of a neural node 230 in the next layer, the output may be multiplied by the corresponding weight / parameter 240.

[0033] The input of encoder 210 may be high-dimensional, and the purpose of encoder 210 may be to reduce the dimensionality of the input. The output of encoder 210 may thus be a low-dimensional representation of the input of encoder 210, which then becomes the input of decoder 220. The purpose of decoder 220 may be to reconstruct the low-dimensional input into a high-dimensional output. It is contemplated that the output of decoder 220 may be the same as the input of encoder 210.

[0034] Figure 3 310 and decoder 320. The data set 330 for the autoencoder may refer to a collection of input data that may be used to train and evaluate the autoencoder model. In the context of an autoencoder, the data set 330 may include a collection of unlabeled examples or samples that the autoencoder may learn to reconstruct. In order to make the input to the encoder 310 the same as the output of the decoder 320, the weights / parameters in the encoder 310 and in the decoder 320 may be adequately updated for any input to the autoencoder. When the weights / parameters in the encoder 310 and in the decoder 320 may be insufficient based on the current input to the autoencoder, some difference or error 340 between the input to the encoder 310 and the output of the decoder 320 may be measured in some form of loss function (e.g., mean squared error, root mean squared error, normalized mean squared error, etc.), such as Figure 3As shown. Subsequently, the weights / parameters in the encoder 310 and the decoder 320 can be updated based on the current difference or error 340. It is understood that the input of the encoder 310 and the output of the decoder 320 can be the same, almost the same, similar, or any other correlation between the two that can be used from training or operation. The goal can be to make them the same, but it is understood that in some cases they may not match. For example, the difference or error 340 can represent the deviation between the input of the encoder 310 and the output of the decoder 320.

[0035] In order to make the input of the encoder the same as the output of the decoder, the weights and parameters in the encoder and decoder can be fully updated for any input to the autoencoder. Many autoencoder architectures have been proposed for applications where autoencoders may be used. The structure of an autoencoder can be specified based on several features, including but not limited to the number of layers and the number of neural nodes in each layer. Back Figure 2 In the example shown in , the encoder 210 shown may have two layers, and the second layer may have two neural nodes. The structure of the autoencoder may have different types of layers. For example, the autoencoder may include a pooling layer, a convolutional layer, etc. The connection architecture between layers may include partial connections, full connections, or any other connectivity between nodes and / or layers that can be determined based on the application. In addition, the structure of the autoencoder may include different types of connections, including forward connections, reverse connections, etc.

[0036] In the autoencoder structure, the neural nodes can implement different types of operations. For example, the neural nodes can implement sigmoid functions, step functions, or other similar data manipulation operations. The structure of the autoencoder can determine the number of weights / parameters in the encoder and in the decoder. For example, Figure 2 As shown, based on the structure determined for the self-encoder, the encoder can have W n Weights 240. The autoencoder structure may determine the type of weights / parameters 240 in the implementation. For example, weights / parameters 240 may include real numbers, complex numbers, integers, floating point numbers, or any other data type that may be determined by the structure of the autoencoder. The autoencoder structure selected for the application may determine what loss function may be used to measure the difference or error.

[0037] Figure 4 is a schematic diagram showing an example of a sandwich structure of an autoencoder, in which the structures of the encoder and the decoder are symmetrical. In the sandwich structure, the input layer 410 and the output layer 420 are located on the outermost layer, and the hidden layer 430 is sandwiched therebetween. Figure 4In the outermost structure of the autoencoder shown, the structures of the encoder and the decoder are symmetrical in several respects. First, the number of layers in the encoder and the decoder can be the same, except that there may be an additional layer in the encoder, called a code or bottleneck layer 440. Second, the number of neural nodes in the input layer 410 of the encoder can be the same as the number of neural nodes in the output layer 420 of the decoder. Third, the number of encoder hidden layers 430 and decoder hidden layers 435 (i.e., if the layer is not the input layer 410, the output layer 420, or the code / bottleneck layer 440, then the layer can be a hidden layer) in the encoder and the decoder can be the same. Fourth, the vector of weights / parameters between the input layer 410 and the first hidden layer 450 in the encoder can be the transpose of the vector of weights / parameters between the output layer 420 and the first hidden layer 480 in the decoder. The vector of weights / parameters between the first hidden layer 450 and the second hidden layer 460 in the encoder can be the transpose of the vector of weights / parameters between the first hidden layer 480 and the second hidden layer 470 in the decoder, and so on. As Figure 4 As shown, W1 T is the transpose of W1, W2 T is the transpose of W2, and so on.

[0038] for Figure 4 The sandwich structure shown, the encoder and decoder of the autoencoder can synchronize their structures and weights / parameters. In the design of the autoencoder, there may be a trade-off between the complexity of the structure and the performance of the reconstruction. The sandwich structure can be designed to significantly reduce the dimension of the encoder input and fully reconstruct the compressed representation at the output of the decoder. However, overfitting problems may occur (i.e., when the input of the autoencoder changes, the difference / error between the input and output of the autoencoder may increase). Conversely, the structure can be designed to be a simpler structure, resulting in a larger difference / error between the input and output of the autoencoder.

[0039] The following describes the implementation of CSI compression using an autoencoder. Figure 5A general architecture for applying an autoencoder to CSI compression for 3GPP NR is presented in . First, a base station 520 (denoted as gNB) sends a reference signal RS 530 (on PDSCH / PDCCH) to a UE 510. After receiving the reference signal RS 530, the UE 510 estimates the channel condition (denoted as H'). An encoder of the autoencoder (denoted as AE encoder) may be located at the UE 510, and a decoder of the autoencoder (denoted as AE decoder) may be located at the base station. After estimating H' at the UE 510, the AE encoder may compress H' into low-dimensional information. Then, a quantization and encoding process may be performed to convert the low-dimensional information into a bit stream with a specific number of bits, which may represent a new low-dimensional version 540 of the CSI. The transceiver (TX / RX) module at the UE 510 may send the low-dimensional version of the CSI to the base station 520. After receiving the low-dimensional version 540 of the CSI via the TX / RX module in the base station 520, the decoding and dequantization process in the base station 520 can be performed to convert the bit stream into low-dimensional information. Finally, the AE decoder in the base station 520 can reconstruct the low-dimensional information into the channel condition (denoted as H'). Figure 5 In the embodiment, the AE encoder and quantization & encoding in the UE 510 can realize CSI bit stream generation, while the AE decoder and decoding & inverse quantization in the base station 520 can realize CSI decoding to reconstruct a high-dimensional version of the CSI to determine the channel condition.

[0040] Under this general architecture, there may be two key challenges. First, in traditional applications where autoencoders are used (e.g., video / audio processing, principal component analysis, etc.), the autoencoders (including both the AE encoder and the AE decoder) may be installed on the same machine. In the case of CSI compression, since both the input and output of the autoencoder reside in the device in the architecture (e.g., on the UE 510 and the base station 520), the architecture may have information about both the input and output of the autoencoder, so that the architecture can determine the measurement of the difference / error between the input and output of the autoencoder. As a result, the architecture may involve updating the weights / parameters of both the AE encoder and the AE decoder based on the measured difference / error. However, when the autoencoder can be applied to CSI feedback compression, the UE 510 (AE encoder) may directly have information related to the channel conditions, while the base station 520 (AE decoder) may only have information related to the channel conditions based on the results of channel reconstruction. Therefore, UE 510 and base station 520 may not have complete information about the difference / error between the input and output of the encoder, and thus may not be able to update the weights / parameters of the AE encoder and AE decoder due to incomplete information about the channel conditions.

[0041] Second, the weights / parameters of the AE encoder and AE decoder may be updated periodically, but the difference / error may not be significantly reduced. In this case, the current structure of the AE encoder and AE decoder may not be sufficient. In this case, the structure of the AE encoder and AE decoder may need to be changed. The derivation of the sufficient structure of the AE encoder and AE decoder can be based on the difference / error between the input and output of the autoencoder. If the UE 510 and the base station 520 may not have information about the difference / error between the input and output of the autoencoder, then both the UE 510 and the base station 520 may not be able to determine the improved weights / parameters and structure for the AE encoder and AE decoder.

[0042] In order to at least partially address these two challenges in applying autoencoders to CSI compression in 3GPP NR, the disclosed embodiments provide methods and apparatus that may include two phases: an initial phase and a second phase. The purpose of the initial phase may be two parts. First, both the UE and the base station may have a common understanding of the structures supported by the AE encoder and the AE decoder. Second, both the UE and the base station may have a common understanding of the adopted CSI bitstream generation and decoding scheme. For the second phase, once the UE and the base station adopt the structures of the UE's AE encoder and the base station's AE decoder, at least some of the disclosed embodiments implement CSI compression and update the weights / parameters and structures accordingly.

[0043] For the initial stage, one or more aspects of the present application may be applicable. A certain number of encoder / decoder structures of autoencoders for CSI feedback may be provided. The base station may communicate the encoder / decoder structure (all or part of the structure) of the supported autoencoder to the UE. The information may be transmitted, for example, via a master information block (Master Information Block, MIB) or a system information block (System Information Block, SIB) in NR. The UE may communicate the encoder / decoder structure (all or part of the structure in the standard) of the supported autoencoder to the base station. The information may be transmitted, for example, via UE capability information (UE Capability Information) of radio resource control (Radio Resource Control, RRC) signaling in NR.

[0044] In some embodiments, the encoder / decoder structure of the adopted autoencoder can be determined by the network (base station or core network (CN)), and the network can notify the UE of this configuration. In other embodiments, the encoder / decoder structure of the adopted autoencoder can be determined by the UE, and the UE can notify the network of this configuration. In this case, the network can allocate radio resources to the UE to send this configuration. In other embodiments, the encoder / decoder structure of the adopted autoencoder can be specified in the standard.

[0045] In some embodiments, the adopted weights / parameters of both the AE encoder and the AE decoder may be determined by the network, and the network may communicate this configuration to the UE. In other embodiments, the adopted weights / parameters of both the AE encoder and the AE decoder may be determined by the UE, and the UE may communicate this configuration to the network. In this case, the network may allocate radio resources to the UE to send this configuration. In other embodiments, the adopted weights / parameters of both the AE encoder and the AE decoder may be specified in the standard.

[0046] A certain number of CSI bit stream generation / decoding schemes may be provided. The base station may communicate the supported CSI bit stream generation / decoding schemes (all or some of the schemes) to the UE. This information may be transmitted, for example, via a master information block (MIB) or a system information block (SIB) in NR. The UE may communicate the supported CSI bit stream generation / decoding schemes (all or some of the schemes) to the base station. This information may be transmitted, for example, via UE capability information of RRC signaling in NR.

[0047] In some embodiments, the adopted CSI bit stream generation / decoding scheme may be determined by the network, and the network may communicate this configuration to the UE. In other embodiments, the adopted CSI bit stream generation / decoding scheme may be determined by the UE, and the UE may communicate this configuration to the network. In this case, the network may allocate radio resources for the UE to send the configuration. In other embodiments, the adopted CSI bit stream generation / decoding scheme may be specified.

[0048] Consistent with some disclosed embodiments, in an alternative design for the second phase, after the initial phase, both the UE and the base station may have information associated with the adopted structure of the AE encoder and AE decoder, the weights / parameters of the AE encoder and AE decoder, and the CSI bitstream generation and decoding scheme. In the design of the second phase, the methods and apparatus described and exemplified herein may provide two alternatives. Figure 6, a block diagram of an example of an implementation of a first alternative for the second stage is shown, wherein the UE may provide updated weights / parameters to the base station in an implementation of CSI compression. The first alternative for the second stage may involve several steps. First, the base station 620 (i.e., gNB) may send a reference signal RS 630 in one or more downlink transmissions. After receiving the reference signal RS 630, the UE 610 may estimate the channel condition H'. The AE encoder 650 in the UE 610 may then reduce the dimensionality of the estimated channel condition to generate low dimensional information.

[0049] The CSI bitstream generator 655 in the UE 610 can pass the low-dimensional information to the CSI bitstream generator with a certain number of bits. The TX / RX module 680 of the UE 610 can then transmit the bitstream to the base station 620 as a low-dimensional version 660 of the CSI. Upon receiving the bitstream including the low-dimensional version 660 of the CSI, the TX / RX module 685 in the base station 620 can provide the bitstream as an input to the CSI bitstream decoder 675. The CSI bitstream decoder 675 in the base station 620 can convert the bitstream into low-dimensional information. The AE decoder 670 in the base station 620 can then reconstruct the channel conditions. The self-encoder 640 in the UE 610 can measure the difference / error between its own input and output based on a certain loss function. Based on these differences / errors, the self-encoder 640 in the UE 610 can update the weights / parameters of both the AE encoder 650 in the UE 610 and the AE decoder 670 in the base station 620. In addition, the autoencoder 640 in the UE 610 may select other structures for both the AE encoder 650 in the UE 610 and the AE decoder 670 in the base station 620. Therefore, the TX / RX module in the UE 610 may send to the base station 620 one or more of the updated weights / parameters for the AE encoder 650 in the UE 610 and / or the AE decoder 670 in the base station 620 and / or the structures of the AE encoder 650 in the UE 610 and the AE decoder 670 in the base station 620.

[0050] After receiving the updated weights / parameters for the AE encoder 650 in the UE 610 and / or the AE decoder 670 in the base station 620 and / or one or more of the structures of the AE encoder 650 in the UE 610 and the AE decoder 670 in the base station 620, the base station 620 may further confirm or determine the adopted weights / parameters of the AE encoder 650 in the UE 610 and / or the AE decoder 670 in the base station 620 and / or one or more of the adopted structures of the AE encoder 650 in the UE 610 and the AE decoder 670 in the base station 620. The base station 620 may then communicate the adopted weights / parameters of the AE encoder 650 to the UE 610 and update the AE decoder 670 in the base station 620, and / or may communicate the adopted structure of the AE encoder 650 to the UE 610 and update the AE decoder 670 in the base station 620. The AE decoder 670 in the base station 620 may then use the adopted weights / parameters or structures determined by the base station 620. The AE encoder 650 in the UE 610 and the autoencoder 640 in the UE 610 may also use the updated weights / parameters or structures determined by the base station 620 .

[0051] Consistent with the disclosed embodiments of aspects of the present application, the UE may include an autoencoder (including an encoder and a decoder), an AE encoder, and a CSI bitstream generator. The base station may include an AE decoder and a CSI bitstream decoder. After the initial stage, the UE may update the weights / parameters of the AE encoder in the UE and the weights / parameters of the AE decoder in the base station (using the autoencoder in the UE), and the UE may communicate the updated weights / parameters to the base station. The base station may allocate radio resources to the UE to upload the updated weights / parameters. After receiving the updated weights / parameters from the UE, the base station may further communicate or confirm information including the adopted weights / parameters to the UE.

[0052] In the UE, when the adopted weights / parameters provided by the base station can be received by the UE, the AE encoder can use the adopted weights / parameters provided by the base station to implement CSI compression. Then, the CSI bitstream generator can convert the compressed CSI into a bitstream with a certain number of bits. The autoencoder can further update the weights / parameters using the adopted weights / parameters provided by the base station. In the base station, the CSI bitstream decoder can convert the bitstream received from the UE to obtain the compressed CSI. The AE decoder can reconstruct the CSI using the adopted weights / parameters to determine the channel conditions.

[0053] After the initial stage, the UE may select a different AE encoder / decoder structure from a set of AE encoder / decoder structures that the UE and the base station can support. The UE may communicate the selected AE encoder / decoder structure to the base station. The base station may allocate radio resources to the UE for the UE to communicate the selected AE encoder / decoder structure to the base station. The base station may further communicate the adopted AE encoder / decoder structure to the UE. Then, the AE decoder in the base station may use the adopted structure.

[0054] Figure 7 700 is a flow chart showing a method 700 for managing channel state information feedback compression in a UE consistent with the disclosed embodiments described and exemplified herein. The method 700 may involve an initial phase ( Figure 7 ), wherein both the UE and the base station may be configured with information associated with the adopted structure of the AE encoder and AE decoder, the weights / parameters of the AE encoder and AE decoder, and the CSI bitstream generation and decoding scheme.

[0055] refer to Figure 7 , method 700 includes step 710: the UE receives at least one reference signal from the base station. In an embodiment, the UE may receive at least one reference signal that may be used for various purposes, including but not limited to synchronization, channel estimation, and signal quality measurement. (One or more) reference signals may be transmitted by the base station and received by the UE to facilitate reliable communication between the two (i.e., the UE may receive at least one reference signal from the base station). Consistent with the disclosed embodiments, the UE may use the at least one reference signal to determine the channel condition of the Uu link between the UE and the base station.

[0056] The method 700 includes step 720: the UE estimates the channel condition based on at least one reference signal. Estimating the channel condition may refer to the UE analyzing the quality and characteristics of the wireless channel between itself and the base station, including determining signal strength, interference level and other factors that can affect communication performance. The UE may estimate the channel condition by analyzing at least one reference signal received from the base station.

[0057] The method 700 includes step 730: the UE uses an encoder of a first autoencoder in the UE to reduce the dimensionality of the estimated channel condition so that a low-dimensional channel condition is generated. Reducing the dimensionality of the estimated channel condition may refer to compressing high-dimensional data into a low-dimensional representation using the encoder of the autoencoder. In some disclosed embodiments, the encoder of the autoencoder in the UE may be based on a deep neural network. The deep neural network may include a plurality of neural nodes, and each of the plurality of neural nodes may include weights and parameters for generating an output based on a neural node input. The reduction may provide a reduction in the number of bits required to transmit the estimated channel condition from the UE to the base station. It should be understood that the decoder of the corresponding autoencoder in the base station may be used to reconstruct the low-dimensional representation of the estimated channel condition received by the base station via the CSI into a high-dimensional representation.

[0058] The method 700 includes step 740: the UE uses an encoder of a second autoencoder in the UE and a decoder of the autoencoder to calculate an error measurement value based on the difference between the UE channel state input and the UE channel state output. For example, the autoencoder in the UE can measure the difference or error by comparing its input and output using a loss function. Therefore, the calculation of the error measurement value can be based on the difference between the UE channel state input and the UE channel state output, and can include using a loss function to perform the calculation.

[0059] The method 700 includes step 750: the UE calculates at least one of updated weights or updated parameters for an encoder of an autoencoder in the UE based on the value indicating the error measurement value. In view of the channel conditions and the expected error between the encoder of the autoencoder of the UE and the decoder of the autoencoder of the base station (corresponding to the value indicating the error measurement value), the UE can generate updated weights or updated parameters for the encoder of its autoencoder to take into account the value indicating the error measurement value. The decoder of the autoencoder in the base station can be provided with corresponding updated weights / parameters and structures consistent with the updated weights / parameters and structures of the encoder of the autoencoder in the UE.

[0060] The method 700 includes step 760: the UE updates at least one of the weights or parameters of the encoder of the first autoencoder in the UE based on at least one of the calculated updated weights or updated parameters. Based on at least one of the calculated updated weights or updated parameters, the UE may update the encoder of its first autoencoder to match the decoder of the autoencoder in the base station.

[0061] The method 700 includes step 770: the UE generates a CSI bitstream or at least one of at least one message based on the low-dimensional channel condition or at least one of the updated weights or updated parameters of the encoder of the first self-encoder in the UE. The CSI bitstream may refer to a bitstream containing information related to the current state of the wireless channel between the base station and the UE. At least one message may refer to at least one packet, PDU, or other data unit that can be sent and / or received in data communication. In the disclosed embodiment, at least one message may exist in the data communication from the UE to the base station (and / or from the base station to the UE). Based on the updated weights and / or updated parameters calculated and updated based on the value indicating the error measurement value, the UE may generate a channel state bitstream or at least one message, which may include the low-dimensional channel condition, at least one of the updated weights or updated parameters of the encoder of the first self-encoder in the UE, or at least one of the updated weights or updated parameters of the decoder of the self-encoder in the base station.

[0062] The method 700 includes step 780: the UE transmits at least one of a CSI bitstream or at least one message. Once the CSI can be generated, it can be transmitted to the base station via the CSI bitstream. In the disclosed embodiment, the bitstream containing information related to the current state of the wireless channel between the base station and the UE may also include information related to the autoencoder components in the base station and the UE (e.g., weights, parameters, and structural information of the autoencoder components for implementing CSI compression). Alternatively (or additionally), the information related to the autoencoder components in the base station and the UE can be transmitted in a message (e.g., at least one message) other than the CSI bitstream. For example, one or more messages can be sent from the UE to convey information related to the autoencoder components in the base station and the UE in a separate data communication.

[0063] In some embodiments, the method 700 may further include: the UE calculating at least one of the updated weights or updated parameters of the decoder of the autoencoder in the base station based on the value indicating the error measurement value; and transmitting at least one of the updated weights or updated parameters of the decoder of the autoencoder in the base station to the base station. Therefore, the UE can communicate the updated weights and / or parameters corresponding to the decoder of the autoencoder in the base station to the base station.

[0064] In some embodiments, the method 700 may further include: the UE receives one or more of an encoder weight of an autoencoder, an encoder parameter of an autoencoder, an encoder structure of an autoencoder, a decoder weight of an autoencoder, a decoder parameter of an autoencoder, or a decoder structure of an autoencoder. For example, the UE may receive an encoder structure of an autoencoder and / or a decoder structure of an autoencoder, for example, which is included in a master information block message and / or a system information block message.

[0065] In some embodiments, method 700 may involve: the UE is configured with information associated with the adopted structure of the encoder of the autoencoder, the adopted structure of the decoder of the autoencoder, the encoder weights of the autoencoder, the decoder weights of the autoencoder, the encoder structure of the autoencoder, the decoder structure of the autoencoder, the channel state information bitstream generation and the decoding scheme.

[0066] In some embodiments, the method 700 may also include: the UE transmits information related to at least one of the calculated weights or parameters of the encoder of the first autoencoder in the UE, the encoder of the second autoencoder in the UE, the decoder of the autoencoder in the UE, the decoder of the autoencoder in the base station, or the encoder of the first autoencoder in the UE, the encoder of the second autoencoder in the UE, the decoder of the autoencoder in the UE, or the decoder of the autoencoder in the base station. For example, the UE may include an algorithm that allows the calculation of weights / parameters and determines the appropriate structure in the autoencoder component to achieve CSI compression, and the UE may transmit the information to the base station.

[0067] Figure 8 8 is a flow chart illustrating a method 800 for managing channel state information feedback compression in a gNB consistent with the disclosed embodiments described and illustrated herein. The method 800 may involve an initial phase ( Figure 8 ), wherein both the base station and the UE may be configured with information associated with the adopted structure of the AE encoder and the AE decoder, the weights / parameters of the AE encoder and the AE decoder, and the CSI bitstream generation and decoding scheme.

[0068] Reference Figure 8 , method 800 includes step 810: a base station transmits at least one reference signal. In an embodiment, the base station may transmit a reference signal to a UE to allow the UE to estimate channel conditions. The reference signal may provide information related to channel quality, such as signal strength and interference level. The UE may use the received reference signal to optimize its transmission and reception on the wireless channel.

[0069] The method 800 includes step 820: the base station receives a CSI bit stream or at least one message in response to the transmitted at least one reference signal. In the disclosed embodiment, the base station can receive a bit stream that provides information about the current state of a channel between the base station and the UE based at least in part on an estimated channel condition determined by the UE. Thus, the at least one reference signal can be transmitted to the UE, and a channel state information bit stream can be received from the UE.

[0070] The method 800 includes step 830: the base station determines at least one of updated weights, updated parameters, or structures for at least one of a first autoencoder in the UE, a second autoencoder in the UE, or a decoder of an autoencoder in the UE based on at least one of a CSI bitstream or one or more messages conveying this information. In some disclosed embodiments, the base station may extract information related to updated weights, updated parameters, or updated structures of autoencoder components in the UE and the base station from at least one of the CSI bitstreams or from one or more messages conveying this information separate from the CSI bitstream.

[0071] Method 800 includes step 840: the base station calculates at least one of the updated weights or updated parameters of the decoder of the autoencoder in the base station based on at least one of the updated weights or updated parameters of the encoder of the autoencoder in the UE. In some disclosed embodiments, the base station may receive the updated weights and / or updated parameters of the encoder of the autoencoder in the UE and calculate the updated weights and / or updated parameters of the decoder of the autoencoder in the base station to correspond to the encoder of the autoencoder in the UE. The base station may determine the updated weights, parameters and / or structures of the associated autoencoder components for implementing CSI compression, and may calculate the corresponding updated weights, parameters and / or structures of the decoder of the autoencoder in the base station.

[0072] The method 800 includes step 850: the base station updates at least one of the weights, parameters or structures of the decoder of the autoencoder in the base station based on at least one of the calculated updated weights or updated parameters. Once the weights, parameters or structures of the decoder for the autoencoder have been calculated, the base station can update the weights, parameters and / or structures of the decoder of the autoencoder in the base station accordingly.

[0073] The method 800 may also include receiving from the UE at least one of the calculated weights, parameters, or updated structures for at least one of a decoder of an autoencoder in the base station, an encoder of a first autoencoder in the UE, an encoder of a second autoencoder in the UE, or a decoder of an autoencoder in the UE. In some embodiments, the updated structure may be a sandwich structure. In some embodiments, the base station may receive from the UE updated weights, parameters, and / or structures for one or more autoencoder components in the UE and the base station.

[0074] The method 800 may also include transmitting to the UE at least one of the calculated weights, parameters, or structures for at least one of a decoder of the autoencoder in the base station, an encoder of the first autoencoder in the UE, an encoder of the second autoencoder in the UE, or a decoder of the autoencoder in the UE. The base station may transmit the weights, parameters, and / or structures back to the UE to allow updating or confirming updates of the autoencoder components throughout the system to ensure that the weights, parameters, and / or structures in the UE and the base station correspond.

[0075] Consistent with some disclosed embodiments, after the initial phase, both the UE and the base station may have information associated with the adopted structure of the AE encoder and AE decoder, the weights / parameters of the AE encoder and AE decoder, and the CSI bitstream generation and decoding scheme. Fig. 9 is a schematic diagram showing an example of an implementation of such an arrangement. Fig. 9 Schematic diagram and Figure 6 There are some similarities between the schematic diagrams of Fig. 9 In the discussion of

[0076] First, the base station 920 (ie, gNB) may send a reference signal RS 930 in one or more downlink transmissions. After receiving the reference signal RS 930, the UE 910 may estimate the channel condition. The AE encoder 950 in the UE 910 may then reduce the dimension of the estimated channel condition to generate low-dimensional information H'.

[0077] The AE encoder 950 in the UE 910 can pass the low-dimensional information H' to a CSI bitstream generator 955 having a certain number of bits. The TX / RX module 990 in the UE 910 can then transmit the bitstream to the base station 920 as a low-dimensional version 960 of the CSI. Upon receiving the bitstream including the low-dimensional version 960 of the CSI, the TX / RX module 995 in the base station 920 can input the bitstream to the CSI bitstream decoder 975. The CSI bitstream decoder 975 in the base station 920 can convert the bitstream into low-dimensional information. The AE decoder 970 in the base station 920 can then reconstruct the channel conditions. The self-encoder 940 in the UE 910 can measure the difference and / or error between its own input and output based on a certain loss function. Based on these differences / errors, the UE 910 can transmit a message about these differences / errors to the base station 920. The UE 910 can transmit this message periodically, or it can transmit this message when the difference and / or error can be greater than a certain threshold. The event for which the message is to be transmitted may be determined by the base station 920 , and the base station 920 may communicate the configuration for the event to the UE 910 .

[0078] When the base station 920 receives the message conveying the difference / error from the UE 910, the autoencoder 980 in the base station 920 may update the weights / parameters for the AE encoder 950 in the UE 910 and the AE decoder 970 in the base station 920, or may select another structure for the AE encoder 950 in the UE 910 and the AE decoder 970 in the base station 920. The AE decoder 970 in the base station 920 may use the updated weights / parameters or the updated structure. The base station 920 may convey the updated weights / parameters for the AE encoder 950 in the UE 910 and / or the AE decoder 970 in the base station or the adopted structure for the AE encoder 950 in the UE 910 and / or the AE decoder 970 in the base station 920 to the UE 910. The autoencoder 940 and the AE encoder 950 in the UE 910 may use the updated weights / parameters or the updated structure.

[0079] Consistent with the disclosed embodiments, the base station may include an autoencoder (including an encoder and a decoder), an AE decoder, and a CSI bitstream decoder. The UE may have an autoencoder (including an encoder and a decoder), an AE encoder, and a CSI bitstream generator. After the initial stage, the base station may update the weights / parameters of the AE encoder in the UE and the AE decoder in the base station based on a message sent by the UE related to the feasibility of the updated weights / parameters, and the base station may communicate the updated weights / parameters to the UE. The base station may allocate radio resources to transmit the updated weights / parameters (e.g., transmitted via a control channel or a shared channel). The UE may send a message to convey information related to the feasibility of the currently adopted weights / parameters (e.g., the difference / error between the estimated channel condition and the output of the autoencoder in the UE). These differences / errors may be derived based on a specific loss function. Messages related to the feasibility of the currently adopted weights / parameters may be sent periodically, or may be sent when the difference / error is greater than a certain threshold. The event for sending the message may be determined by the base station, and the base station may communicate this configuration (for the event) to the UE. The base station may allocate radio resources for the UE to send the message.

[0080] In the UE, the autoencoder can compress the channel and reconstruct the channel using the adopted weights / parameters provided by the base station. The autoencoder can derive the error / difference between the channel condition and the reconstructed channel condition to measure the feasibility of the adopted weights / parameters communicated by the base station. The AE encoder in the UE can compress the CSI using the adopted weights / parameters provided by the base station. Then, the CSI bitstream generator can convert the compressed CSI into a bitstream with a certain number of bits. In the base station, the autoencoder can update the weights / parameters used for the autoencoder and AE encoder in the UE and the AE decoder in the base station. The CSI bitstream decoder can convert the bitstream received from the UE into compressed CSI. The AE decoder can reconstruct the CSI using the updated weights / parameters.

[0081] After the initial stage, the base station can select other structures for the autoencoder and AE encoder / decoder from the set of autoencoder and AE encoder / decoder structures supported by the UE and the base station (based on the message sent by the UE about the feasibility of the currently adopted weights / parameters), and the base station can communicate the selected structure to the UE. The base station can allocate radio resources to communicate the selected structure to the UE (for example, transmitted via a control channel or a shared channel). The UE can send a message to the base station to convey the feasibility of the currently adopted structure (for example, the difference / error between the estimated channel condition and the output of the autoencoder in the UE). These differences / errors can be derived based on a specific loss function. This message (related to the feasibility of the currently adopted structure) can be sent periodically, or it can be sent when the difference / error is greater than a certain threshold. The event to send the message can be determined by the base station, and the base station can communicate the configuration (for the event) to the UE. The base station can allocate radio resources for the UE to send the message. Then, the AE decoder in the base station can use the adopted structure. Then, the autoencoder and AE encoder in the UE can use the adopted structure.

[0082] Fig.10 1 is a flow chart showing a method 1000 for managing channel state information feedback compression in a UE consistent with the disclosed embodiments described and exemplified herein. The method 1000 may involve an initial phase ( Fig.10 ), wherein both the UE and the base station may be configured with information associated with the adopted structure of the AE encoder and AE decoder, the weights / parameters of the AE encoder and AE decoder, and the CSI bitstream generation and decoding scheme.

[0083] refer to Fig.10 , method 1000 includes step 1010: the UE receives at least one reference signal from the base station. In an embodiment, the UE may receive at least one reference signal that may be used for various purposes, including but not limited to synchronization, channel estimation, and signal quality measurement. These reference signals may be transmitted by the base station and received by the UE to facilitate reliable communication between the two (i.e., at least one reference signal may be received from the base station). Consistent with the disclosed embodiments, the UE may use the reference signal to determine the channel condition of the Uu link between the UE and the base station.

[0084] The method 1000 includes step 1020: the UE estimates the channel condition based on at least one reference signal. Estimating the channel condition may refer to the UE analyzing the quality and characteristics of the wireless channel between itself and the base station, including determining the signal strength, interference level and other factors that can affect the communication performance. The UE may estimate the channel condition by analyzing at least one reference signal received from the base station.

[0085] Method 1000 includes step 1030: the UE uses an encoder of a first autoencoder in the UE (i.e., an AE encoder) to reduce the dimensionality of the estimated channel condition so that a low-dimensional channel condition is generated. As previously described in the present disclosure, the autoencoder can be used for dimensionality reduction. In some disclosed embodiments, the estimated channel condition input to the encoder of the first autoencoder can be a high-dimensional channel condition. In the disclosed embodiments, the encoder of the first autoencoder in the UE can be used to convert the high-dimensional estimated channel condition into a low-dimensional estimated channel condition. In some disclosed embodiments, the encoder of the first autoencoder can be based on a deep neural network. The deep neural network may include a plurality of neural nodes, and each of the plurality of neural nodes may include weights and parameters for generating an output based on the neural node input. The reduction may involve a reduction in the number of bits used to transmit the estimated channel condition from the UE to the base station.

[0086] Method 1000 includes step 1040: the UE uses an encoder of a second autoencoder in the UE and a decoder of the autoencoder (which form an autoencoder) to calculate an error measurement value based on the difference between the UE channel state input and the UE channel state output. For example, the autoencoder in the UE can measure the difference or error in comparing its input and output based on a loss function. Therefore, the calculation of the error measurement value can be based on the difference between the UE channel state input and the UE channel state output, and can include using a loss function to perform the calculation. Based on the difference or error determined according to the error measurement calculated by the autoencoder, the UE can update the weights / parameters and / or structures for both the AE encoder at the UE and the AE decoder at the base station. The training of the autoencoder in the UE can configure the autoencoder to determine the difference or error between its input and its output.

[0087] Method 1000 includes step 1050: the UE transmits at least one of a CSI bit stream or at least one message, wherein the CSI bit stream or at least one of the messages is based on at least one of a low-dimensional channel condition or a value indicating an error measurement value. The CSI bit stream can be transmitted to a base station. The transmitted CSI bit stream can be received by the base station, and the CSI bit stream decoder in the base station can decode it to restore the original high-dimensional estimated channel condition. Alternatively (or additionally), one or more messages can be transmitted to the base station separately from the CSI bit stream. The CSI bit stream or one or more messages transmitted to the base station may contain at least one of a low-dimensional channel condition or a value indicating an error measurement value.

[0088] The method 1000 may also include: the UE generates at least one of a CSI bitstream or at least one message based on at least one of a low-dimensional channel condition or a value indicating an error measurement value. In the disclosed embodiments, the base station may use the low-dimensional channel condition, or the difference or error measured by the autoencoder in the UE, or both, to determine changes to weights / parameters that may be used by a decoder of the base station autoencoder (i.e., an AE decoder) to match an encoder of the autoencoder in the UE (i.e., an AE encoder).

[0089] In the initial stage, method 1000 may also include: the UE receives the encoder structure of the autoencoder, and the base station receives the decoder structure of the autoencoder. In some disclosed embodiments, the encoder of the autoencoder and the decoder structure of the autoencoder may be included in at least one of the MIB message or the SIB message. Therefore, the encoder and the decoder of the autoencoder in the UE and the autoencoder in the base station can be allocated in the initial stage of the configuration architecture to operate in accordance with the disclosed embodiments.

[0090] In some embodiments, the method 1000 may further include: the UE receives one or more of an encoder weight of an autoencoder, an encoder parameter of an autoencoder, an encoder structure of an autoencoder, a decoder weight of an autoencoder, a decoder parameter of an autoencoder, or a decoder structure of an autoencoder. For example, the UE may receive an encoder structure of an autoencoder and / or a decoder structure of an autoencoder, for example, which is included in a master information block message and / or a system information block message.

[0091] In some embodiments, method 1000 may involve: the UE is configured with information associated with the adopted structure of the encoder of the autoencoder, the adopted structure of the decoder of the autoencoder, the encoder weights of the autoencoder, the decoder weights of the autoencoder, the encoder structure of the autoencoder, the decoder structure of the autoencoder, the channel state information bitstream generation and the decoding scheme.

[0092] In some embodiments, method 1000 may also include: the UE transmits information related to at least one of the calculated weights or parameters of the encoder of the first autoencoder in the UE, the encoder of the second autoencoder in the UE, the decoder of the autoencoder in the UE, the decoder of the autoencoder in the base station, or the encoder of the first autoencoder in the UE, the encoder of the second autoencoder in the UE, the decoder of the autoencoder in the UE, or the decoder of the autoencoder in the base station. For example, the UE may include an algorithm that allows the calculation of weights / parameters and determines the appropriate structure in the autoencoder component to achieve CSI compression, and the UE may transmit the information to the base station.

[0093] Fig.111 is a flow chart illustrating a method 1100 for managing CSI feedback compression in a gNB consistent with the disclosed embodiments described and illustrated herein. The method 1100 may involve an initial phase ( Fig.11 ), wherein both the base station and the UE may be configured with information associated with the adopted structure of the AE encoder and the AE decoder, the weights / parameters of the AE encoder and the AE decoder, and the CSI bitstream generation and decoding scheme.

[0094] Reference Fig.11 , method 1100 includes step 1110: a base station transmits at least one reference signal. In an embodiment, the base station may transmit a reference signal to a UE to allow the UE to estimate channel conditions. The reference signal may provide information related to channel quality, such as signal strength and interference level. The UE may use the received reference signal to optimize its transmission and reception.

[0095] The method 1100 includes step 1120: the base station receives at least one of a CSI bit stream or at least one message in response to the transmitted at least one reference signal. In the disclosed embodiments, the base station may receive a bit stream that provides information about the current state of a channel between the base station and the UE based at least in part on an estimated channel condition determined by the UE. Thus, at least one reference signal may be transmitted to the UE, and a CSI bit stream may be received from the UE. Alternatively, the base station may receive at least one message from the UE, including the information described in the disclosed embodiments.

[0096] Method 1100 includes step 1130: the base station determines at least one of a low-dimensional channel condition or a value indicating an error measurement value based on a CSI bitstream or at least one message. As described and exemplified herein, the base station may extract the low-dimensional channel condition and / or the value indicating the error measurement value received from the UE from the CSI bitstream using a CSI bitstream decoder. Alternatively, the base station may also receive the value indicating the error measurement value from one or more messages separately from the CSI bitstream. The base station may input the recovered low-dimensional channel condition information into the decoder of the autoencoder (i.e., the AE decoder) to restore the original high-dimensional channel condition determined by the UE. Therefore, the decoder of the autoencoder in the base station may be used to determine the estimated channel condition to decode the low-dimensional channel condition and restore the original estimated channel condition. In some disclosed embodiments, the decoder of the autoencoder in the base station may be based on a deep neural network. The deep neural network implementing the decoder of the autoencoder may include a plurality of neural nodes, and each of the plurality of neural nodes may include weights and parameters for generating an output based on the neural node input. In some embodiments, the decoder of the autoencoder in the base station may be used to convert the low-dimensional channel condition into a high-dimensional channel condition.

[0097] The method 1100 includes step 1140: the base station calculates at least one of the updated weights or updated parameters of the decoder of the autoencoder in the base station based on at least one of the estimated channel conditions or the value indicating the error measurement value, wherein the estimated channel conditions can be based on the low-dimensional channel conditions. In the disclosed embodiment, the UE can provide information related to the channel conditions via the CSI bitstream, so that the difference caused by the change in the channel conditions can be compensated by updating the weights / parameters and / or structures of the autoencoder components in the UE and the base station to correct the inaccuracy in the CSI compression. Therefore, calculating at least one of the updated weights or updated parameters of the decoder of the autoencoder in the base station may include: calculating at least one of the updated weights or updated parameters of the decoder of the autoencoder in the base station based on the value indicating the error measurement value. For example, the error measurement value can be used to determine the changes that may need to be made to the weights or parameters of the decoder of the autoencoder in order to match the encoder of the autoencoder in the UE. Therefore, the error measurement value can be used in the algorithm implemented in the base station to generate the updated weights and / or updated parameters of the decoder of the autoencoder in the base station. At least one of the updated weights or updated parameters of the decoder of the autoencoder in the base station may correspond to at least one of the updated weights or updated parameters of the encoder of the autoencoder in the UE. At least one of the encoder structure of the autoencoder or the decoder structure of the autoencoder may be transmitted to the UE.

[0098] The method 1100 includes step 1150: the base station updates at least one of the weights or parameters of the decoder of the autoencoder in the base station based on at least one of the calculated updated weights or updated parameters. Step 1150 implements the update of the decoder of the autoencoder in the base station to correspond to the update of the encoder of the autoencoder in the UE and compensate for the channel condition change that may have caused the difference or error of CSI compression between the UE and the base station.

[0099] In some disclosed embodiments, calculating at least one of the updated weights or updated parameters may also include: calculating at least one of the updated weights or updated parameters for each of the encoder of the first autoencoder in the UE (i.e., the AE encoder), the encoder of the second autoencoder in the UE, and the decoder of the autoencoder in the UE, wherein the encoder of the second autoencoder in the UE and the decoder of the autoencoder belong to the same autoencoder pair. For example, the autoencoder components in the UE and the base station may need to be calculated to provide the same (or almost the same) output weights and parameters for a specific input. Therefore, each autoencoder component in the UE and the base station may need updated weights, updated parameters, and / or updated structures to maintain matching outputs. In addition, calculating at least one of the updated weights or updated parameters may include: updating at least one of the weights or parameters of the encoder for the first autoencoder in the UE, the encoder for the second autoencoder in the UE, and the decoder for the autoencoder in the UE. Once the weights and / or parameters are calculated, these weights and parameters can be used to update the weights and parameters in the associated autoencoder components.

[0100] In some embodiments, method 1100 may involve: the base station is configured with information associated with the adopted structure of the encoder of the autoencoder, the adopted structure of the decoder of the autoencoder, the encoder weights of the autoencoder, the decoder weights of the autoencoder, the encoder structure of the autoencoder, the decoder structure of the autoencoder, the channel state information bitstream generation and the decoding scheme.

[0101] Fig.12 1 shows a block diagram of a device 1200 consistent with some embodiments of the present disclosure. The device 1200 may be a network node, a roadside unit, a relay node, a base station (e.g., Figure 6 Base station 620 or Fig. 9 920) or a UE (eg, Figure 6 UE 610 or Fig. 9UE 910 in ). The device 1200 may take any form, including but not limited to a computer system, a vehicle, a component installed in a vehicle, a roadside unit, a laptop computer, a wireless terminal including a mobile phone, a wireless handheld device, or a wireless personal device, or any other form. The device 1200 may include an antenna 1202, which may be used to transmit electromagnetic signals to a base station, a UE or other UE, or to receive electromagnetic signals from a base station, a UE or other device. The antenna 1202 may include one or more antenna elements, and may enable different input-output antenna configurations, such as a multiple input multiple output (MIMO) configuration, a multiple input single output (MISO) configuration, and a single input multiple output (SIMO) configuration. In some embodiments, the antenna 1202 may include multiple (e.g., tens or hundreds) of antenna elements, and may enable multi-antenna functions such as beamforming. In some embodiments, the antenna 1202 is a single antenna.

[0102] Device 1200 may include a transceiver 1204 coupled to antenna 1202. Transceiver 1204 may be a wireless transceiver at device 1200 and may communicate bidirectionally with a base station, UE, or other device. Figure 6 TX / RX module 680 or TX / RX module 685, or Fig. 9 The TX / RX module 990 or TX / RX module 995 in the embodiment may receive / transmit wireless signals from / to the UE or gNB in ​​Uu communication. The transceiver 1204 may include a modem for modulating packets and providing the modulated packets to the antenna 1202 for transmission, and for demodulating packets received from the antenna 1202.

[0103] Device 1200 may include memory 1206. Memory 1206 may be any type of computer-readable storage medium, including volatile or non-volatile memory devices, or a combination thereof. Computer-readable storage media include, but are not limited to, non-transitory computer storage media. Non-transitory storage media may be accessed by general-purpose or special-purpose computers. Examples of non-transitory storage media include, but are not limited to, portable computer floppy disks, hard disks, random access memory (RAM), erasable programmable read-only memory (EPROM), read-only memory (ROM), electrically erasable programmable ROM (EEPROM), digital versatile disks (DVD), flash memory, compact disk (CD) ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, etc. Non-transitory media may be used to carry or store desired program code devices (e.g., instructions and / or data structures), and may be accessed by general-purpose or special-purpose computers, or general-purpose or special-purpose processors. In some examples, the software / program code may be transmitted from a remote source (e.g., a website, a server, etc.) using a coaxial cable, a fiber optic cable, a twisted pair, a digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwaves. In such examples, coaxial cables, fiber optic cables, twisted pairs, DSL, or wireless technologies such as infrared, radio, and microwaves are within the scope of the medium definition. Combinations of the above examples are also within the scope of computer-readable media.

[0104] The memory 1206 may store information related to the identification of the device 1200 and the signals and / or data received by the antenna 1202. The memory 1206 may also store post-processed signals and / or data. The memory 1206 may also store computer-readable program instructions, mathematical models, and algorithms used in signal processing in the transceiver 1204 and calculations in the processor 1208. The memory 1206 may also store computer-readable program instructions for execution by the processor 1208 to operate the device 1200 to perform various functions described in the present disclosure. In some examples, the memory 1206 may include a basic input / output system (BIOS), which may control basic hardware or software operations, such as interaction with peripheral components or devices.

[0105] The computer-readable program instructions of the present disclosure can be assembly instructions, instruction set architecture (Instruction-Set-Architecture, ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages ​​(including object-oriented programming languages ​​and traditional procedural programming languages). The computer-readable program instructions can be executed completely on a computing device as an independent software package, or partly executed on a first computing device and partly executed on a second computing device away from the first computing device. In the latter scenario, the second computing device away can be connected to the first computing device through any type of network (including a local area network (Local Area Network, LAN) or a wide area network (Wide Area Network, WAN)).

[0106] The device 1200 may include a processor 1208, which may include a hardware device with processing capabilities. The processor 1208 may include at least one of a general-purpose processor, a digital signal processor (DSP), a central processing unit (CPU), a microcontroller, an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), a programmable logic device, a discrete gate or transistor logic component, a discrete hardware component, or other programmable logic devices. Examples of general-purpose processors include, but are not limited to, microprocessors, any conventional processors, controllers, microcontrollers, or state machines. In some embodiments, the processor 1208 may be implemented using a combination of devices (e.g., a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors in combination with a DSP core, or any other such configuration). The processor 1208 may receive downlink signals, uplink signals, reference signals from the transceiver 1204, and further process these signals. The processor 1208 may also receive data packets from the transceiver 1204, and further process these packets. In some embodiments, the processor 1208 may be configured to operate the memory using a memory controller. In some embodiments, the memory controller may be integrated into the processor 1208. The processor 1208 may be configured to execute computer-readable instructions stored in a memory (e.g., memory 1206) to enable the device 1200 to perform various functions.

[0107] Device 1200 may include a global positioning system (GPS) 1210. GPS 1210 may be used to enable location-based services or other services based on the geographic location of device 1200. GPS 1210 may receive a global navigation satellite system (GNSS) signal from a single satellite or multiple satellite signals via antenna 1202 and provide the geographic location of device 1200 (e.g., coordinates of device 1200). In some embodiments, for example, when device 1200 may be a base station in a substantially fixed position, GPS 1210 may be omitted.

[0108] The device 1200 may include an input / output (I / O) device 1212, which may be used to communicate the results of signal processing and calculation to a user or other device. The I / O device 1212 may include a user interface, which includes a display and an input device for transmitting user commands to the processor 1208. The display may be configured to display the state of signal reception at the device 1200, the data stored at the memory 1206, the state of signal processing and the results of calculation, etc. The display may include, but is not limited to, a cathode ray tube (CRT), a liquid crystal display (LCD), a light-emitting diode (LED), a plasma gas display, a touch screen, or other image projection devices for displaying information to a user. The input device may be any type of computer hardware device for receiving data and control signals from a user. The input device may include, but is not limited to, a keyboard, a mouse, a scanner, a digital camera, a joystick, a trackball, a cursor direction key, a touch screen monitor, or an audio / video commander, etc.

[0109] The device 1200 may also include a machine interface 1214 , such as an electrical bus that connects the transceiver 1204 , the memory 1206 , the processor 1208 , the GPS 1210 , and the I / O devices 1212 .

[0110] In some embodiments, the device 1200 may be configured or programmed to manage channel state information feedback compression. The processor 1208 may be configured to execute instructions stored in the memory 1206 to perform the combined Figure 7 , Figure 8 , Fig.10 and Fig.11 At least one of methods 700, 800, 1000, or 1110, respectively described.

[0111] In some embodiments, the device 1200 may be configured or programmed to transmit CSI over the Uu interface. For example, the device 1200 may be a UE in Uu interface communication, and the processor 1208 may be configured to execute instructions stored in the memory 1206 to determine channel state information; store the channel state information; and transmit the channel state information to the base station. The device 1200 may include any known UE elements.

[0112] In some embodiments, device 1200 may be a UE, and processor 1208 may execute instructions stored in memory 1206 to: receive at least one reference signal; estimate channel conditions based on at least one reference signal; use an encoder of a first autoencoder in the UE to reduce the dimension of the estimated channel conditions so as to generate low-dimensional channel conditions; use an encoder of a second autoencoder in the UE and a decoder of the autoencoder to calculate an error measurement value based on a difference between a UE channel state input and a UE channel state output; and transmit a CSI bit stream, wherein the CSI bit stream is based on at least one of the low-dimensional channel conditions or a value indicating an error measurement value.

[0113] In some embodiments, device 1200 may be a base station (e.g., a gNB), and processor 1208 may execute instructions stored in memory 1206 to: transmit at least one reference signal; receive a channel state information bit stream in response to the transmitted at least one reference signal; determine, based on the CSI bit stream, at least one of a low-dimensional channel condition or a value indicating an error measurement value; calculate, based on at least one of the estimated channel condition or the value indicating the error measurement value, at least one of updated weights or updated parameters for a decoder of an autoencoder in the base station, wherein the estimated channel condition is based on the low-dimensional channel condition; and update at least one of the weights or parameters of the decoder of the autoencoder in the base station based on at least one of the calculated updated weights or updated parameters.

[0114] One or more aspects of the present disclosure may be related to or otherwise incorporate features of “Study on AI / ML for NR Air Interface” in 3GPP Release 18.

[0115] All processes described herein can be fully automated via software code modules, including one or more specific computer executable instructions executed by a computing system. The computing system may include one or more computers or processors. The code modules may be stored in any type of non-transitory computer-readable medium or other computer storage device. Some or all of the methods may be embodied in dedicated computer hardware. In at least some embodiments consistent with the present disclosure, the weights / parameters of the AE encoder and AE decoder are updated at the UE side or at the gNB side.

[0116] In addition to what is described herein, many variations will be apparent from this disclosure. For example, depending on the embodiment, certain actions, events, or functions in any algorithm described herein may be performed in a different order, may be added, merged, or omitted entirely (e.g., not all described actions or events are necessary for the practice of the algorithm). In addition, in some embodiments, actions or events may be performed simultaneously, e.g., by multithreading, interrupt handling, or multiple processors or processor cores or other parallel architectures, rather than sequentially. In addition, different tasks or processes may be performed by different machines and / or computing systems that can run together.

[0117] The various illustrative logic blocks and modules described herein in conjunction with the disclosed embodiments may be implemented or executed by a machine, such as a processing unit or processor designed to perform the functions described herein, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof. The processor may be a microprocessor, but in an alternative, the processor may be a controller, a microcontroller or a state machine, or a combination thereof, etc. The processor may include a circuit configured to process computer executable instructions. In another embodiment, the processor includes an FPGA or other programmable device that performs logical operations without processing computer executable instructions. The processor may also be implemented as a combination of computing devices, for example, a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in combination with a DSP core, or any other such configuration. Although this article is mainly described with respect to digital technology, the processor may also mainly include analog components. The computing environment may include any type of computer system including, but not limited to, a microprocessor-based computer system, a mainframe computer, a digital signal processor, a portable computing device, a device controller, or a computing engine within an appliance, to name a few.

[0118] Conditional language such as "can," "can," "might," or "may," unless otherwise specifically stated, is understood in the context of common usage to convey that some embodiments include, while other embodiments do not include, certain features, elements, and / or steps. Thus, such conditional language is generally not intended to imply that such features, elements, and / or steps are in any way required for one or more embodiments, or that one or more embodiments necessarily include logic for determining (with or without user input or prompting) whether such features, elements, and / or steps are included or to be performed in any particular embodiment.

[0119] Any process description, element or block in the flowcharts described herein and / or depicted in the accompanying drawings should be understood to represent modules, segments or code portions that include one or more executable instructions for implementing a specific logical function or element in the process. Alternative implementations are included within the scope of the embodiments described herein in which, as will be appreciated by those skilled in the art, elements or functions may be deleted, executed in a different order than shown or discussed (including substantially simultaneously or in reverse order), depending on the functions involved.

[0120] Although the example embodiments provided in this disclosure refer to examples of 3GPP NR radio access technology, the scope of the solutions in this disclosure is by no means limited to these examples. They can be applied to, for example, 3GPP LTE or 3GPP 6G. They can also be applied to non-3GPP radio access technologies, such as IEEE 802.11 technology (for example, but not limited to 802.11n, 802.11u or 802.11p).

[0121] As used in this disclosure, the use of the term "or" in a list of items indicates an inclusive list. A list of items may be prefixed with phrases such as "at least one" or "one or more". For example, a list of at least one of A, B, or C includes A or B or C or AB (i.e., A and B) or AC or BC or ABC (i.e., A and B and C). In addition, as used in this disclosure, prefixing a list of conditions with the phrase "based on" should not be interpreted as "based only on" a set of conditions, but rather should be interpreted as "based at least in part on" a set of conditions. For example, a result described as "based on condition A" may be based on both condition A and condition B without departing from the scope of this disclosure.

[0122] In this specification, the terms "include", "contain" or "comprises" are used interchangeably and have the same meaning and are interpreted as inclusive and open-ended. The terms "include", "contain" or "comprises" may be used before a list of elements and indicate that at least all of the listed elements in the list are present, but other elements that are not in the list may also be present. For example, if A includes B and C, then {B, C} and {B, C, D} are both within the scope of A.

[0123] In conjunction with the accompanying drawings, the present disclosure describes example configurations that do not represent all examples that can be implemented or all configurations within the scope of the present disclosure. The term "exemplary" should not be interpreted as "preferred" or "advantageous compared to other examples", but should be interpreted as "illustration, instance, or example". By reading this disclosure, including the description of the embodiments and the accompanying drawings, a person of ordinary skill in the art will understand that alternative embodiments can be used to implement the technology disclosed herein. Those skilled in the art will understand that the embodiments described herein or certain features of the embodiments can be combined to obtain other embodiments for practicing the technology described in the present disclosure. Therefore, the present disclosure is not limited to the examples and designs described herein, but should conform to the widest scope consistent with the principles and novel features disclosed herein.

[0124] The flowcharts and block diagrams in the figure show examples of the architecture, functions and operations of possible implementations of the systems, methods and devices according to various embodiments. It should be noted that in some alternative implementations, the functions marked in the box may occur outside the order marked in the figure. For example, depending on the functions involved, the two boxes shown in succession can actually be executed substantially simultaneously, or the boxes can sometimes be executed in reverse order. Similarly, in the method consistent with various embodiments, additional steps may be included in such a method, and some steps may be omitted or combined.

[0125] It should be understood that the described embodiments are not mutually exclusive, and elements, components, materials or steps described in conjunction with one exemplary embodiment may be combined with or eliminated from other embodiments in a suitable manner to achieve the desired design goals.

[0126] References herein to "some embodiments" or "some exemplary embodiments" mean that a particular feature, structure, or characteristic described in conjunction with the embodiment may be included in at least one embodiment. The appearance of the phrases "one embodiment," "some embodiments," or "another embodiment" throughout this disclosure does not necessarily refer to the same embodiment, nor are they necessarily separate or alternative embodiments that are mutually exclusive with other embodiments.

[0127] In addition, the articles "a" and "an" as used in this disclosure and the appended claims should generally be construed to mean "one or more" unless specified otherwise or clearly directed to a singular form by context.

[0128] Unless expressly stated otherwise, each numerical value and range should be interpreted as being approximate, as would the word "about" or "approximately" preceding the value of the numerical value or range.

[0129] Although elements in the following method claims, if any, are recited in a specific order, these elements are not necessarily intended to be limited to being implemented in that specific order unless the claim recitation otherwise implies a specific order for implementing some or all of these elements.

[0130] It should be understood that certain features of the present disclosure described in the context of separate embodiments for the sake of clarity may also be provided in combination in a single embodiment. Conversely, various features of the present specification described in the context of a single embodiment for the sake of brevity may also be provided separately, or in any suitable sub-combination, or as appropriate in any other described embodiment of the present specification. Certain features described in the context of various embodiments are not essential features of these embodiments unless otherwise stated.

[0131] It will be further understood that those skilled in the art may make various modifications, substitutions and changes to the details, materials and arrangements of the components described and illustrated for the purpose of explaining the nature of the described embodiments without departing from the scope. Accordingly, the appended claims cover all such substitutions, modifications and changes falling within the aspects of the claims.

[0132] Clause 1: A method for managing compression of channel state information feedback, the method comprising: receiving at least one reference signal; estimating a channel condition based on the at least one reference signal; Using an encoder of a first autoencoder in a user equipment (UE), reducing the dimensionality of the estimated channel state so as to generate a low dimensional channel state; calculating an error measurement value based on a difference between a UE channel state input and a UE channel state output using an encoder of a second autoencoder and a decoder of the autoencoder in the UE; and Transmitting at least one of a channel state information bitstream or at least one message, wherein the at least one of the channel state information bitstream or at least one message is based on at least one of the low dimensional channel condition or a value indicative of the error measurement value.

[0133] Clause 2: The method according to clause 1, further comprising: At least one of the channel state information bitstream or at least one message is generated based on at least one of the low dimensional channel condition or a value indicative of the error measurement value.

[0134] Clause 3: The method of clause 1, wherein the at least one reference signal is received from a base station.

[0135] Clause 4: The method of clause 1, wherein at least one of the channel state information bitstream or at least one message is transmitted to the base station.

[0136] Clause 5: A method according to clause 4, wherein at least one of the channel state information bit stream or at least one message transmitted to the base station includes at least one of the low-dimensional channel condition or a value indicative of the error measurement value.

[0137] Clause 6: The method of clause 1, wherein the encoder of the first autoencoder is based on a deep neural network.

[0138] Clause 7: The method of clause 6, wherein the deep neural network comprises a plurality of neural nodes, and each of the plurality of neural nodes comprises weights and parameters for generating an output based on a neural node input.

[0139] Clause 8: The method of clause 1, further comprising: inputting the estimated channel conditions into an encoder of the first autoencoder, wherein the estimated channel conditions input into the encoder of the first autoencoder are high-dimensional channel conditions.

[0140] Clause 9: The method of clause 1, wherein calculating the error measure based on the difference between the UE channel state input and the UE channel state output comprises using a loss function.

[0141] Clause 10: The method according to clause 1 further includes: calculating updated weights and parameters of the encoder of the first autoencoder and the decoder of the autoencoder based on the error measurement value.

[0142] Clause 11: The method according to clause 1 also includes: receiving at least one of the encoder weights of the autoencoder, the encoder parameters of the autoencoder, the encoder structure of the autoencoder, the decoder weights of the autoencoder, the decoder parameters of the autoencoder or the decoder structure of the autoencoder.

[0143] Clause 12: A method according to clause 1, wherein the UE is configured with information associated with the adopted structure of the encoder of the autoencoder, the adopted structure of the decoder of the autoencoder, the encoder weights of the autoencoder, the decoder weights of the autoencoder, the encoder structure of the autoencoder, the decoder structure of the autoencoder, the channel state information bitstream generation, and the decoding scheme.

[0144] Clause 13: The method of clause 1, further comprising: receiving at least one of an encoder structure from an encoder or a decoder structure from an encoder.

[0145] Clause 14: The method of clause 13, wherein at least one of an encoder structure of the autoencoder or a decoder structure of the autoencoder is included in at least one of a master information block message or a system information block message.

[0146] Clause 15: An apparatus for channel state information feedback compression, the apparatus comprising: a memory storing instructions; and a processor configured to execute instructions stored in the memory to: receiving at least one reference signal; estimating a channel condition based on the at least one reference signal; Using an encoder of a first autoencoder in a user equipment (UE), reducing the dimensionality of the estimated channel state so as to generate a low dimensional channel state; calculating an error measurement value based on a difference between a UE channel state input and a UE channel state output using an encoder of a second autoencoder and a decoder of the autoencoder in the UE; and Transmitting at least one of a channel state information bitstream or at least one message, wherein the at least one of the channel state information bitstream or at least one message is based on at least one of the low dimensional channel condition or a value indicative of the error measurement value.

[0147] Clause 16: The apparatus of clause 15, wherein the processor is further configured to execute instructions stored in the memory to: At least one of the channel state information bitstream or at least one message is generated based on at least one of the low dimensional channel condition or a value indicative of the error measurement value.

[0148] Clause 17: The apparatus of clause 15, wherein the at least one reference signal is received from a base station.

[0149] Clause 18: The apparatus of clause 15, wherein at least one of the channel state information bitstream or at least one message is transmitted to the base station.

[0150] Clause 19: An apparatus according to clause 18, wherein at least one of the channel state information bit stream or at least one message transmitted to the base station includes at least one of the low-dimensional channel condition or a value indicative of the error measurement value.

[0151] Clause 20: The apparatus of clause 15, wherein the encoder of the first autoencoder is based on a deep neural network.

[0152] Clause 21: The apparatus of clause 20, wherein the deep neural network comprises a plurality of neural nodes, and each of the plurality of neural nodes comprises weights and parameters for generating an output based on a neural node input.

[0153] Clause 22. The apparatus of clause 15, wherein the processor is further configured to execute instructions stored in the memory to: The estimated channel conditions are input into an encoder of the first autoencoder, wherein the estimated channel conditions input into the first autoencoder are high-dimensional channel conditions.

[0154] Clause 23: The apparatus of clause 15, wherein the error measure is calculated based on a difference between a UE channel state input and a UE channel state output.

[0155] Clause 24: The apparatus of clause 15, wherein the processor is further configured to execute instructions stored in the memory to: Based on the error measurement value, updated weights and parameters of the encoder of the first autoencoder and the decoder of the autoencoder are calculated.

[0156] Clause 25: An apparatus according to clause 15, wherein the processor is further configured to execute instructions stored in the memory to: receive at least one of an encoder weight of an autoencoder, an encoder parameter of an autoencoder, an encoder structure of an autoencoder, a decoder weight of an autoencoder, a decoder parameter of an autoencoder, or a decoder structure of an autoencoder.

[0157] Clause 26: An apparatus according to clause 15, wherein the UE is configured with information associated with the adopted structure of the encoder of the autoencoder, the adopted structure of the decoder of the autoencoder, the encoder weights of the autoencoder, the decoder weights of the autoencoder, the encoder structure of the autoencoder, the decoder structure of the autoencoder, the channel state information bitstream generation and the decoding scheme.

[0158] Clause 27: The apparatus of clause 15, the processor further configured to execute instructions stored in the memory to: receive at least one of an encoder structure from an encoder or a decoder structure from an encoder.

[0159] Clause 28: The apparatus of clause 27, wherein at least one of an encoder structure of the autoencoder or a decoder structure of the autoencoder is included in at least one of a master information block message or a system information block message.

[0160] Clause 29: A non-transitory computer-readable medium storing instructions executable by one or more processors of an apparatus for managing channel state information feedback compression to perform a method comprising: receiving at least one reference signal; estimating a channel condition based on the at least one reference signal; Using an encoder of a first autoencoder in a user equipment UE, reducing the dimensionality of the estimated channel state so as to generate a low dimensional channel state; calculating an error measurement value based on a difference between a UE channel state input and a UE channel state output using an encoder of a second autoencoder and a decoder of the autoencoder in the UE; and A channel state information bitstream or at least one message is transmitted, wherein at least one of the channel state information bitstream or at least one message is based on at least one of the low dimensional channel condition or a value indicative of the error measurement value.

[0161] Clause 30: A method for managing compression of channel state information feedback, the method comprising: transmitting at least one reference signal; receiving at least one of a channel state information bitstream or at least one message in response to the transmitted at least one reference signal; determining at least one of a low dimensional channel condition or a value indicative of an error measure based on at least one of the channel state information bitstream or the at least one message; Calculating at least one of updated weights or updated parameters for a decoder of an autoencoder in a base station based on at least one of an estimated channel condition or a value indicative of an error measurement, wherein the estimated channel condition is based on the low-dimensional channel condition; and At least one of the weights or parameters of the decoder of the autoencoder in the base station is updated based on at least one of the calculated updated weights or updated parameters.

[0162] Clause 31: The method of clause 30, wherein the estimated channel conditions are determined by decoding the low-dimensional channel conditions using a decoder of the autoencoder in the base station.

[0163] Clause 32: The method of clause 30, wherein the at least one reference signal is transmitted to a user equipment (UE) and at least one of the channel state information bitstream or the at least one message is received from the UE.

[0164] Clause 33: A method according to clause 30, wherein determining at least one of the values ​​of the low-dimensional channel condition or the indicative error measurement value comprises: extracting at least one of the values ​​of the low-dimensional channel condition or the indicative error measurement value from at least one of the channel state information bit stream or at least one message.

[0165] Clause 34: A method according to clause 30, wherein the decoder of the autoencoder in the base station is based on a deep neural network.

[0166] Clause 35: The method of clause 34, wherein the deep neural network comprises a plurality of neural nodes, and each of the plurality of neural nodes comprises at least one of a weight or a parameter for generating an output based on a neural node input.

[0167] Clause 36: The method of clause 30, further comprising: converting the low-dimensional channel conditions into high-dimensional channel conditions using a decoder of the autoencoder.

[0168] Clause 37: A method according to clause 30, wherein calculating at least one of the updated weights or updated parameters of the decoder of the autoencoder in the base station includes: calculating at least one of the updated weights or updated parameters of the decoder of the autoencoder in the base station based on the value of the indicative error measurement value.

[0169] Clause 38: A method according to clause 37, wherein at least one of the updated weights or updated parameters for the decoder of the autoencoder corresponds to at least one of the updated weights or updated parameters for the encoder of the autoencoder in the user equipment.

[0170] Clause 39: A method according to clause 30, wherein calculating at least one of the updated weights or updated parameters further includes: calculating at least one of the updated weights or updated parameters for each of an encoder of a first autoencoder in a user equipment (UE), an encoder of a second autoencoder in the UE, and a decoder of the autoencoder in the UE, and wherein the encoder of the second autoencoder in the UE and the decoder of the autoencoder in the UE belong to the same autoencoder pair.

[0171] Clause 40: A method according to Clause 39, wherein calculating at least one of the updated weights or updated parameters also includes: updating at least one of the weights or parameters of the encoder for the first autoencoder in the UE, the encoder for the second autoencoder in the UE, and the decoder for the autoencoder in the UE.

[0172] Clause 41: A method according to clause 30, wherein the base station is configured with information associated with the adopted structure of the encoder of the autoencoder, the adopted structure of the decoder of the autoencoder, the encoder weights of the autoencoder, the decoder weights of the autoencoder, the encoder structure of the autoencoder, the decoder structure of the autoencoder, the channel state information bitstream generation and the decoding scheme.

[0173] Clause 42: The method of clause 30, further comprising: transmitting at least one of an encoder structure of the autoencoder or a decoder structure of the autoencoder to a user equipment.

[0174] Clause 43: An apparatus for managing compression of channel state information feedback, the apparatus comprising: a memory storing instructions; and a processor configured to execute instructions stored in the memory to: transmitting at least one reference signal; receiving at least one of a channel state information bitstream or at least one message in response to the transmitted at least one reference signal; determining at least one of a low dimensional channel condition or a value indicative of an error measure based on at least one of the channel state information bitstream or the at least one message; Calculating at least one of updated weights or updated parameters for a decoder of an autoencoder in a base station based on at least one of an estimated channel condition or a value indicative of an error measurement, wherein the estimated channel condition is based on the low dimensional channel condition; and At least one of the weights or parameters of the decoder of the autoencoder in the base station is updated based on at least one of the calculated updated weights or updated parameters.

[0175] Clause 44: The apparatus of clause 43, wherein the estimated channel conditions are determined by decoding the low-dimensional channel conditions using a decoder of the autoencoder in the base station.

[0176] Clause 45: The apparatus of clause 43, wherein the at least one reference signal is transmitted to a user equipment (UE) and at least one of the channel state information bitstream or the at least one message is received from the UE.

[0177] Clause 46: An apparatus according to clause 43, wherein determining at least one of the values ​​of the low-dimensional channel condition or the indicative error measurement value comprises: extracting at least one of the values ​​of the low-dimensional channel condition or the indicative error measurement value from at least one of the channel state information bit stream or at least one message.

[0178] Clause 47: An apparatus according to clause 43, wherein the decoder of the autoencoder in the base station is based on a deep neural network.

[0179] Clause 48: The apparatus of clause 47, wherein the deep neural network comprises a plurality of neural nodes, and each of the plurality of neural nodes comprises at least one of a weight or a parameter for generating an output based on a neural node input.

[0180] Clause 49: The apparatus of clause 43, wherein the processor is further configured to execute instructions stored in the memory to: The low-dimensional channel conditions are converted into high-dimensional channel conditions using a decoder of the autoencoder.

[0181] Clause 50: An apparatus according to clause 43, wherein calculating at least one of the updated weights or updated parameters of the decoder of the autoencoder in the base station includes: calculating at least one of the updated weights or updated parameters of the decoder of the autoencoder in the base station based on the value of the indicative error measurement value.

[0182] Clause 51: An apparatus according to clause 50, wherein at least one of the updated weights or updated parameters for the decoder of the autoencoder corresponds to at least one of the updated weights or updated parameters for the encoder of the autoencoder in the user equipment.

[0183] Clause 52: An apparatus according to claim 43, wherein calculating at least one of the updated weights or updated parameters further includes: calculating at least one of the updated weights or updated parameters for each of an encoder of a first autoencoder in a user equipment (UE), an encoder of a second autoencoder in the UE, and a decoder of the autoencoder in the UE, and wherein the encoder of the second autoencoder in the UE and the decoder of the autoencoder in the UE belong to the same autoencoder pair.

[0184] Clause 53: An apparatus according to clause 52, wherein calculating at least one of the updated weights or updated parameters further comprises updating at least one of the weights, parameters or structures of an encoder for the first autoencoder in the UE, an encoder for the second autoencoder in the UE and a decoder for the autoencoder in the UE.

[0185] Clause 54: An apparatus according to clause 43, wherein the base station is configured with information associated with the adopted structure of the encoder of the autoencoder, the adopted structure of the decoder of the autoencoder, the encoder weights of the autoencoder, the decoder weights of the autoencoder, the encoder structure of the autoencoder, the decoder structure of the autoencoder, the channel state information bitstream generation and the decoding scheme.

[0186] Clause 55: The apparatus of clause 43, wherein the processor is further configured to execute instructions stored in the memory to: At least one of an encoder structure of the encoder or a decoder structure of the encoder is transmitted to the user equipment.

[0187] Clause 56: A non-transitory computer-readable medium storing instructions executable by one or more processors of an apparatus for managing channel state information feedback compression to perform a method comprising: transmitting at least one reference signal; receiving at least one of a channel state information bitstream or at least one message in response to the transmitted at least one reference signal; determining at least one of a low dimensional channel condition or a value indicative of an error measure based on at least one of the channel state information bitstream or the at least one message; Calculating at least one of updated weights or updated parameters for a decoder of an autoencoder in a base station based on at least one of an estimated channel condition or a value indicative of an error measurement, wherein the estimated channel condition is based on the low-dimensional channel condition; and At least one of the weights or parameters of the decoder of the autoencoder in the base station is updated based on at least one of the calculated updated weights or updated parameters.

[0188] Clause 57: A method for managing compression of channel state information feedback, the method comprising: receiving at least one reference signal; estimating a channel condition based on the at least one reference signal; Using an encoder of a first autoencoder in a user equipment (UE), reducing the dimensionality of the estimated channel state so as to generate a low dimensional channel state; calculating an error measurement value based on a difference between a UE channel state input and a UE channel state output using an encoder of a second autoencoder and a decoder of the autoencoder in the UE; calculating at least one of updated weights or updated parameters of an encoder for the first autoencoder in the UE based on a value indicative of the error measure; Based on at least one of the calculated updated weights or updated parameters, updating at least one of the weights or parameters of the encoder of the first autoencoder in the UE; generating at least one of a channel state information bitstream or at least one message based on at least one of the low-dimensional channel condition or at least one of updated weights or updated parameters of an encoder of the first autoencoder in the UE; and At least one of the channel state information bitstream or at least one message is transmitted.

[0189] Clause 58. The method according to clause 57, further comprising: Based on the value indicating the error measurement value, at least one of the updated weights or updated parameters of the decoder of the autoencoder in the base station is calculated; and at least one of the updated weights or updated parameters of the decoder of the autoencoder in the base station is transmitted to the base station.

[0190] Clause 59: The method according to Clause 57 also includes: transmitting information related to at least one of the calculated weights, parameters or structures of the encoder of the first autoencoder in the UE, the encoder of the second autoencoder in the UE, the decoder of the autoencoder in the UE, the decoder of the autoencoder in the base station, or the encoder of the first autoencoder in the UE, the encoder of the second autoencoder in the UE, the decoder of the autoencoder in the UE, or the decoder of the autoencoder in the base station.

[0191] Clause 60: The method according to Clause 59 also includes: receiving a response including a confirmation, the confirmation including the calculated weights, parameters or structure of at least one of the encoder of the first autoencoder in the UE, the encoder of the second autoencoder in the UE, the decoder of the autoencoder in the UE or the decoder of the autoencoder in the base station; and based on the confirmation, updating the structure or at least one of the calculated weights or parameters of at least one of the encoder of the first autoencoder in the UE, the encoder of the second autoencoder in the UE or the decoder of the autoencoder in the UE.

[0192] Clause 61: A method of managing compression of channel state information feedback, the method comprising: transmitting at least one reference signal; receiving at least one of a channel state information bitstream or at least one message in response to the transmitted at least one reference signal; determining at least one of updated weights, updated parameters, or structures for at least one of a first autoencoder in a user equipment (UE), a second autoencoder in the UE, or a decoder of an autoencoder in the UE; Calculating at least one of updated weights or updated parameters for a decoder of an autoencoder in a base station based on at least one of updated weights or updated parameters for an encoder of the autoencoder in the UE; and At least one of the weights, parameters or structures of the decoder of the autoencoder in the base station is updated based on the calculated updated weights or at least one of the updated parameters or structures.

[0193] Clause 62: The method according to clause 61, further comprising: At least one of calculated weights, parameters, or updated structures for at least one of a decoder of the autoencoder in the base station, an encoder of the first autoencoder in a UE, an encoder of the second autoencoder in the UE, or a decoder of the autoencoder in the UE is received from the UE.

[0194] Clause 63: The method according to clause 62, wherein the updated structure is a sandwich structure.

[0195] Clause 64. The method according to clause 61 also includes: transmitting to the UE at least one of the calculated weights, parameters or structures for at least one of a decoder of an autoencoder in a base station, an encoder of a first autoencoder in a UE, an encoder of a second autoencoder in a UE or a decoder of an autoencoder in a UE. [List of abbreviations]

[0196] 3GPP (Third Generation Partnership Project) AE (Autoencoder) CSI (Channel State Information) DL (Downlink) gNB (Base Station or Next Generation Base Station) base station or next generation base station LTE (Long-Term Evolution) long-term evolution NR (New Radio) PDU (Protocol Data Unit) PSCCH (Physical Sidelink Control Channel) PSSCH (Physical Sidelink Shared Channel) RS (Reference Signal) TX / RX (Transmit / Receive) transmission / reception UE (User Equipment) UL (Uplink) Uu (UMTS Air Interface) UMTS air interface

Claims

1. A method for managing channel state information feedback compression, the method comprising: receiving at least one reference signal; estimating a channel condition based on the at least one reference signal; Using an encoder of a first autoencoder in a user equipment (UE), reducing the dimensionality of the estimated channel state so as to generate a low dimensional channel state; calculating an error measurement value based on a difference between a UE channel state input and a UE channel state output using an encoder of a second autoencoder and a decoder of the autoencoder in the UE; as well as Transmitting at least one of a channel state information bitstream or at least one message, wherein the at least one of the channel state information bitstream or at least one message is based on at least one of the low dimensional channel condition or a value indicative of the error measurement value.

2. The method according to claim 1, further comprising: At least one of the channel state information bitstream or at least one message is generated based on at least one of the low dimensional channel condition or a value indicative of the error measurement value.

3. The method according to claim 1, wherein: At least one of the channel state information bit stream or at least one message is transmitted to the base station.

4. The method according to claim 1, wherein: The encoder of the first autoencoder is based on a deep neural network.

5. The method according to claim 1, further comprising: The estimated channel conditions are input into an encoder of the first autoencoder, wherein the estimated channel conditions input into the encoder of the first autoencoder are high-dimensional channel conditions.

6. The method according to claim 1, wherein: Calculating the error measurement value based on the difference between the UE channel state input and the UE channel state output includes: using a loss function.

7. The method according to claim 1, further comprising: Based on the error measurement value, updated weights and parameters of the encoder of the first autoencoder and the decoder of the autoencoder are calculated.

8. The method according to claim 1, further comprising: At least one of an encoder weight of an autoencoder, an encoder parameter of an autoencoder, an encoder structure of an autoencoder, a decoder weight of an autoencoder, a decoder parameter of an autoencoder, or a decoder structure of an autoencoder is received.

9. A method for managing channel state information feedback compression, the method comprising: transmitting at least one reference signal; receiving at least one of a channel state information bitstream or at least one message in response to the transmitted at least one reference signal; determining at least one of a low dimensional channel condition or a value indicative of an error measure based on at least one of the channel state information bitstream or the at least one message; calculating at least one of updated weights or updated parameters for a decoder of an autoencoder in a base station based on at least one of an estimated channel condition or a value indicative of an error measurement, wherein the estimated channel condition is based on the low-dimensional channel condition; as well as At least one of the weights or parameters of the decoder of the autoencoder in the base station is updated based on at least one of the calculated updated weights or updated parameters.

10. The method according to claim 9, wherein: The estimated channel conditions are determined by decoding the low-dimensional channel conditions using a decoder of the autoencoder in the base station.

11. The method according to claim 9, wherein: The at least one reference signal is transmitted to a user equipment (UE), and at least one of the channel state information bitstream or the at least one message is received from the UE.

12. The method according to claim 9, wherein: Determining at least one of a low dimensional channel condition or a value indicative of an error measure comprises extracting at least one of a low dimensional channel condition or a value indicative of an error measure from at least one of the channel state information bitstream or at least one message.

13. The method according to claim 9, wherein: The decoder of the autoencoder in the base station is based on a deep neural network.

14. The method according to claim 9, further comprising: The low-dimensional channel conditions are converted into high-dimensional channel conditions using a decoder of the autoencoder.

15. The method according to claim 9, wherein: Calculating at least one of the updated weights or updated parameters of the decoder of the autoencoder in the base station includes: calculating at least one of the updated weights or updated parameters of the decoder of the autoencoder in the base station based on the value of the indicative error measurement value.

16. The method according to claim 9, wherein: Calculating at least one of the updated weights or updated parameters also includes: calculating at least one of the updated weights or updated parameters for each of an encoder of a first autoencoder in a user equipment (UE), an encoder of a second autoencoder in the UE, and a decoder of the autoencoder in the UE, and wherein the encoder of the second autoencoder in the UE and the decoder of the autoencoder in the UE belong to the same autoencoder pair.

17. A method for managing channel state information feedback compression, the method comprising: receiving at least one reference signal; estimating a channel condition based on the at least one reference signal; Using an encoder of a first autoencoder in a user equipment (UE), reducing the dimensionality of the estimated channel state so as to generate a low dimensional channel state; calculating an error measurement value based on a difference between a UE channel state input and a UE channel state output using an encoder of a second autoencoder and a decoder of the autoencoder in the UE; calculating at least one of updated weights or updated parameters of an encoder for the first autoencoder in the UE based on a value indicative of the error measure; Based on at least one of the calculated updated weights or updated parameters, updating at least one of the weights or parameters of the encoder of the first autoencoder in the UE; generating at least one of a channel state information bitstream or at least one message based on at least one of the low-dimensional channel condition or at least one of updated weights or updated parameters of an encoder of the first autoencoder in the UE; and At least one of the channel state information bitstream or at least one message is transmitted.

18. The method according to claim 17, further comprising: calculating at least one of updated weights or updated parameters for a decoder of an autoencoder in a base station based on a value indicative of the error measure; And transmitting at least one of the updated weights or updated parameters of the decoder of the autoencoder in the base station to the base station.

19. The method according to claim 17, further comprising: Transmitting information related to at least one of the calculated weights, parameters or structures of the encoder of the first autoencoder in the UE, the encoder of the second autoencoder in the UE, the decoder of the autoencoder in the UE, the decoder of the autoencoder in the base station, or the encoder of the first autoencoder in the UE, the encoder of the second autoencoder in the UE, the decoder of the autoencoder in the UE, or the decoder of the autoencoder in the base station.

20. A method for managing channel state information feedback compression, the method comprising: transmitting at least one reference signal; receiving at least one of a channel state information bitstream or at least one message in response to the transmitted at least one reference signal; determining at least one of updated weights, updated parameters, or structures for at least one of a first autoencoder in a user equipment (UE), a second autoencoder in the UE, or a decoder of an autoencoder in the UE; Calculating at least one of updated weights or updated parameters for a decoder of an autoencoder in a base station based on at least one of updated weights or updated parameters for an encoder of the autoencoder in the UE; as well as At least one of the weights, parameters or structures of the decoder of the autoencoder in the base station is updated based on the calculated updated weights or at least one of the updated parameters or structures.

Images