Method and device for sending and receiving phase shift configuration of intelligent reflection surface IRS

Abstract

A method and device for sending and receiving a phase shift configuration of an intelligent reflective surface can be applied to the technical field of communications, and the method executed by a network device comprises: receiving a pilot signal sent by a terminal device; performing decorrelation processing on the pilot signal; inputting the decorrelated pilot signal into a first network model to obtain an initial reflection phase shift vector; the initial reflection phase shift vector is input into a second network model, a reflection phase shift vector to be configured is obtained, and the scale of the reflection phase shift vector to be configured is smaller than that of the initial reflection phase shift vector; and sending the reflection phase shift vector to be configured to the IRS. Therefore, by reducing the scale of the to-be-configured reflection phase shift vector, the overhead of configuring the reflection phase shift vector to the IRS by the network equipment and the loss of rate performance are reduced.

Description

Method and device for transmitting and receiving phase shift configuration of intelligent reflecting surface (IRS) Technical Field

[0001] The present disclosure relates to the field of communication technology, and in particular to a method and device for transmitting and receiving a phase shift configuration of an intelligent reflecting surface (IRS). Background Art

[0002] In related technologies, intelligent reflecting surfaces (IRS) are deployed on the surfaces of various objects in a wireless transmission environment. Appropriate phase shifts are set according to the channel information of the system to achieve dynamic adjustment of the communication environment.

[0003] To implement beamforming in IRS-assisted communication systems, network devices must calculate and configure phase shifts for the IRS. Reducing the overhead of the IRS phase shift configuration process is a pressing issue.

[0004] Summary of the Invention

[0005] The embodiments of the present disclosure provide a method and apparatus for transmitting and receiving phase shift configuration of an intelligent reflecting surface (IRS). By reducing the size of the reflection phase shift vector to be configured, the overhead and rate performance loss of a network device in configuring the reflection phase shift vector to the IRS are reduced.

[0006] In a first aspect, an embodiment of the present disclosure provides a method for transmitting a phase shift configuration of an intelligent reflecting surface (IRS). The method, executed by a network device, includes: receiving a pilot signal transmitted by a terminal device; performing decorrelation processing on the pilot signal; inputting the decorrelated pilot signal into a first network model to obtain an initial reflection phase shift vector; inputting the initial reflection phase shift vector into a second network model to obtain a reflection phase shift vector to be configured, wherein the scale of the reflection phase shift vector to be configured is smaller than the scale of the initial reflection phase shift vector; and transmitting the reflection phase shift vector to be configured to the IRS.

[0007] In the present disclosure, after receiving a pilot signal from a terminal device, a network device first performs decorrelation processing on the pilot signal. The decorrelation-processed pilot signal is then input into a first network model to obtain an initial reflection phase shift vector. The initial reflection phase shift vector is then input into a second network model to obtain a smaller reflection phase shift vector to be configured. This reflection phase shift vector is then sent to the IRS. This reduces the size of the reflection phase shift vector to be configured, thereby reducing the overhead and rate performance loss of the network device in configuring the reflection phase shift vector to the IRS.

[0008] In a second aspect, embodiments of the present disclosure provide a method for receiving a phase shift configuration of an intelligent reflecting surface (IRS). The method, executed by the IRS, includes: receiving a reflection phase shift vector to be configured, sent by a network device; inputting the reflection phase shift vector to be configured into a phase shift conversion model to obtain a reflection phase shift vector to be applied, wherein the scale of the reflection phase shift vector to be applied is larger than the scale of the reflection phase shift vector to be configured; and determining a first reflection phase shift corresponding to each unit in the IRS based on the reflection phase shift vector to be applied.

[0009] In this disclosure, after receiving a reflection phase shift vector to be configured from a network device, the IRS uses a shift conversion model to convert it into a larger reflection phase shift vector to be applied. Based on this reflection phase shift vector, the IRS then determines the first reflection phase shift corresponding to each unit. This allows the network device to configure a smaller reflection phase shift vector for the IRS, reducing the overhead and performance loss associated with configuring the reflection phase shift vector.

[0010] In a third aspect, an embodiment of the present disclosure provides a network device, including:

[0011] A transceiver module, configured to receive a pilot signal sent by a terminal device;

[0012] a processing module, configured to perform decorrelation processing on the pilot signal;

[0013] The processing module is further configured to input the decorrelated pilot signal into the first network model to obtain an initial reflection phase shift vector;

[0014] The processing module is further configured to input the initial reflection phase shift vector into the second network model to obtain a reflection phase shift vector to be configured, wherein the scale of the reflection phase shift vector to be configured is smaller than the scale of the initial reflection phase shift vector;

[0015] The transceiver module is further configured to send the reflection phase shift vector to be configured to the IRS.

[0016] In a fourth aspect, an embodiment of the present disclosure provides an intelligent reflective surface (IRS), comprising:

[0017] A transceiver module, configured to receive a reflection phase shift vector to be configured sent by a network device;

[0018] a processing module, configured to input the reflection phase shift vector to be configured into a phase shift conversion model to obtain a reflection phase shift vector to be applied, wherein the scale of the reflection phase shift vector to be applied is larger than the scale of the reflection phase shift vector to be configured;

[0019] The processing module is further configured to determine a first reflection phase shift corresponding to each unit in the IRS based on the reflection phase shift vector to be applied.

[0020] In a fifth aspect, an embodiment of the present disclosure provides a communication device, which includes a processor. When the processor calls a computer program in a memory, the method described in the first aspect is executed.

[0021] In a sixth aspect, an embodiment of the present disclosure provides a communication device, which includes a processor. When the processor calls a computer program in a memory, the method described in the second aspect is executed.

[0022] In the seventh aspect, an embodiment of the present disclosure provides a communication device, which includes a processor and a memory, in which a computer program is stored; the processor executes the computer program stored in the memory so that the communication device executes the method described in the first aspect above.

[0023] In an eighth aspect, an embodiment of the present disclosure provides a communication device, which includes a processor and a memory, in which a computer program is stored; the processor executes the computer program stored in the memory so that the communication device executes the method described in the second aspect above.

[0024] In a ninth aspect, an embodiment of the present disclosure provides a communication device, which includes a processor and an interface circuit, wherein the interface circuit is used to receive code instructions and transmit them to the processor, and the processor is used to run the code instructions to enable the device to execute the method described in the first aspect above.

[0025] In the tenth aspect, an embodiment of the present disclosure provides a communication device, which includes a processor and an interface circuit, wherein the interface circuit is used to receive code instructions and transmit them to the processor, and the processor is used to run the code instructions to enable the device to execute the method described in the second aspect above.

[0026] In the eleventh aspect, an embodiment of the present disclosure provides a communication system, which includes the communication device described in the third aspect and the communication device described in the fourth aspect, or the system includes the communication device described in the fifth aspect and the communication device described in the sixth aspect, or the system includes the communication device described in the seventh aspect and the communication device described in the eighth aspect, or the system includes the communication device described in the ninth aspect and the communication device described in the tenth aspect.

[0027] In a twelfth aspect, an embodiment of the present invention provides a computer-readable storage medium for storing instructions for the above-mentioned terminal device, and when the instructions are executed, the terminal device executes the method described in the first aspect.

[0028] In a thirteenth aspect, an embodiment of the present invention provides a readable storage medium for storing instructions used by the above-mentioned network device, and when the instructions are executed, the network device executes the method described in the above-mentioned second aspect.

[0029] In a fourteenth aspect, the present disclosure further provides a computer program product comprising a computer program, which, when executed on a computer, enables the computer to execute the method described in the first aspect above.

[0030] In a fifteenth aspect, the present disclosure further provides a computer program product comprising a computer program, which, when executed on a computer, enables the computer to execute the method described in the second aspect above.

[0031] In a seventeenth aspect, the present disclosure provides a chip system comprising at least one processor and an interface for supporting a network device in implementing the functions described in the first aspect, such as determining or processing at least one of the data and information described in the aforementioned method. In one possible design, the chip system further comprises a memory for storing computer programs and data necessary for the terminal device. The chip system may consist of a chip alone or may include a chip and other discrete components.

[0032] In an eighteenth aspect, the present disclosure provides a chip system comprising at least one processor and an interface for supporting an IRS device in implementing the functions described in the second aspect, such as determining or processing at least one of the data and information described in the aforementioned method. In one possible design, the chip system further comprises a memory for storing computer programs and data necessary for the network device. The chip system may consist solely of a chip or may include a chip and other discrete components.

[0033] In a nineteenth aspect, the present disclosure provides a computer program, which, when executed on a computer, enables the computer to execute the method described in the first aspect above.

[0034] In a twentieth aspect, the present disclosure provides a computer program which, when executed on a computer, enables the computer to execute the method described in the second aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure or the background technology, the drawings required for use in the embodiments of the present disclosure or the background technology will be described below.

[0036] FIG1 is a schematic diagram of the architecture of a communication system provided by an embodiment of the present disclosure;

[0037] FIG2 is a flow chart of a method for transmitting a phase shift configuration of an intelligent reflecting surface (IRS) according to an embodiment of the present disclosure;

[0038] FIG3 is a flow chart of another method for transmitting a phase shift configuration of an intelligent reflective surface (IRS) according to an embodiment of the present disclosure;

[0039] FIG4 is a flow chart of another method for transmitting a phase shift configuration of an intelligent reflective surface (IRS) according to an embodiment of the present disclosure;

[0040] FIG5 is a flow chart of another method for transmitting a phase shift configuration of an intelligent reflective surface (IRS) according to an embodiment of the present disclosure;

[0041] FIG6 is a flow chart of another method for transmitting a phase shift configuration of an intelligent reflective surface (IRS) according to an embodiment of the present disclosure;

[0042] FIG7 is a flow chart of a training method for determining a model for a reflection phase shift to be configured, provided in an embodiment of the present disclosure;

[0043] FIG7a is a schematic diagram of the structure of the network model provided by the present disclosure;

[0044] FIG7 b is a schematic structural diagram of a first network model;

[0045] FIG7 c is a schematic structural diagram of a second network model;

[0046] FIG7 d is a schematic structural diagram of a phase shift conversion model;

[0047] FIG7e is a schematic diagram of the transmission time slot structure of the IRS-assisted communication system;

[0048] FIG8 is a flow chart of another method for receiving a phase shift configuration of an intelligent reflecting surface (IRS) provided in an embodiment of the present disclosure;

[0049] FIG9 is a flow chart of another method for receiving a phase shift configuration of an intelligent reflecting surface (IRS) according to an embodiment of the present disclosure;

[0050] FIG10 is an interactive schematic diagram of a receiving method of a phase shift configuration of an intelligent reflecting surface IRS provided by an embodiment of the present disclosure;

[0051] FIG11 is a schematic structural diagram of a communication device provided in an embodiment of the present disclosure;

[0052] FIG12 is a schematic structural diagram of another communication device provided in an embodiment of the present disclosure;

[0053] FIG13 is a schematic structural diagram of a chip provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0054] To facilitate understanding, the terms involved in this disclosure are first introduced.

[0055] 1. Intelligent reflecting surface (IRS)

[0056] Smart reflective surfaces, also known as "reconfigurable smart surfaces" or "smart metasurfaces," are known as IRS (Intelligent Reflection Surface) or RIS (Reconfigurable Intelligence Surface). While superficially unremarkable, IRSs can be flexibly deployed in wireless communication environments to manipulate the frequency, phase, and polarization characteristics of reflected or refracted electromagnetic waves, thereby reshaping wireless channels. Specifically, IRSs utilize precoding technology to reflect incident signals in a specific direction, thereby enhancing signal strength at the receiving end and achieving channel control.

[0057] 2. Pilot signal

[0058] A pilot signal, also known as a reference signal, is a known signal provided by the transmitter to the receiver for channel estimation or sounding. It can be used for coherent detection and demodulation in terminal devices, beamforming, coherent detection and monitoring of network equipment, and channel quality measurement.

[0059] In order to better understand the method for sending and receiving a phase shift configuration of an intelligent reflecting surface (IRS) disclosed in an embodiment of the present disclosure, the communication system to which the embodiment of the present disclosure is applicable is first described below.

[0060] Please refer to Figure 1, which is a schematic diagram of the architecture of a communication system provided by an embodiment of the present disclosure. The communication system may include, but is not limited to, a network device, an intelligent reflective surface, and a terminal device. The number and configuration of devices shown in Figure 1 are for example purposes only and do not limit the embodiments of the present disclosure. In actual applications, two or more network devices, two or more intelligent relays, and two or more terminal devices may be included. The communication system shown in Figure 1 includes, for example, a network device 11, a terminal device 12, and an intelligent reflective surface 13.

[0061] It should be noted that the technical solutions of the embodiments of the present disclosure can be applied to various communication systems, such as long-term evolution (LTE) systems, fifth-generation (5G) mobile communication systems, 5G new radio (NR) systems, or other future new mobile communication systems.

[0062] The network device 11 in the embodiment of the present disclosure is an entity on the network side for transmitting or receiving signals. For example, the network device 101 can be an evolved NodeB (eNB), a transmission reception point (TRP), a next generation NodeB (gNB) in an NR system, a base station in other future mobile communication systems, or an access node in a wireless fidelity (WiFi) system. The embodiment of the present disclosure does not limit the specific technology and specific device form adopted by the network device. The network device provided in the embodiment of the present disclosure can be composed of a centralized unit (CU) and a distributed unit (DU), wherein the CU can also be called a control unit. The CU-DU structure can be used to split the protocol layer of the network device, such as the base station, and the functions of some protocol layers are placed in the CU for centralized control, and the functions of the remaining part or all of the protocol layers are distributed in the DU, and the DU is centrally controlled by the CU.

[0063] The terminal device 12 in the embodiment of the present disclosure is an entity on the user side for receiving or transmitting signals, such as a mobile phone. The terminal device can also be called a terminal device (terminal), user equipment (UE), mobile station (MS), mobile terminal device (MT), etc. The terminal device can be a car with communication function, a smart car, a mobile phone, a wearable device, a tablet computer (Pad), a computer with wireless transceiver function, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal device in industrial control (industrial control), a wireless terminal device in self-driving, a wireless terminal device in remote medical surgery, a wireless terminal device in smart grid (smart grid), a wireless terminal device in transportation safety (transportation safety), a wireless terminal device in smart city (smart city), a wireless terminal device in smart home (smart home), etc. The embodiment of the present disclosure does not limit the specific technology and specific device form adopted by the terminal device.

[0064] In this system, the network device can implement the method shown in any embodiment of Figures 2 to 7 of the present disclosure, and the IRS can execute the method shown in any embodiment of Figures 8 to 10 of the present disclosure.

[0065] It can be understood that the communication system described in the embodiment of the present disclosure is for the purpose of more clearly illustrating the technical solution of the embodiment of the present disclosure, and does not constitute a limitation on the technical solution provided by the embodiment of the present disclosure. Ordinary technicians in this field can know that with the evolution of the system architecture and the emergence of new business scenarios, the technical solution provided by the embodiment of the present disclosure is also applicable to similar technical problems.

[0066] To implement beamforming in an IRS-assisted communication system, the network device needs to estimate the channel state information (CSI). Based on the CSI estimate, the network device calculates the appropriate beamforming matrix for transmitting the beam and the phase shift for reflecting the beam from the IRS. The calculated IRS phase shift is then assigned to the IRS.

[0067] Because network equipment design beamforming before IRS phase shift configuration occurs, the impact of IRS phase shift configuration overhead can cause spectrum efficiency loss. This disclosure proposes a method for sending and receiving phase shift configuration for an intelligent reflective surface (IRS). This method considers the IRS phase configuration overhead during the beamforming design process and adaptively adjusts the phase configuration overhead based on channel statistics, minimizing system and rate performance losses caused by the phase configuration overhead.

[0068] Please refer to Figure 2, which is a flow chart of a method for transmitting a phase shift configuration of an intelligent reflective surface (IRS) according to an embodiment of the present disclosure. The method is executed by a network device. As shown in Figure 2, the method may include but is not limited to the following steps:

[0069] Step 201: Receive a pilot signal sent by a terminal device.

[0070] During beamforming design, each terminal device must send a pilot signal to the network device, allowing the network device to determine its corresponding beamforming matrix and IRS phase shift based on the corresponding pilot signal. In other words, in this disclosure, the network device can receive pilot signals sent by one or more terminal devices and calculate the beamforming matrix and IRS phase shift corresponding to each terminal device.

[0071] Step 202: Perform decorrelation processing on the pilot signal.

[0072] Optionally, the network device may perform decorrelation processing on the received pilot signal by utilizing the characteristic that channels between terminal devices communicating with the network device are orthogonal to each other.

[0073] Step 203: Input the decorrelated pilot signal into the first network model to obtain an initial reflection phase shift vector.

[0074] The first network model is a model obtained by training the network device based on training data and used to generate a reflection phase shift vector of the IRS according to an input decorrelated pilot signal.

[0075] Optionally, when the network device performs decorrelation processing on the received pilot signal, it can perform decorrelation processing on the currently received pilot signal based on the training data corresponding to the first network model, assuming the number of terminal devices communicating with it at the same time when decorrelation processing on the pilot signal.

[0076] For example, if the network device performs three-directional decorrelation processing on the pilot signals sent by the terminal devices when preparing training data corresponding to the first network model, that is, the network device assumes that it can communicate with three terminal devices simultaneously through the IRS. Then, when performing beamforming design based on the first network model, the network device can also perform three-directional decorrelation processing on the pilot signals sent by the received terminal devices.

[0077] Referring to the above example, after the network device receives the pilot signal sent by the terminal device, it can divide it into multiple subframes, each of which has 3 pilot symbols. After the network device performs decorrelation processing on the received pilot signal, the obtained decorrelated pilot signal can be expressed as follows:

[0078] y k (t) = F k q(t)+n(t)

[0079] Among them, y k (t) represents the component of user k in the t-th subframe after decorrelation;

[0080] represents the real end-to-end channel of terminal device k, represents the direct link channel between the network device and the terminal device k, represents the reflection link cascade channel between the network device and the terminal device k, G represents the channel between the network device and the IRS, represents the channel between the IRS and the terminal device k, diag(·) represents the diagonal matrix with the input vector as the diagonal element;

[0081] q(t)=[1,v(t) T ] T , (·) T represents the transpose, v(t) represents the reflected phase shift vector of the IRS in the t-th subframe, and n(t) represents the noise component after decorrelation of the t-th subframe.

[0082] Optionally, after processing the de-processed pilot signal, the first network model can also output a beamforming matrix. Then, when the network device performs downlink data transmission, it can process the downlink data based on the beamforming matrix, obtain the processed downlink data, and send the processed downlink data to the terminal device.

[0083] Step 204: Input the initial reflection phase shift vector into the second network model to obtain a reflection phase shift vector to be configured. The scale of the reflection phase shift vector to be configured is smaller than the scale of the initial reflection phase shift vector.

[0084] The second network model is a model obtained by joint training when the network device trains the first network model and is used to compress the scale of the initial reflection phase shift vector output by the first network model.

[0085] Optionally, the second network model may be used to compress the scale of the initial reflection phase shift vector to obtain a smaller-scale reflection phase shift vector to be configured.

[0086] Step 205: Send the reflection phase shift vector to be configured to the IRS.

[0087] In the present disclosure, after determining the reflection phase shift vector to be configured, the network device may send it to the IRS, so that the IRS may reflect and transmit the signal based on the reflection phase shift vector to be configured.

[0088] After processing the decorrelated pilot signal using the first network model to obtain an initial reflection phase shift vector, the network device further processes the initial network reflection phase shift vector using the second network model to obtain a smaller reflection phase shift vector to be configured, which it then sends to the IRS. The smaller size of the reflection phase shift vector to be configured reduces the overhead and rate performance loss of configuring the reflection phase shift vector with the IRS.

[0089] In the present disclosure, after receiving a pilot signal from a terminal device, a network device first performs decorrelation processing on the pilot signal. The decorrelation-processed pilot signal is then input into a first network model to obtain an initial reflection phase shift vector. The initial reflection phase shift vector is then input into a second network model to obtain a smaller reflection phase shift vector to be configured. This reflection phase shift vector is then sent to the IRS. This reduces the size of the reflection phase shift vector to be configured, thereby reducing the overhead and rate performance loss of the network device in configuring the reflection phase shift vector to the IRS.

[0090] Please refer to Figure 3, which is a flow chart of another method for transmitting a phase shift configuration of an intelligent reflective surface (IRS) according to an embodiment of the present disclosure. The method is executed by a network device. As shown in Figure 3, the method may include but is not limited to the following steps:

[0091] Step 301: Receive a pilot signal sent by a terminal device.

[0092] Step 302: Perform decorrelation processing on the pilot signal.

[0093] Step 303: Input the decorrelated pilot signal into the first network model to obtain an initial reflection phase shift vector.

[0094] The specific implementation process of the above steps 301 to 303 can refer to the detailed description of any embodiment of the present disclosure and will not be repeated here.

[0095] Step 304: Input the initial reflection phase shift vector into the compression sub-network in the second network model to obtain a compressed reflection phase shift vector.

[0096] Optionally, the compression sub-network can be any network that can compress vectors.

[0097] For example, the compression subnetwork can be composed of a cascade of convolutional layers, straightening layers, and fully connected layers, where the number of convolutional layers, straightening layers, and fully connected layers can be adjusted as needed.

[0098] Optionally, the number of neurons in the fully connected layer can be set based on the desired number of output elements after compression. For example, if the size of the initial reflection phase shift vector output by the first network model is 100×2×1, and the number of neurons in the fully connected layer is set to 50, the size of the compressed reflection phase shift vector obtained after passing through the compression subnetwork is reduced by half.

[0099] Step 305: Input the compressed reflection phase shift vector into the quantization subnetwork in the second network model to obtain the reflection phase shift vector to be configured.

[0100] To further reduce the amount of information fed back to the IRS by the network device, the second network model may further include a quantization subnetwork for quantizing the compressed phase shift vector to convert the floating-point vector into a vector in bit form. That is, the resulting reflection phase shift vector to be configured is a vector of bit tree values, thereby further reducing the configuration loss of the reflection phase shift vector.

[0101] Optionally, the quantization function can be selected as needed, for example, a quantization subnetwork designed using the Tanh function can be used. The quantization function is as follows:

[0102] Q1(v q )=tanh(c1vq )

[0103] Quantizer Q1, used to quantize vector v q Each element in performs a quantization operation based on the Tanh function. c1 is the quantization parameter. The larger its value is, the closer Q1 is to the real quantizer. However, if c1 is too large, it will affect the model effect.

[0104] Step 306: Send the reflection phase shift vector to be configured to the IRS.

[0105] The specific implementation process of the above step 306 can be described in detail with reference to any embodiment of the present disclosure, and will not be repeated here.

[0106] In the present disclosure, after receiving a pilot signal from a terminal device, a network device first performs decorrelation processing on the pilot signal. The decorrelation-processed pilot signal is then input into a first network model to obtain an initial reflection phase shift vector. The initial reflection phase shift vector is then compressed and quantized to obtain a smaller reflection phase shift vector to be configured. This reflection phase shift vector is then sent to the IRS. By compressing the size of the reflection phase shift vector to be configured, the network device reduces the overhead and rate performance loss of configuring the reflection phase shift vector to the IRS.

[0107] Please refer to Figure 4, which is a flow chart of another method for transmitting the phase shift configuration of an intelligent reflective surface (IRS) according to an embodiment of the present disclosure. The method is executed by a network device. As shown in Figure 4, the method may include but is not limited to the following steps:

[0108] Step 401: Receive a pilot signal sent by a terminal device.

[0109] Step 402: Perform decorrelation processing on the pilot signal.

[0110] Step 403: Input the decorrelated pilot signal into the first network model to obtain an initial reflection phase shift vector.

[0111] Step 404: Input the initial reflection phase shift vector into the compression sub-network in the second network model to obtain a compressed reflection phase shift vector.

[0112] Step 405: Input the compressed reflected phase shift vector into the quantizer in the quantization subnetwork to obtain a quantized phase shift vector.

[0113] The specific implementation process of the above steps 401 to 405 can refer to the detailed description of any embodiment of the present disclosure and will not be repeated here.

[0114] Step 406: Input the compressed reflected phase shift vector into the strategy sub-network in the quantization sub-network to obtain a mask vector.

[0115] Optionally, the corresponding reflection phase shift amount of the IRS may vary depending on the channel state. For example, when the channel state is good, only a small number of elements in the IRS need to process the data to be reflected or transmitted based on the corresponding phase shift angle; when the channel state is poor, all elements in the IRS may need to process the data to be reflected or transmitted based on the corresponding phase shift angle. The pilot signal transmitted by the terminal device can reflect the channel state. Therefore, in the present disclosure, the compressed reflection phase shift vector can also be input into the strategy sub-network. The strategy sub-network determines the current channel state based on the compressed reflection phase shift vector, and based on the current channel state, determines the scale of the reflection phase shift vector to be configured, thereby outputting the corresponding mask vector.

[0116] In the present disclosure, the policy sub-network can be implemented using any form of network structure, and the present disclosure does not limit this.

[0117] (21) For example, the strategy sub-network may include a fully connected layer, a Softmax layer, a Gumbel Softmax layer, and a layer for converting one-hot encoding to a mask operation. The number of neurons in the fully connected layer of the strategy sub-network needs to be the same as the number of neurons in the last fully connected layer of the compression sub-network. The input of the strategy sub-network is the compressed reflected phase shift vector in step 404. The output of the Softmax layer is represented by p, and the output of the Gumbel Softmax layer is represented by Then the output of the Gumbel Softmax layer is:

[0118]

[0119] Among them, g i Obey the standard Gumbel distribution, g i =-ln(-lnu i ),u i Obey the uniform distribution between (0,1), τ is a control Discreteness parameter. When τ tends to infinity, tends to obey uniform distribution; when τ tends to 0, Tends to be a one-hot code, so in this embodiment, τ can take a number close to 0, such as 0.005. Convert to mask m. For example,

[0120] Step 407: Fusing the mask vector with the quantized phase shift vector to obtain a reflection phase shift vector to be configured.

[0121] Optionally, the network device may perform a Hadmard product operation on the mask vector and the quantized phase shift vector to obtain the reflection phase shift vector to be configured.

[0122] Step 408: Send the reflection phase shift vector to be configured to the IRS.

[0123] The specific implementation process of the above step 408 can be described in detail with reference to any embodiment of the present disclosure, and will not be repeated here.

[0124] In the present disclosure, after receiving a pilot signal sent by a terminal device, a network device first performs decorrelation processing on the pilot signal. The decorrelation-processed pilot signal is then input into a first network model to obtain an initial reflection phase shift vector. The initial reflection phase shift vector is then compressed and quantized. A policy subnetwork is then used to determine a mask vector based on the current channel state. Finally, based on the mask vector and the quantized vector, a smaller reflection phase shift vector suitable for the current channel state is obtained and sent to the IRS. Thus, by adaptively compressing the size of the reflection phase shift vector to be configured, the configured reflection phase shift vector is ensured to be suitable for the current channel state, while reducing the overhead and rate performance loss of the network device configuring the reflection phase shift vector to the IRS.

[0125] Please refer to Figure 5, which is a flowchart of another method for transmitting the phase shift configuration of an intelligent reflective surface (IRS) provided by an embodiment of the present disclosure. The method is executed by a network device. As shown in Figure 5, the method may include but is not limited to the following steps:

[0126] Step 501: Receive a pilot signal sent by a terminal device.

[0127] The specific implementation process of the above step 501 can refer to the detailed description of any embodiment of the present disclosure and will not be repeated here.

[0128] Step 502: Determine current pilot information based on the pilot signal sent by the terminal device.

[0129] The pilot information includes the number of subframes included in the pilot signal and the number of pilot symbols included in each subframe.

[0130] For example, if the first network model and the second network model are trained and generated based on the assumption that three terminal devices communicate with the network device at the same time, then after receiving the pilot signal sent by the terminal device, the network device can decode the pilot signal according to the three pilot symbols contained in each subframe to determine the number of subframes contained in the pilot signal.

[0131] Step 503: Send the current pilot information to the IRS.

[0132] In the present disclosure, in order to assist the IRS in determining the applicable reflection phase shift during the uplink transmission process, the network device, after receiving the pilot signal sent by the terminal device, immediately determines the pilot information based on the pilot signal and sends the pilot information to the IRS, so that the IRS configures the uplink reflection phase based on the pilot information.

[0133] Optionally, to further reduce transmission overhead between the network device and the IRS, the network device may send current pilot information to the IRS only when the current pilot information differs from historical pilot information. Thus, the IRS configures the uplink reflection phase shift based on the new pilot information only upon receipt of the new pilot information. Otherwise, it continues to reflect and transmit uplink data based on the previous uplink reflection phase shift.

[0134] Step 504: perform decorrelation processing on the pilot signal.

[0135] Step 505: Input the decorrelated pilot signal into the first network model to obtain an initial reflection phase shift vector.

[0136] Step 506: Input the initial reflection phase shift vector into the second network model to obtain the reflection phase shift vector to be configured.

[0137] It should be noted that steps 502 and 503 can be performed synchronously with steps 504 to 506. That is, the network device can synchronously perform decorrelation processing on the pilot signal when determining and sending the pilot information to the IRS. This disclosure does not limit this.

[0138] Step 507: Send the reflection phase shift vector to be configured to the IRS.

[0139] The specific implementation process of the above steps 504 to 507 can refer to the detailed description of any embodiment of the present disclosure and will not be repeated here.

[0140] In the present disclosure, after receiving a pilot signal from a terminal device, a network device determines pilot information based on the pilot signal and sends the pilot information to an IRS. Simultaneously, the pilot signal is decorrelated and then input into a first network model to obtain an initial reflection phase shift vector. This initial reflection phase shift vector is then input into a second network model to obtain a smaller reflection phase shift vector to be configured, which is then sent to the IRS. This reduces the size of the reflection phase shift vector to be configured, thereby reducing the overhead and rate performance loss of the network device in configuring the reflection phase shift vector to the IRS.

[0141] Please refer to Figure 6, which is a flowchart of another method for transmitting the phase shift configuration of an intelligent reflective surface (IRS) provided by an embodiment of the present disclosure. The method is executed by a network device. As shown in Figure 6, the method may include but is not limited to the following steps:

[0142] Step 601: Receive a pilot signal sent by a terminal device.

[0143] The specific implementation process of the above step 601 can refer to the detailed description of any embodiment of the present disclosure and will not be repeated here.

[0144] Step 602: Determine current channel statistics.

[0145] Optionally, since the pilot signal sent by the terminal device can reflect the current channel state, the network device can determine the current channel statistical characteristics by analyzing the pilot signal sent by the terminal device.

[0146] Step 603: Acquire a first network model and a second network model associated with the current channel statistical characteristics from a preset model library.

[0147] Because different pilot signals in the training data correspond to different channel statistical characteristics, the model parameters of the first and second network models generated through training may also differ. Therefore, in the present disclosure, the network device associates the channel statistical characteristics corresponding to the pilot signals in the training data with the first and second network models generated through training. This allows the network device to obtain the corresponding usable first and second network models based on the current channel statistical characteristics during the reflection phase shift calculation process.

[0148] Step 604: perform decorrelation processing on the pilot signal.

[0149] Among them, step 604 can be executed simultaneously with step 603, or step 604 can be executed first and then step 603, which is not limited in this disclosure.

[0150] Step 605: Input the decorrelated pilot signal into the first network model to obtain an initial reflection phase shift vector.

[0151] Step 606: Input the initial reflection phase shift vector into the second network model to obtain the reflection phase shift vector to be configured.

[0152] Step 607: Send the reflection phase shift vector to be configured to the IRS.

[0153] The specific implementation process of the above steps 604 to 607 can refer to the detailed description of any embodiment of the present disclosure and will not be repeated here.

[0154] Step 608: Send the identifier of the first network model and / or the identifier of the second network model to the IRS.

[0155] Among them, step 608 can be executed before step 607 and after step 603, or can be executed synchronously with step 607, which is not limited in this disclosure.

[0156] Because the reflection phase shift vector to be configured sent by the network device to the IRS is a compressed phase shift vector, the IRS must first decompress the received reflection phase shift vector before configuring the reflection phase shift. In other words, the IRS needs to know how the network device compresses the reflection phase shift vector to be configured. Therefore, the network device can send the IRS the identifier of the currently used first network model and / or the identifier of the second network model. Based on the identifier of the first network model and / or the identifier of the second network model, the IRS can then determine the processing performed by the network device on the reflection phase shift vector to be configured.

[0157] Optionally, to minimize transmission between the network device and the IRS, the network device may send the identifier of the first network model and / or the identifier of the second network model to the IRS only when the current channel statistics differ from the historical channel statistics. Thus, if the IRS does not receive the identifier of the first network model and / or the identifier of the second network model, it may assume that the first network model and the second network model currently used by the network device are the same as those used historically. Consequently, the IRS may use the same phase shift conversion model as used historically to convert the received reflection phase shift vector to be configured.

[0158] In the present disclosure, after receiving a pilot signal sent by a terminal device, a network device determines the current channel statistics based on the pilot signal and, based on the current channel statistics, determines the first and second network models to be used. The network device also performs decorrelation processing on the pilot signal. The decorrelation-processed pilot signal is then input into the first network model to obtain an initial reflection phase shift vector. The initial reflection phase shift vector is then input into the second network model to obtain a smaller reflection phase shift vector to be configured. The smaller reflection phase shift vector is then sent to the IRS. This reduces the size of the reflection phase shift vector to be configured, thereby reducing the overhead and rate performance loss of the network device in configuring the reflection phase shift vector to the IRS.

[0159] Please refer to Figure 7, which is a flowchart of a training method for determining a model of a reflection phase shift to be configured, provided by an embodiment of the present disclosure. The method is executed by a network device. As shown in Figure 7, the method may include but is not limited to the following steps:

[0160] Step 701: Obtain a training data set.

[0161] Among them, the training data set includes the decorrelated pilot signals corresponding to multiple terminal devices, the first matrix corresponding to the direct link channel between each terminal device and the network device, and the second matrix corresponding to the reflected link cascade channel between the network device and the terminal device.

[0162] Optionally, the network device may first determine the possible scenarios in which it communicates with the terminal device via the IRS. For example, assume that it can communicate with three terminal devices with the assistance of the IRS. The total time slot of the pilot signal sent by the terminal device to the network device is L. The network device can then divide it into x subframes, with each subframe having 3 pilot symbols. The pilot symbols of all terminal devices are designed to be orthogonal to each other. Utilizing the orthogonality of the pilot sequences between terminal devices, the received pilot signal is decorrelated to obtain:

[0163] y k (t) = F k q(t)+n(t)

[0164] Among them, y k (t) represents the component of terminal device k in the t-th subframe after decorrelation, represents the real end-to-end channel of terminal device k, represents the first matrix corresponding to the direct link channel between the network device and the terminal device k, represents the second matrix corresponding to the reflection link cascade channel between the network device and the terminal device k, G represents the channel between the network device and the IRS, represents the channel between the IRS and the terminal device k, diag(·) represents the diagonal matrix with the input vector as the diagonal element, q(t)=[1,v(t) T ] T , (·) T represents the transpose, v(t) represents the reflected phase shift vector of the IRS in the t-th subframe, and n(t) represents the noise component after decorrelation of the t-th subframe.

[0165] Since the pilot signal of the terminal device is divided into x subframes, the training data set determined by the pilot signal sent by each terminal device can be expressed as follows:

[0166] Y k =F k Q+N

[0167] where Y k =[y k (1),...,y k (x)], Q = [q(1), ..., q(x)], N = [n(1), ..., n(x)], where Y kThis is the training data set required to train the neural network. Considering that there are three terminal devices in total, the input of the neural network should be Y = [Y1, Y2, Y3].

[0168] Step 702: Input the training data set into the network model to be trained, and obtain the reference beamforming matrix output by the first initial network model and the reference reflection phase shift vector output by the initial phase shift conversion model in the network model to be trained.

[0169] The network to be trained includes a cascaded first initial network model, a second initial network model, and an initial phase shift conversion model.

[0170] The schematic diagram of the network model provided by the present disclosure may be shown in FIG7 a . As shown in FIG7 a , the network model includes a first initial network model 71 , a second initial network model 72 and an initial phase shift conversion model 73 .

[0171] The input of the first initial network model 71 is the training data set Y, and the output is the beamforming matrix And the initial reflection phase shift vector V. The initial reflection phase shift vector V is input to the second initial network model 72, and the second initial network model 72 outputs the reflection phase shift vector to be configured After being input into the initial phase shift conversion model 73, the initial phase shift conversion model 73 can output the reference reflection phase shift vector vector applicable to the IRS.

[0172] The first initial network model 71, the second initial network model 72, and the initial phase-shift conversion model 73 can all be implemented using any network structure. The functions of the first initial network model 71, the second initial network model 72, and the initial phase-shift conversion model 73 are further described below using the network structure shown in FIG. 7b as an example.

[0173] FIG7 b is a schematic structural diagram of the first network model 71 .

[0174] As shown in Figure 7b, the first initial network model 71 can be composed of three groups of fully connected layers 711, activation function layers 712, and batch normalization layers 713. Each fully connected layer 711 can include 300 neurons as a hidden layer, and the activation function can be ReLu. In addition, due to the limitation of downlink transmission power, the beamforming matrix W can be normalized to where ||·|| F It means to find the Frobenius norm of the matrix, and P is the downlink transmit power.

[0175] As shown in FIG. 7 c , the second initial network model 72 may include a compression sub-network 721 , a quantization sub-network 722 , and a strategy sub-network 723 .

[0176] The compression sub-network 721 can be composed of a convolution layer, a straightening layer, and a fully connected layer. The form of the convolution kernel used in the convolution layer can be designed as needed. For example, three convolution kernels of size 2×2 are selected; the number of neurons in the fully connected layer is the number of output elements after compression. For example, in this embodiment, the number of neurons in the fully connected layer is set to 50. The reflected phase shift vector after compression of the output of the fully connected layer is v q In order to further reduce the amount of information fed back to the IRS by the network device while ensuring the sum rate performance, a quantization subnetwork 722 designed using the Tanh function can be added after the fully connected layer. For example, the quantization function can be as follows:

[0177] Q1(v q )=tanh(c1v q )

[0178] The quantization sub-network 722 performs the vector v q Each element in performs a quantization operation based on the Tanh function, and the output is a quantized vector The larger c1 is, the closer Q1 is to the real quantizer. However, if c1 is too large, it will affect the training effect. In this disclosure, c1 can be set to 10.

[0179] In addition, the strategy sub-network 723 can be composed of a fully connected layer (for example, the number of neurons is 50), a Softmax layer, a Gumbel Softmax layer, and a layer for converting one-hot code to mask operation. The input of the strategy sub-network 723 is the compressed reflected phase shift vector v output by the compression sub-network 721. q , the output of the Softmax layer is represented by p, and the output of the Gumbel Softmax layer is represented by Then the output of the Gumbel Softmax layer is:

[0180]

[0181] where g i Obey the standard Gumbel distribution, g i =-ln(-lnu i ),u i Obey the uniform distribution between (0,1), τ is a control Discreteness parameter. When τ tends to infinity, tends to obey uniform distribution; when τ tends to 0, Tends to be a one-hot code, so in this embodiment, τ takes a number close to 0, 0.005. Convert to mask vector m:

[0182] after, Perform the Hadmard product with the mask m to obtain the reflection phase shift vector to be configured Right now

[0183] As shown in FIG7d , the phase shift conversion model 73 may be composed of a fully connected layer 731 , a convolutional layer 732 , a straightening layer 733 and another fully connected layer 734 , 734 being used to output the reflection phase shift vector to be applied The number of neurons in the two fully connected layers can be the same or different. For example, both fully connected layers contain 200 neurons, and the convolutional layer uses three convolution kernels of size 2×2.

[0184] Optionally, if the IRS supports low-precision phase configuration, a quantization subnetwork Q2 (not shown in the figure) can be added after the above network. The input of the quantizer is the real and imaginary parts of the decompressed IRS reflection phase shift. For the IRS reflection phase shift v=e jθ , find the phase angle θ∈[0,2π), then the quantized phase angle is:

[0185]

[0186] Where B = 2 b , b is the number of quantization bits; The larger c2 is, the closer Q2 is to the real quantizer. However, if c2 is too large, it will affect the training effect. In this disclosure, for example, c2 = 10 can be selected; b i is the quantization threshold. Quantization subnetwork Q2 performs a tanh-based quantization operation on each element of vector θ. Next, the trigonometric functions cos(·) and sin(·) are used to convert the quantized IRS phase shift Q2(θ) into its real and imaginary parts. The trigonometric operations here are also performed on each element of Q2(θ).

[0187] In addition, the outputs of the last layer of the fully connected neural network of the above initial phase shift conversion model 73 are the real and imaginary parts of the IRS reflection phase shift. In order to reduce the scale of the neural network at the IRS, the present disclosure can also further simplify the IRS reflection phase shift. For example, the real and imaginary parts of the reflection phase shift angle output by the output neural network are converted into the IRS reflection phase shift angle value output by the output neural network, so that the number of neurons in the second fully connected layer can be adjusted from 200 to 100. Taking into account the periodicity of trigonometric functions, the present disclosure can first use the trigonometric functions cos(·) and sin(·) to convert them into real and imaginary parts, and then use the real and imaginary parts to calculate the angle value within the range [0,2π).

[0188] Step 703 : Determine a loss value according to the reference reflection phase shift vector, the reference beamforming matrix, the first matrix, and the second matrix.

[0189] In the present disclosure, the Adam optimizer and end-to-end training method can be used to minimize the cost function to train the model to be trained. Consider the data transmission phase of the downlink, including the phase in which the network device sends feedback data to the IRS and the phase in which the network device transmits data to the terminal device.

[0190] As shown in Figure 7e, it is a schematic diagram of the transmission time slot structure of the IRS-assisted communication system. The duration of a communication process between the network device and the terminal device is t0, which includes the pilot transmission phase, the phase in which the network device sends feedback data to the IRS, and the phase in which the network device transmits data to the terminal device. The total duration of the latter two phases is represented by T, and the duration of the network device sending feedback data to the IRS is represented by T. F Indicates that the amount of feedback data is The expected optimized system rate can be expressed as:

[0191] Among them, R F Represents the data transmission rate of the feedback link, R k It represents the rate of terminal device k. The specific expression is:

[0192]

[0193] in:

[0194]

[0195] Here |·| means modulo, and Respectively represent the real and imaginary parts, represents the normalized reference beamforming matrix of the ith terminal device, represents the reference reflection phase shift vector, Represents the downlink noise power.

[0196] Then the loss function that can be used in this disclosure is:

[0197]

[0198] Where E[·] represents the expectation.

[0199] After determining the reference reflection phase shift vector, the reference beamforming matrix, the first matrix and the second matrix, the network device calculates the value of the upper loss function.

[0200] Step 704 : Modify the initial network model based on the loss value until the loss value determined based on the output reflection phase shift vector and the beamforming matrix of the modified network model is less than a threshold.

[0201] The modified network model includes the first network model, the second network model and the phase shift conversion model in cascade.

[0202] In the present disclosure, after calculating the loss value, the network device can determine the correction gradient of each layer in the network based on the loss value, and then correct the weights and biases of each layer in the initial phase shift conversion model, the second initial network model and the first initial network model, until the loss value determined based on the reflected phase shift vector and beamforming matrix output by the corrected network model is less than the threshold, and the model training can be determined to be completed.

[0203] Optionally, the network device in the present disclosure may further send the phase shift conversion network model and the associated model identifier to the IRS, wherein the associated model identifier includes the identifier of the first network model and / or the identifier of the second network model.

[0204] In the present disclosure, since the phase shift conversion model is the model used by the IRS to decompress and perform other processing on the reflection phase shift to be configured by the received network device, the network device needs to send the trained phase shift conversion model to the IRS after model training is completed. In addition, since the phase shift conversion model is jointly trained with the first network model and the second network model, the network device also needs to send the identifier of the first network model and / or the identifier of the second network model to the IRS. This allows the IRS to store the phase shift conversion network model and the associated model identifier. During the reflection phase shift configuration process, the IRS selects the phase shift conversion model to be used based on the obtained reflection phase shift vector to be configured and the identifier of the first network model and / or the identifier of the second network model used when generating it. The selected phase shift conversion model is then used to configure the received reflection phase shift vector to be configured.

[0205] Optionally, the model training process may also be performed by a server. After the server generates a revised network model by executing the above process, the server may send the first network model and the second network model in the generated network model to the network device, and send the phase-shifted network model and the associated model identifier to the IRS. This disclosure is not limited to this.

[0206] Please refer to Figure 8, which is a flowchart of another method for receiving the phase shift configuration of an intelligent reflective surface (IRS) provided by an embodiment of the present disclosure. The method is performed by the IRS. As shown in Figure 8, the method may include but is not limited to the following steps:

[0207] Step 801: Receive a reflection phase shift vector to be configured sent by a network device.

[0208] In this disclosure, the reflection phase shift vector to be configured received by the IRS is obtained by processing the pilot signal sent by the terminal device using the trained first and second network models. The specific process for determining the reflection phase shift vector to be configured can be found in the detailed description of any embodiment of this disclosure and is not further elaborated here.

[0209] Step 802: Input the reflection phase shift vector to be configured into the phase shift conversion model to obtain the reflection phase shift vector to be applied.

[0210] The scale of the reflection phase shift vector to be applied is larger than the reflection phase shift vector to be configured.

[0211] Because the reflection phase shift vector to be configured sent by the network device is a compressed vector, the IRS needs to decompress and process the reflection phase shift vector before it can be applied. Therefore, the IRS can input the reflection phase shift vector to be configured into the phase shift conversion model to decompress and process the reflection phase shift vector to obtain the reflection phase shift vector to be applied.

[0212] Optionally, the phase shift conversion model can be implemented using a network of any structure, for example, a network structure as shown in FIG7 b-c . This disclosure does not limit this.

[0213] In addition, the phase shift conversion model can be generated by network equipment training and then configured to the IRS, or can be generated by server training and then configured in the IRS, which is not limited in this disclosure.

[0214] Step 803: Determine a first reflection phase shift corresponding to each unit in the IRS based on the reflection phase shift vector to be applied.

[0215] After determining the reflection phase shift vector to be applied, the IRS can determine the reflection phase shift corresponding to each IRS unit according to the arrangement information of the units in the IRS, and then reflect or transmit the downlink data based on the first reflection phase shift corresponding to each unit in the downlink transmission.

[0216] In this disclosure, after receiving a reflection phase shift vector to be configured from a network device, the IRS uses a shift conversion model to convert it into a larger reflection phase shift vector to be applied. Based on this reflection phase shift vector, the IRS then determines the first reflection phase shift corresponding to each unit. This allows the network device to configure a smaller reflection phase shift vector for the IRS, reducing the overhead and performance loss associated with configuring the reflection phase shift vector.

[0217] Please refer to Figure 9, which is a flowchart of another method for receiving the phase shift configuration of an intelligent reflective surface (IRS) provided by an embodiment of the present disclosure. The method is performed by the IRS. As shown in Figure 9, the method may include but is not limited to the following steps:

[0218] Step 901: Receive a reflection phase shift vector to be configured sent by a network device.

[0219] Step 902 , determining whether the IRS supports the first precision phase configuration, if so, executing step 903 , otherwise, executing step 904 .

[0220] The first precision is higher than the second precision.

[0221] The phase accuracy supported by the IRS is used to indicate the configurable phase accuracy of the IRS unit, for example, supporting 4 decimal places or only supporting integer places.

[0222] Step 903: Input the reflection phase shift vector to be configured into the first decompression model to obtain the reflection phase shift vector to be applied.

[0223] Step 904: input the reflection phase shift vector to be configured into the cascaded second decompression model and quantization model to obtain the reflection phase shift vector to be applied.

[0224] That is, if the IRS supports higher phase accuracy, the decompressed vector of the received reflection phase shift vector to be configured can be directly determined as the reflection phase shift vector to be applied. However, if the RS supports lower phase accuracy, the decompressed vector of the received reflection phase shift vector to be configured needs to be quantized to obtain the reflection phase shift vector to be applied.

[0225] Among them, the structure and implementation principle of the first decompression model, the second decompression model and the quantization model can refer to the detailed description of any embodiment of the present disclosure and will not be repeated here.

[0226] Optionally, in order to further reduce the scale of the phase shift conversion model on the IRS side, the last fully connected layer of the second decompression model in the phase shift conversion model may not output the imaginary and real parts corresponding to each phase shift angle, but output the specific phase shift angle, thereby halving the scale of the last fully connected layer.

[0227] That is, in the present disclosure, the IRS can input the reflection phase shift vector to be configured into the second decompression model, obtain the phase shift angle value corresponding to each element in the reflection phase shift vector to be configured output by the second decompression model; then input the phase shift angle value into the quantization model to obtain the quantized phase shift angle; and then determine the reflection phase shift vector to be applied based on the quantized phase shift angle.

[0228] That is, the IRS first determines the specific reflection phase shift angle output based on the real and imaginary parts of the reflection phase shift angle, then quantizes the reflection phase shift angle, and then uses trigonometric function calculations to determine the real and imaginary parts of the reflection phase shift angle to be applied based on the quantized reflection phase shift angle.

[0229] Step 905 : Determine a first reflection phase shift corresponding to each unit in the IRS based on the reflection phase shift vector to be applied.

[0230] In this disclosure, after receiving a reflection phase shift vector to be configured from a network device, the IRS can convert it into a larger reflection phase shift vector to be applied based on the phase shift accuracy supported by the IRS using a corresponding shift conversion model. The IRS then determines the corresponding reflection phase shift for each unit based on this reflection phase shift vector. This allows the network device to configure a smaller reflection phase shift vector for the IRS, reducing the overhead and performance loss associated with configuring the reflection phase shift vector.

[0231] Please refer to Figure 10, which is a flow chart of another method for receiving the phase shift configuration of an intelligent reflective surface (IRS) provided by an embodiment of the present disclosure. The method is performed by the IRS. As shown in Figure 10, the method may include but is not limited to the following steps:

[0232] Step 1001: Receive pilot information sent by a network device.

[0233] The pilot information includes the number of subframes contained in the pilot signal and the number of pilot symbols contained in each subframe, which is determined by the network device based on the pilot signal sent by the terminal device.

[0234] For example, if the phase shift conversion model is generated based on the assumption that three terminal devices communicate with the network device at the same time, then after receiving the pilot signal sent by the terminal device, the network device can deprocess the pilot signal according to the principle that each subframe contains three pilot symbols to determine the number of subframes contained in the pilot signal.

[0235] Step 1002: Determine the second reflection phase shift corresponding to each unit in the IRS during uplink transmission according to the pilot information.

[0236] In the present disclosure, in order to assist the IRS in determining the applicable reflection phase shift during the uplink transmission process, the network device, after receiving the pilot signal sent by the terminal device, immediately determines the pilot information based on the pilot signal and sends the pilot information to the IRS, so that the IRS configures the uplink reflection phase based on the pilot information.

[0237] Optionally, to further reduce transmission overhead between the network device and the IRS, the network device may send current pilot information to the IRS only when the current pilot information differs from historical pilot information. Thus, the IRS configures the uplink reflection phase shift based on the new pilot information only upon receipt of the new pilot information. Otherwise, it continues to reflect and transmit uplink data based on the previous uplink reflection phase shift.

[0238] Step 1003: Receive a reflection phase shift vector to be configured sent by a network device.

[0239] Step 1004: Receive the associated model identifier sent by the network device.

[0240] The associated model is a model used by the network device to generate the reflection phase shift vector to be configured. The associated model identifier can be the identifier of the first network model and / or the identifier of the second network model. The relationship between the first and second network models and the phase shift conversion model can be described in detail in any embodiment of the present disclosure and will not be repeated here.

[0241] Optionally, in the present disclosure, step 1003 and step 1004 may be performed in parallel, i.e., the network device sends the reflection phase shift vector to be configured and the associated model identifier to the IRS via a single message. Alternatively, step 1004 may be performed first and then step 1003, which is not limited in the present disclosure.

[0242] Step 1005: Determine the phase shift conversion model to be used currently according to the model identifier.

[0243] Because the reflection phase shift vector to be configured sent by the network device to the IRS is a compressed phase shift vector, the IRS must first decompress the received reflection phase shift vector before configuring the reflection phase shift. In other words, the IRS needs to know how the network device compresses the reflection phase shift vector to be configured. Therefore, the network device can send the IRS the identifier of the currently used first network model and / or the identifier of the second network model. Based on the identifier of the first network model and / or the identifier of the second network model, the IRS can then determine the processing performed by the network device on the reflection phase shift vector to be configured.

[0244] Optionally, to minimize transmission between the network device and the IRS, the network device may send the identifier of the first network model and / or the identifier of the second network model to the IRS only when the current channel statistics differ from the historical channel statistics. Thus, if the IRS does not receive the identifier of the first network model and / or the identifier of the second network model, it may assume that the first network model and the second network model currently used by the network device are the same as those used historically. Consequently, the IRS may use the same phase shift conversion model as used historically to convert the received reflection phase shift vector to be configured.

[0245] Optionally, the IRS can pre-receive a decompressed network model and associated model identifier from a network device, and then associate and store the decompressed network model and associated model identifier. Thus, when configuring reflection phase shift, upon receiving the associated model identifier from the network device, the IRS can determine the phase shift conversion model to be used based on the stored correspondence.

[0246] Step 1006: Input the reflection phase shift vector to be configured into the phase shift conversion model to obtain the reflection phase shift vector to be applied.

[0247] Step 1007: Determine a first reflection phase shift corresponding to each unit in the IRS based on the reflection phase shift vector to be applied.

[0248] In the present disclosure, upon receiving pilot information sent by a network device, the IRS can determine the second reflection phase shift corresponding to each unit in the uplink transmission based on the pilot information. Furthermore, upon receiving the reflection phase shift vector to be configured and the associated model identifier sent by the network device, the IRS can determine the phase shift conversion model to be used based on the associated model identifier. The current phase shift conversion model to be used can then be used to convert it into a larger reflection phase shift vector to be applied. Based on the reflection phase shift vector to be applied, the IRS can then determine the reflection phase shift corresponding to each unit. This allows the network device to configure a smaller reflection phase shift vector for the IRS, reducing the overhead and rate performance loss of configuring the reflection phase shift vector for the IRS.

[0249] Please refer to Figure 11, which is a schematic diagram of the structure of a communication device provided in an embodiment of the present disclosure. The communication device 1100 shown in Figure 11 may include a processing module 1101 and a transceiver module 1102. The transceiver module 1102 may include a sending module and / or a receiving module. The sending module is used to implement a sending function, and the receiving module is used to implement a receiving function. The transceiver module 1102 can implement the sending function and / or the receiving function.

[0250] It is understandable that the communication device 1100 may be a network device, a device in a network device, or a device that can be used in conjunction with a network device.

[0251] The communication device 1100 is on the network device side, wherein:

[0252] The transceiver module 1102 is configured to receive a pilot signal sent by a terminal device;

[0253] The processing module 1101 is configured to perform decorrelation processing on the pilot signal;

[0254] The processing module 1101 is further configured to input the decorrelated pilot signal into the first network model to obtain an initial reflection phase shift vector;

[0255] The processing module 1101 is further configured to input the initial reflection phase shift vector into the second network model to obtain a reflection phase shift vector to be configured, where the scale of the reflection phase shift vector to be configured is smaller than the scale of the initial reflection phase shift vector;

[0256] The transceiver module 1102 is further configured to send the reflection phase shift vector to be configured to the IRS.

[0257] Optionally, the processing module 1101 is further configured to:

[0258] Inputting the initial reflection phase shift vector into the compression subnetwork in the second network model to obtain a compressed reflection phase shift vector;

[0259] The compressed reflection phase shift vector is input into the quantization subnetwork in the second network model to obtain the reflection phase shift vector to be configured.

[0260] Optionally, the processing module 1101 is further configured to:

[0261] Inputting the compressed reflected phase shift vector into a quantizer in the quantization subnetwork to obtain a quantized phase shift vector;

[0262] Inputting the compressed reflected phase shift vector into the strategy sub-network in the quantization sub-network to obtain a mask vector;

[0263] The mask vector and the quantized phase shift vector are merged to obtain the reflection phase shift vector to be configured.

[0264] Optionally, the processing module 1101 is further configured to:

[0265] Determining current pilot information according to the pilot signal sent by the terminal device, the pilot information including the number of subframes included in the pilot signal and the number of pilot symbols included in each subframe;

[0266] The transceiver module 1102 is further configured to send the current pilot information to the IRS.

[0267] Optionally, the transceiver module 1102 is further configured to send the current pilot information to the IRS when the current pilot information is different from the historical pilot information.

[0268] Optionally, the processing module 1101 is further configured to:

[0269] Determine current channel statistics;

[0270] A first network model and a second network model associated with the current channel statistical characteristics are obtained from a preset model library.

[0271] Optionally, the transceiver module 1102 is further configured to:

[0272] The identifier of the first network model and / or the identifier of the second network model is sent to the IRS.

[0273] Optionally, the transceiver module 1102 is further configured to:

[0274] The current channel statistical characteristic is different from the historical channel statistical characteristic, and the identifier of the first network model and / or the identifier of the second network model is sent to the IRS.

[0275] Optionally, the processing module 1101 is further configured to:

[0276] Obtaining a training data set, the training data set including decorrelated pilot signals corresponding to a plurality of terminal devices, a first matrix corresponding to a direct link channel between each terminal device and the network device, and a second matrix corresponding to a reflected link cascade channel between the network device and the terminal device;

[0277] Inputting the training data set into a network model to be trained, obtaining a reference beamforming matrix output by a first initial network model and a reference reflection phase shift vector output by an initial phase shift conversion model in the network model to be trained, wherein the network to be trained includes a cascaded first initial network model, a second initial network model, and an initial phase shift conversion model;

[0278] determining a loss value according to the reference reflection phase shift vector, the reference beamforming matrix, the first matrix, and the second matrix;

[0279] The initial network model is modified based on the loss value until a loss value determined based on a reflection phase shift vector and a beamforming matrix output by the modified network model is less than a threshold value, wherein the modified network model includes the cascaded first network model, the second network model, and the phase shift conversion model.

[0280] Optionally, the transceiver module 1102 is further configured to:

[0281] The phase shift conversion network model and the associated model identifier are sent to the IRS, where the associated model identifier includes the identifier of the first network model and / or the identifier of the second network model.

[0282] In the present disclosure, after receiving a pilot signal from a terminal device, a network device first performs decorrelation processing on the pilot signal. The decorrelation-processed pilot signal is then input into a first network model to obtain an initial reflection phase shift vector. The initial reflection phase shift vector is then input into a second network model to obtain a smaller reflection phase shift vector to be configured. This reflection phase shift vector is then sent to the IRS. This reduces the size of the reflection phase shift vector to be configured, thereby reducing the overhead and rate performance loss of the network device in configuring the reflection phase shift vector to the IRS.

[0283] It is understandable that the communication device 1100 may be an IRS, or a device in the IRS, or a device that can be used in conjunction with the IRS.

[0284] Communication device 1100, on the IRS side, wherein:

[0285] The transceiver module 1102 is configured to receive a reflection phase shift vector to be configured sent by a network device;

[0286] The processing module 1101 is configured to input the reflection phase shift vector to be configured into a phase shift conversion model to obtain a reflection phase shift vector to be applied, where the scale of the reflection phase shift vector to be applied is larger than the scale of the reflection phase shift vector to be configured;

[0287] The processing module 1101 is further configured to determine a first reflection phase shift corresponding to each unit in the IRS based on the reflection phase shift vector to be applied.

[0288] Optionally, the processing module 1101 is further configured to:

[0289] The IRS supports first precision phase configuration, inputs the reflection phase shift vector to be configured into a first decompression model, and obtains the reflection phase shift vector to be applied; or

[0290] The IRS supports second-precision phase configuration, inputs the reflection phase shift vector to be configured into the cascaded second decompression model and quantization model, and obtains the reflection phase shift vector to be applied;

[0291] The first precision is higher than the second precision.

[0292] Optionally, the processing module 1101 is further configured to:

[0293] Inputting the reflection phase shift vector to be configured into the second decompression model, and obtaining a phase shift angle value corresponding to each element of the reflection phase shift vector to be configured output by the second decompression model;

[0294] Inputting the phase shift angle value into the quantization model to obtain a quantized phase shift angle;

[0295] The reflection phase shift vector to be applied is determined according to the quantized phase shift angle.

[0296] Optionally, the transceiver module 1102 is further configured to:

[0297] receiving pilot information sent by the network device;

[0298] The processing module 1101 is further configured to determine, according to the pilot information, a second reflection phase shift corresponding to each unit in the IRS during uplink transmission.

[0299] Optionally, the transceiver module 1102 is further configured to receive an associated model identifier sent by the network device, where the associated model is a model used by the network device to generate the reflection phase shift vector to be configured;

[0300] The processing module 1101 is further configured to determine the phase shift conversion model to be currently used according to the model identifier.

[0301] Optionally, the transceiver module 1102 is further configured to receive a phase shift conversion model and an associated model identifier sent by the network device;

[0302] The processing module 1101 is further configured to store the phase shift conversion model and the associated model identifier in association with each other.

[0303] In this disclosure, after receiving a reflection phase shift vector to be configured from a network device, the IRS uses a shift conversion model to convert it into a larger reflection phase shift vector to be applied. Based on this reflection phase shift vector, the IRS then determines the first reflection phase shift corresponding to each unit. This allows the network device to configure a smaller reflection phase shift vector for the IRS, reducing the overhead and performance loss associated with configuring the reflection phase shift vector.

[0304] Please refer to Figure 12, which is a schematic diagram of the structure of another communication device provided in an embodiment of the present disclosure. Communication device 1200 can be a network device, an IRS, a chip, a chip system, or a processor that supports the network device to implement the above-mentioned method, or a chip, a chip system, or a processor that supports the IRS to implement the above-mentioned method. This device can be used to implement the method described in the above-mentioned method embodiment. For details, please refer to the description of the above-mentioned method embodiment.

[0305] The communication device 1200 may include one or more processors 1201. The processor 1201 may be a general-purpose processor or a dedicated processor. For example, it may be a baseband processor or a central processing unit. The baseband processor may be used to process communication protocols and communication data, and the central processing unit may be used to control the communication device (e.g., a base station, a baseband chip, a terminal device, a terminal device chip, a DU or CU, etc.), execute computer programs, and process computer program data.

[0306] Optionally, the communication device 1200 may further include one or more memories 1202, on which a computer program 1204 may be stored. The processor 1201 executes the computer program 1204 to cause the communication device 1200 to perform the method described in the above method embodiment. Optionally, the memory 1202 may also store data. The communication device 1200 and the memory 1202 may be provided separately or integrated together.

[0307] Optionally, the communication device 1200 may further include a transceiver 1205 and an antenna 1206. The transceiver 1205 may be referred to as a transceiver unit, a transceiver, or a transceiver circuit, and is configured to implement transceiver functions. The transceiver 1205 may include a receiver and a transmitter. The receiver may be referred to as a receiver or a receiving circuit, and is configured to implement a receiving function; the transmitter may be referred to as a transmitter or a transmitting circuit, and is configured to implement a transmitting function.

[0308] Optionally, the communication device 1200 may further include one or more interface circuits 1207. The interface circuit 1207 is configured to receive code instructions and transmit the code instructions to the processor 1201. The processor 1201 executes the code instructions to enable the communication device 1200 to perform the method described in the above method embodiment.

[0309] The communication device 1200 is a network device: the processor 1201 is used to execute steps 202, 203, and 204 in Figure 2, steps 302, 303, and 304 in Figure 3, etc., and the transceiver 1205 is used to execute steps 201 and 205 in Figure 2, or steps 301 and 306 in Figure 3, etc.

[0310] The communication device 1200 is an IRS: the transceiver 1205 is used to execute step 801 in FIG. 8 , or step 901 in FIG. 9 , etc., and the processor 1201 is used to execute step 802 , step 803 , etc. in FIG. 8 .

[0311] In one implementation, processor 1201 may include a transceiver for implementing receiving and transmitting functions. For example, the transceiver may be a transceiver circuit, an interface, or an interface circuit. The transceiver circuit, interface, or interface circuit for implementing the receiving and transmitting functions may be separate or integrated. The transceiver circuit, interface, or interface circuit may be used for reading and writing code / data, or may be used for transmitting or delivering signals.

[0312] In one implementation, processor 1201 may store a computer program 1203. Computer program 1203, when executed on processor 1201, enables communication device 1200 to perform the method described in the above method embodiment. Computer program 1203 may be embedded in processor 1201, in which case processor 1201 may be implemented by hardware.

[0313] In one implementation, the communication device 1200 may include a circuit that can implement the functions of sending, receiving, or communicating in the aforementioned method embodiments. The processor and transceiver described in the present disclosure can be implemented on an integrated circuit (IC), an analog IC, a radio frequency integrated circuit RFIC, a mixed signal IC, an application specific integrated circuit (ASIC), a printed circuit board (PCB), an electronic device, etc. The processor and transceiver can also be manufactured using various IC process technologies, such as complementary metal oxide semiconductor (CMOS), N-type metal oxide semiconductor (nMetal-oxide-semiconductor, NMOS), P-type metal oxide semiconductor (positive channel metal oxide semiconductor, PMOS), bipolar junction transistor (bipolar junction transistor, BJT), bipolar CMOS (BiCMOS), silicon germanium (SiGe), gallium arsenide (GaAs), etc.

[0314] The communication device described in the above embodiments may be a network device or an intelligent relay, but the scope of the communication device described in this disclosure is not limited thereto, and the structure of the communication device may not be limited to FIG12. The communication device may be an independent device or may be part of a larger device. For example, the communication device may be:

[0315] (1) An independent integrated circuit (IC), or chip, or chip system or subsystem;

[0316] (2) a collection of one or more ICs, optionally including a storage component for storing data and computer programs;

[0317] (3) ASIC, such as modem;

[0318] (4) Modules that can be embedded in other devices;

[0319] (5) Receivers, terminal devices, intelligent terminal devices, cellular phones, wireless devices, handheld devices, mobile units, vehicle-mounted devices, network devices, cloud devices, artificial intelligence devices, etc.;

[0320] (6)Others, etc.

[0321] If the communication device can be a chip or a chip system, please refer to the schematic diagram of the chip structure shown in Figure 13. The chip shown in Figure 13 includes a processor 1301 and an interface 1303. The number of processors 1301 can be one or more, and the number of interfaces 1303 can be multiple.

[0322] For the case where the chip is used to implement the functions of the network device in the embodiments of the present disclosure:

[0323] Interface 1303 is used to execute step 201, step 205, etc. in FIG2 .

[0324] For the case where the chip is used to implement the IRS function in the embodiments of the present disclosure:

[0325] Interface 15303 is used to execute step 801 in FIG8 , etc.

[0326] Optionally, the chip further includes a memory 1303, which is used to store necessary computer programs and data.

[0327] Those skilled in the art will also appreciate that the various illustrative logical blocks and steps listed in the embodiments of the present disclosure may be implemented by electronic hardware, computer software, or a combination of both. Whether such functionality is implemented by hardware or software depends on the specific application and the design requirements of the entire system. Those skilled in the art may use various methods to implement the described functionality for each specific application, but such implementation should not be construed as exceeding the scope of protection of the embodiments of the present disclosure.

[0328] The present disclosure also provides a readable storage medium having instructions stored thereon, which implement the functions of any of the above method embodiments when executed by a computer.

[0329] The present disclosure also provides a computer program product, which implements the functions of any of the above method embodiments when executed by a computer.

[0330] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware or any combination thereof. When implemented using software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer programs. When the computer program is loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present disclosure are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer program can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer program can be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media integrated therein. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a high-density digital video disc (DVD)), or a semiconductor medium (eg, a solid state disk (SSD)).

[0331] Those skilled in the art will understand that the various numerical numbers such as first and second involved in the present disclosure are only for the convenience of description and are not used to limit the scope of the embodiments of the present disclosure, and also indicate the order of precedence.

[0332] The at least one in the present disclosure can also be described as one or more, and the multiple can be two, three, four or more, which is not limited in the present disclosure. In the embodiments of the present disclosure, for a technical feature, the technical features in the technical feature are distinguished by "first", "second", "third", "A", "B", "C" and "D", and there is no order of precedence or size between the technical features described by "first", "second", "third", "A", "B", "C" and "D".

[0333] The correspondences shown in the tables of the present disclosure can be configured or predefined. The values ​​of the information in each table are merely examples and can be configured to other values, which are not limited by the present disclosure. When configuring the correspondence between information and parameters, it is not necessary to configure all the correspondences shown in each table. For example, in the tables of the present disclosure, the correspondences shown in certain rows may not be configured. For another example, appropriate modifications and adjustments can be made based on the above tables, such as splitting, merging, etc. The names of the parameters shown in the titles of the above tables may also use other names that are understandable to the communication device, and the values ​​or representations of the parameters may also use other values ​​or representations that are understandable to the communication device. When implementing the above tables, other data structures may also be used, such as arrays, queues, containers, stacks, linear lists, pointers, linked lists, trees, graphs, structures, classes, heaps, hash tables, or hash tables.

[0334] The predefined in the present disclosure may be understood as defined, predefined, stored, pre-stored, pre-negotiated, pre-configured, solidified, or pre-burned.

[0335] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this disclosure.

[0336] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0337] The above description is merely a specific embodiment of the present disclosure, but the scope of protection of the present disclosure is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this disclosure should be included in the scope of protection of the present disclosure. Therefore, the scope of protection of the present disclosure should be based on the scope of protection of the claims.

Claims

1. A method for transmitting a phase shift configuration of an intelligent reflecting surface (IRS), characterized in that: Executed by a network device, the method includes: receiving a pilot signal sent by a terminal device; performing decorrelation processing on the pilot signal; Inputting the decorrelated pilot signal into the first network model to obtain an initial reflection phase shift vector; Inputting the initial reflection phase shift vector into the second network model to obtain a reflection phase shift vector to be configured, wherein the scale of the reflection phase shift vector to be configured is smaller than the scale of the initial reflection phase shift vector; The reflection phase shift vector to be configured is sent to the IRS.

2. The method according to claim 1, wherein Inputting the initial reflection phase shift vector into the second network model to obtain the reflection phase shift vector to be configured includes: Inputting the initial reflection phase shift vector into the compression subnetwork in the second network model to obtain a compressed reflection phase shift vector; The compressed reflection phase shift vector is input into the quantization subnetwork in the second network model to obtain the reflection phase shift vector to be configured.

3. The method according to claim 2, wherein The step of inputting the compressed reflected phase shift vector into the quantization subnetwork in the second network model includes: Inputting the compressed reflected phase shift vector into a quantizer in the quantization subnetwork to obtain a quantized phase shift vector; Inputting the compressed reflected phase shift vector into the strategy sub-network in the quantization sub-network to obtain a mask vector; The mask vector and the quantized phase shift vector are merged to obtain the reflection phase shift vector to be configured.

4. The method according to claim 1, wherein Also includes: Determining current pilot information according to the pilot signal sent by the terminal device, the pilot information including the number of subframes included in the pilot signal and the number of pilot symbols included in each subframe; The current pilot information is sent to the IRS.

5. The method according to claim 4, wherein The sending the current pilot information to the IRS includes: The current pilot information is different from the historical pilot information, and the current pilot information is sent to the IRS 6. The method according to claim 1, wherein Also includes: Determine current channel statistics; The first network model and the second network model associated with the current channel statistical characteristics are obtained from a preset model library.

7. The method according to claim 6, wherein Also includes: The identifier of the first network model and / or the identifier of the second network model is sent to the IRS.

8. The method according to claim 7, wherein The current channel statistical characteristics are different from historical channel statistical characteristics.

9. The method according to any one of claims 1 to 8, wherein: Also includes: Obtaining a training data set, the training data set including decorrelated pilot signals corresponding to a plurality of terminal devices, a first matrix corresponding to a direct link channel between each terminal device and the network device, and a second matrix corresponding to a reflected link cascade channel between the network device and the terminal device; Inputting the training data set into a network model to be trained, obtaining a reference beamforming matrix output by a first initial network model and a reference reflection phase shift vector output by an initial phase shift conversion model in the network model to be trained, wherein the network to be trained includes a cascaded first initial network model, a second initial network model, and an initial phase shift conversion model; determining a loss value according to the reference reflection phase shift vector, the reference beamforming matrix, the first matrix, and the second matrix; The initial network model is modified based on the loss value until a loss value determined based on a reflection phase shift vector and a beamforming matrix output by the modified network model is less than a threshold value, wherein the modified network model includes the cascaded first network model, the second network model, and the phase shift conversion model.

10. The method according to claim 9, wherein Also includes: The phase shift conversion network model and the associated model identifier are sent to the IRS, where the associated model identifier includes the identifier of the first network model and / or the identifier of the second network model.

11. A method for receiving a phase shift configuration of an intelligent reflecting surface (IRS), characterized in that: Executed by the IRS, the method comprises: receiving a reflection phase shift vector to be configured sent by a network device; Inputting the reflection phase shift vector to be configured into a phase shift conversion model to obtain a reflection phase shift vector to be applied, wherein the scale of the reflection phase shift vector to be applied is larger than the scale of the reflection phase shift vector to be configured; Based on the reflection phase shift vector to be applied, a first reflection phase shift corresponding to each unit in the IRS is determined.

12. The method according to claim 11, wherein Inputting the reflection phase shift vector to be configured into a phase shift conversion model to obtain the reflection phase shift vector to be applied includes: The IRS supports first precision phase configuration, inputs the reflection phase shift vector to be configured into a first decompression model, and obtains the reflection phase shift vector to be applied; or The IRS supports second-precision phase configuration, inputs the reflection phase shift vector to be configured into the cascaded second decompression model and quantization model, and obtains the reflection phase shift vector to be applied; The first precision is higher than the second precision.

13. The method according to claim 12, wherein: Inputting the reflection phase shift vector to be configured into the cascaded second decompression model and quantization model to obtain the IRS reflection phase shift vector to be applied includes: Inputting the reflection phase shift vector to be configured into the second decompression model to obtain a phase shift angle value corresponding to each element of the reflection phase shift vector to be configured output by the second decompression model; Inputting the phase shift angle value into the quantization model to obtain a quantized phase shift angle; The reflection phase shift vector to be applied is determined according to the quantized phase shift angle.

14. The method according to claim 11, wherein Also includes: receiving pilot information sent by the network device; A second reflection phase shift corresponding to each unit in the IRS during uplink transmission is determined according to the pilot information.

15. The method according to claim 11, wherein Also includes: receiving an associated model identifier sent by the network device, where the associated model is a model used by the network device to generate the reflection phase shift vector to be configured; The phase shift conversion model to be currently used is determined according to the model identifier.

16. The method according to claim 15, wherein Also includes: receiving a phase shift conversion model and an associated model identifier sent by the network device; The phase shift conversion model and the associated model identifier are stored in association with each other.

17. A network device, characterized in that: The device comprises: A transceiver module, configured to receive a pilot signal sent by a terminal device; a processing module, configured to perform decorrelation processing on the pilot signal; The processing module is further configured to input the decorrelated pilot signal into the first network model to obtain an initial reflection phase shift vector; The processing module is further configured to input the initial reflection phase shift vector into the second network model to obtain a reflection phase shift vector to be configured, wherein the scale of the reflection phase shift vector to be configured is smaller than the scale of the initial reflection phase shift vector; The transceiver module is further configured to send the reflection phase shift vector to be configured to the IRS.

18. An intelligent reflective surface (IRS), characterized in that: The device comprises: A transceiver module, configured to receive a reflection phase shift vector to be configured sent by a network device; a processing module, configured to input the reflection phase shift vector to be configured into a phase shift conversion model to obtain a reflection phase shift vector to be applied, wherein the scale of the reflection phase shift vector to be applied is larger than the scale of the reflection phase shift vector to be configured; The processing module is further configured to determine a first reflection phase shift corresponding to each unit in the IRS based on the reflection phase shift vector to be applied.

19. A communication device, characterized in that: The device includes a processor and a memory, wherein a computer program is stored in the memory, and the processor executes the computer program stored in the memory to enable the device to perform the method according to any one of claims 1 to 10, or the method according to any one of claims 11 to 16.

20. A communication system, characterized in that: The communication system includes a network device and an intelligent reflective surface (IRS), wherein the network device executes the method according to any one of claims 1 to 10, and the IRS is configured to execute the method according to any one of claims 11 to 16.

21. A computer-readable storage medium storing instructions, which, when executed, enable the method according to any one of claims 1 to 10 to be implemented, or enable the method according to any one of claims 11 to 16 to be implemented.

Images