An overhead-splitting cascaded channel state information feedback method, system and device

Abstract

The application discloses a kind of overhead apportioning cascade channel state information feedback method, system and equipment, belong to wireless communication technical field, the angular domain support similarity between adjacent subcarriers is calculated for the same user, and adjacent subcarriers are adaptively divided into several sharing windows according to angular domain support similarity;The feedback budget of each user on each subcarrier is decomposed into support description overhead, scale description overhead and coefficient quantization overhead;In each sharing window, support description information and scale description information are fed back once, and they are reused in multiple subcarriers in the sharing window;Calculate the effective coefficient quantization bit number, quantize the complex coefficient in the angular domain support description information, and reconstruct the cascade channel on each subcarrier according to the quantization result.The application can relieve the feedback resource pressure in multi-user scenario, improve the expansibility of system under limited feedback condition.

Description

Technical Field

[0001] This invention relates to the field of wireless communication technology, and in particular to a method, system, and device for overhead-sharing cascaded channel state information feedback. Background Technology

[0002] Reconfigurable smart surfaces (RIS) are a novel wireless communication auxiliary structure composed of a large number of low-cost passive or near-passive reflective elements. By programmably controlling the phase of each reflective element, RIS can alter the wireless propagation environment, enhance the desired signal propagation path, and improve the coverage and transmission performance of the communication system. In wideband orthogonal frequency division multiplexing (OFDM) systems, RIS can work in conjunction with multiple subcarriers, thereby providing additional array gain and spatial control capabilities for wideband communication links.

[0003] In reconfigurable smart surface communication systems, there is no direct communication link between the base station and the user. Effective transmission relies on the reflection link of the reconfigurable smart surface. The overall equivalent channel is a concatenated channel formed by the concatenation of the channel from the base station to the reconfigurable smart surface and the channel from the reconfigurable smart surface to the user. Therefore, reconfigurable smart surface communication systems need to acquire concatenated channel state information to achieve reliable transmission and phase shift control. In broadband orthogonal frequency division multiplexing systems, the number of subcarriers is large, and the number of reconfigurable smart surface reflection units is also large. The concatenated channels of each user on different subcarriers need to be described and fed back, which will bring significant feedback overhead.

[0004] Existing cascaded channel state information (CSI) feedback methods typically feed back CSI independently for each subcarrier. This means each user feeds back corresponding channel support information, scale description, and channel coefficient quantization information on each subcarrier. Repeatedly feeding back similar or even identical CSI information on each subcarrier leads to redundant consumption of fixed feedback overhead. When the feedback budget is limited, the redundant support and scale overhead consumes a large proportion of bits, reducing the effective bits available for complex coefficient quantization and thus lowering the reconstruction accuracy of the cascaded channel. Cascaded channel reconstruction errors further affect the phase configuration of reconfigurable smart surfaces. If the feedback budget is mainly consumed by fixed overhead, the reconstructed cascaded channel cannot accurately reflect the true channel characteristics, and the reflection phase configuration obtained based on this reconstructed channel will deviate from the optimal configuration, ultimately leading to a decrease in the system's achievable rate. This problem is particularly pronounced in scenarios with low feedback budgets, multiple users, and multiple subcarriers.

[0005] Therefore, it is urgent to solve the problem of high overhead in cascaded channel state information feedback. Summary of the Invention

[0006] This invention aims to provide a method, system, and device for overhead-sharing cascaded channel state information feedback. By dividing a shared window and reusing support description information and scale description information within the shared window, the repetitive fixed overhead in the subcarrier-by-subcarrier feedback process is reduced. The reuse of support information and scale information increases the number of effective coefficient quantization bits. Under the same feedback budget, complex coefficients can achieve higher quantization accuracy, thus solving the problem of high overhead in cascaded channel state information feedback.

[0007] To achieve the above objectives, the technical solution of the present invention is as follows: A method for overhead-sharing cascaded channel state information feedback, applied to a reconfigurable smart surface communication system, includes the following steps: Step S1: Construct a reconfigurable intelligent surface communication system model and determine the system parameters in the reconfigurable intelligent surface communication system model according to the deployment scenario; the system parameters include the corner domain dictionary, the number of users, and the number of subcarriers; Step S2: Map the cascaded channels of each user on each subcarrier to the corner domain dictionary to obtain the support set on each subcarrier; Step S3 involves dividing the shared window based on the support set on each subcarrier to obtain the length of the shared window after the division. This includes the following steps: Step S31: Based on the support set on each subcarrier, filter to obtain the dominant angular domain support set on each subcarrier; Step S32: For the same user, calculate the angular domain support similarity based on the dominant angular domain support set between adjacent subcarriers; Step S33: Based on the corner domain support similarity and shared window length constraints, the adjacent subcarriers are divided into shared windows to obtain the shared window length after the shared window division is completed; Step S4: After completing the shared window partitioning, decompose the feedback budget of each user on each subcarrier into support description overhead, scale description overhead, and coefficient quantization overhead. Step S5: Within each shared window, the supporting description information and scale description information are fed back in one go and reused for multiple subcarriers within the shared window.

[0008] Further, in step S2, the cascaded channels of each user on each subcarrier are mapped into the corner domain dictionary, as follows: in, This represents the cascaded channel of user k on the nth subcarrier, where A represents the corner domain dictionary. Represents the angular domain sparse coefficient vector of user k on the nth subcarrier; The support set is represented as follows: in, Let k be the support set for user k on the nth subcarrier.

[0009] Further, in step S3, the corner domain support similarity is represented as: in, This indicates that user k is in the nth adjacent subcarrier and the nth... Corner domain support similarity between subcarriers and These represent the values ​​of user k on the nth subcarrier and the nth subcarrier, respectively. Dominant angular domain support set on each subcarrier.

[0010] Further, in step S3, the method for dividing the shared window is as follows: when the average support similarity within the current shared window is greater than or equal to a preset support similarity threshold and the shared window length is less than the maximum shared window length, the current shared window continues to expand backward; when the two conditions are not simultaneously met, that is, when the average support similarity is less than the support similarity threshold and the shared window length is greater than or equal to the maximum shared window length, the expansion stops and the current shared window is determined, and the current shared window is determined as a shared window; the above process is repeated from the next ungrouped subcarrier until all subcarriers have been grouped; The shared window length constraint refers to setting upper and lower boundary limits for the adaptively partitioned subcarrier shared window, including a minimum shared window length and a maximum shared window length. When expanding the window and determining whether to add adjacent subcarriers, the shared window length satisfies the following: In the formula, To share window length, To minimize the shared window length, This is the maximum shared window length.

[0011] Furthermore, the aforementioned overhead-sharing cascaded channel state information feedback method further includes: Step S6: Calculate the number of effective coefficient quantization bits after cost amortization based on the length of the shared window after the shared window partitioning, the support description overhead, and the scale description overhead. Step S7: Quantize the complex coefficients in the corner domain support description information using the effective coefficients quantization bit number, and reconstruct the cascaded channel on each subcarrier based on the quantization result; Step S8: Based on the reconstructed cascaded channel, perform reconfigurable smart surface phase configuration to obtain a reconfigurable smart surface phase vector, and use the reconfigurable smart surface phase vector for data transmission in the broadband orthogonal frequency division multiplexing system.

[0012] Furthermore, the feedback budget satisfies: in, User k represents the feedback budget for the scale description overhead on the nth subcarrier. This indicates the overhead of the supporting description. Indicate the cost of scale description. This represents the cost of coefficient quantization used for complex coefficient quantization.

[0013] Furthermore, the scale description information is the shared dynamic range scale within the shared window, expressed as: in, This represents the scale of the shared dynamic range of user k in the g-th shared window. This indicates that user k is on the nth subcarrier. The complex coefficients corresponding to each support position and These represent the operations of extracting the real part and the imaginary part, respectively.

[0014] This invention introduces corner domain support similarity calculation before shared window partitioning. Utilizing the approximate consistency of corner domain support for the same user on adjacent subcarriers, and based on the corner domain support similarity, support similarity threshold, and shared window length constraint, adjacent subcarriers are partitioned into shared windows. Support description information is reused within the shared window, avoiding redundant feedback of similar support information and effectively reducing the fixed overhead in the multi-subcarrier cascaded channel state information feedback process. Furthermore, dynamic range scale description information is reused within the shared window, eliminating the need for independent feedback of scale information on each subcarrier. By jointly amortizing the support description overhead and scale description overhead, this invention can reserve more effective bits for complex coefficient quantization under the same per-user, per-subcarrier feedback budget, thereby improving the utilization efficiency of limited feedback resources.

[0015] Furthermore, the effective coefficient quantization bit number is expressed as: in, Let represent the number of effective coefficient quantization bits for user k on the nth subcarrier, and b represent the feedback budget per user subcarrier. This indicates the number of bits used for scale description information. This indicates the length of the shared window after the shared window partitioning is completed. Based on the shared window length, support description overhead, and scale description overhead, this invention calculates the effective coefficient quantization bits after overhead amortization. This clearly depicts the relationship between fixed overhead reuse and coefficient quantization accuracy, making the feedback resource allocation process more explicit and providing a basis for system design under different feedback budgets and shared window lengths.

[0016] Furthermore, after obtaining the effective coefficient quantization bits after cost amortization, the complex coefficients on the shared support set are quantized. The complex coefficients at the support locations are decomposed into real and imaginary parts, as follows: in, This indicates the operation of taking the real part. This indicates the operation of taking the imaginary part. This indicates that user k is on the nth subcarrier. The complex coefficients corresponding to each support position.

[0017] A cascaded channel state information feedback system with cost-sharing includes: The system model building module constructs a model of the reconfigurable intelligent surface communication system and determines the system parameters. The angular domain sparse representation module maps the cascaded channels of each user on each subcarrier to the angular domain dictionary, and obtains the support set on each subcarrier; The shared window partitioning module filters out the dominant angular domain support set on each subcarrier based on the support set on each subcarrier; for the same user, it calculates the angular domain support similarity based on the dominant angular domain support set between adjacent subcarriers, and performs shared window partitioning on adjacent subcarriers based on the angular domain support similarity and shared window length constraints, determines the shared support set of the user in each shared window, and obtains the shared window length after the shared window partitioning is completed; The feedback budget decomposition module, after completing the shared window partitioning, decomposes the feedback budget of each user on each subcarrier into support description overhead, scale description overhead, and coefficient quantization overhead. The support and scale sharing module provides one-time feedback of support description information and scale description information within each sharing window, and reuses them for multiple subcarriers within the sharing window.

[0018] Furthermore, the aforementioned overhead-sharing cascaded channel state information feedback system further includes: The effective bit calculation module calculates the number of effective coefficient quantization bits after cost amortization based on the length of the shared window after the shared window partitioning, the support description overhead, and the scale description overhead. The coefficient quantization and channel reconstruction module uses the effective coefficient quantization bit number to quantize the complex coefficients in the corner domain support description information, and reconstructs the cascaded channel on each subcarrier based on the quantization result; The phase configuration module performs reconfigurable smart surface phase configuration based on the reconstructed cascaded channel to obtain a reconfigurable smart surface phase vector, which is then used for data transmission in a broadband orthogonal frequency division multiplexing system.

[0019] A communication device includes a controller and a memory, wherein a computer program is stored in the memory, and when the computer program is executed by the controller, it implements a cascaded channel state information feedback method for overhead sharing as described in any of the preceding claims.

[0020] Furthermore, the controller employs an overhead-sharing cascaded channel state information feedback system as described in any of the above-mentioned embodiments.

[0021] A computer-readable storage medium storing a computer program, which, when executed by a controller, implements the overhead-sharing cascaded channel state information feedback method described in any one of the preceding claims.

[0022] Beneficial effects: This invention reduces repetitive fixing overhead in the subcarrier-by-subcarrier feedback process by dividing a shared window and reusing support description information and scale description information within the shared window. The reuse of support and scale information increases the effective coefficient quantization bits, resulting in higher quantization accuracy for complex coefficients under the same feedback budget. This invention is applicable to multi-user broadband communication scenarios. When the number of users increases or the total feedback budget is limited, the overhead amortization mechanism within the shared window can alleviate feedback resource pressure in multi-user scenarios and improve the system's scalability under limited feedback conditions.

[0023] To make the above-mentioned features and advantages of the invention more apparent and understandable, specific embodiments are described below, and detailed descriptions are provided in conjunction with the accompanying drawings. Attached Figure Description

[0024] Figure 1 This is a flowchart of a cascaded channel state information feedback method for overhead sharing according to the present invention.

[0025] Figure 2 This is a schematic diagram of a reconfigurable smart surface communication system model structure in one embodiment of the present invention.

[0026] Figure 3 This is a schematic diagram of the structure of a cascaded channel state information feedback system with overhead sharing according to the present invention.

[0027] Figure 4This is a waveform comparison of cascaded channel reconstruction errors under different feedback budgets in one embodiment of the present invention.

[0028] Figure 5 This is a waveform diagram comparing the average reachable rate of the system under different feedback budgets in one embodiment of the present invention.

[0029] Figure 6 This is a waveform diagram comparing the average reachability of the system under different numbers of users in one embodiment of the present invention, wherein (a) is a waveform diagram comparing the average reachability of the system when the total feedback budget is fixed at 30, and (b) is a waveform diagram comparing the average reachability of the system when the total feedback budget is fixed at 40.

[0030] Figure 7 This is a schematic diagram of the structure of a communication device in one embodiment of the present invention. Detailed Implementation

[0031] To make the objectives and technical solutions of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0032] In one embodiment, see Figure 1 This invention provides a method for overhead-sharing cascaded channel state information feedback, applied to a reconfigurable smart surface communication system, comprising steps S1 to S8: Step S1: Construct a reconfigurable intelligent surface communication system model, and determine the system parameters and cascaded channel state information to be fed back in the reconfigurable intelligent surface communication system model according to the deployment scenario; the system parameters include the corner domain dictionary, the number of users, and the number of subcarriers; Step S2: Map the cascaded channels of each user on each subcarrier to the corner domain dictionary to obtain the support set on each subcarrier; Step S3 involves dividing the shared window based on the support set on each subcarrier to obtain the length of the shared window after the division. This includes the following steps: Step S31: Based on the support set on each subcarrier, filter to obtain the dominant angular domain support set on each subcarrier; Step S32: For the same user, calculate the angular domain support similarity based on the dominant angular domain support set between adjacent subcarriers; Step S33: Based on the corner domain support similarity and shared window length constraints, the adjacent subcarriers are divided into shared windows to obtain the shared window length after the shared window division is completed; Step S4: After completing the shared window partitioning, decompose the feedback budget of each user on each subcarrier into support description overhead, scale description overhead, and coefficient quantization overhead. Step S5: Within each shared window, the supporting description information and scale description information are fed back at once and reused for multiple subcarriers within the shared window; Step S6: Calculate the number of effective coefficient quantization bits after cost amortization based on the length of the shared window after the shared window partitioning, the support description overhead, and the scale description overhead. Step S7: Quantize the complex coefficients in the corner domain support description information using the effective coefficient quantization bit number, and reconstruct the cascaded channel on each subcarrier based on the quantization result; Step S8: Based on the reconstructed cascaded channel, perform reconfigurable smart surface phase configuration to obtain a reconfigurable smart surface phase vector, and use the reconfigurable smart surface phase vector for data transmission in the broadband orthogonal frequency division multiplexing system.

[0033] This invention reduces repetitive fixing overhead in the subcarrier-by-subcarrier feedback process by dividing a shared window and reusing support description information and scale description information within the shared window. By reusing support information and scale information, the number of effective coefficient quantization bits is increased, and complex coefficients can achieve higher quantization accuracy under the same feedback budget.

[0034] In step S1, please refer to Figure 1 In step S1, a reconfigurable intelligent surface communication system model is constructed, and system parameters and cascaded channel state information to be fed back are determined according to the deployment scenario; the system parameters include the corner domain dictionary, the number of users, and the number of subcarriers.

[0035] As an example, the system parameters also include the number of reconfigurable smart surface reflective units.

[0036] As an example, the model of the reconfigurable smart surface communication system includes The reconfigurable smart surface includes 1 subcarrier and K users. There are 10 reflection units. For a cascaded channel of user k on the nth subcarrier, denoted as […]. .in, , .

[0037] The cascaded channel is used to describe the equivalent channel formed by the channel from the base station to the reconfigurable smart surface and the channel from the reconfigurable smart surface to the user.

[0038] The cascaded channel state information is used to describe the equivalent channel characteristics from the base station to the user via the reconfigurable smart surface, including angular domain support information, scale description information, and channel complex coefficient information.

[0039] For example, please refer to Figure 2 The reconfigurable smart surface communication system model includes a base station 21, a reconfigurable smart surface 22, and a user area 23. The user area 23 includes multiple users, and the base station 21 provides auxiliary transmission to the user area 23 through the reconfigurable smart surface 22. The reconfigurable smart surface 22 includes multiple adjustable reflective elements, each of which can apply a controllable phase change to the incident signal, thereby altering the wireless propagation environment and enhancing the effective transmission link in the direction of the target user. Figure 2 middle This represents the channel from base station 21 to reconfigurable smart surface 22. This represents the channel from the reconfigurable smart surface 22 to the first user 231. This represents the channels from the reconfigurable smart surface 22 to the Kth user 23K.

[0040] As an example, the reconfigurable smart surface communication system model adopts a broadband orthogonal frequency division multiplexing transmission mode.

[0041] In step S2, please refer to Figure 1 In step S2, before performing cascaded channel feedback, a corner domain dictionary A is constructed in step S1 to map the cascaded channels of each user on each subcarrier to the corner domain dictionary A. For the cascaded channel of user k on the nth subcarrier, it can be represented as: in, This represents the cascaded channel of user k on the nth subcarrier, where A represents the corner domain dictionary. This represents the angular domain sparse coefficient vector of user k on the nth subcarrier.

[0042] Specifically, since reconfigurable smart surface cascaded channels are typically determined by a finite number of dominant propagation paths, cascaded channels exhibit sparse properties in the angular domain. That is, the angular domain sparse coefficient vector... If only a few elements have large amplitude values ​​and the rest are close to zero, then the support set of user k on the nth subcarrier can be represented as: ,in, This represents the set of positions of non-zero elements.

[0043] In step S3, please refer to Figure 1 In step S3, for the same user, adjacent subcarriers in a broadband orthogonal frequency division multiplexing system are relatively close in the frequency domain, and the angular domain positions corresponding to their dominant propagation paths usually do not change drastically. That is, the dominant angular domain support set of the same user within the range of locally adjacent subcarriers has approximately the same consistency. Based on this characteristic, adjacent subcarriers are divided into several shared windows.

[0044] Specifically, for user k on the nth subcarrier, the angular domain sparse coefficient vector Sort the sparsity coefficients of each corner domain from largest to smallest, and select the one with the largest amplitude. Each angular domain sparse coefficient, and the said The position indices corresponding to the sparse coefficients of each angular domain are determined as the dominant angular domain support set of user k on the nth subcarrier, denoted as: in, Denotes the dominant angular domain support set of user k on the nth subcarrier. Represents the angular domain sparse coefficient vector of user k on the nth subcarrier. The Middle There are 10 elements, where E represents the number of grid cells in the corner domain dictionary. Indicates the sparsity of user k. Represents a set The number of elements in the middle.

[0045] Specifically, based on the dominant angular domain support set on each subcarrier Calculate the support similarity between candidate adjacent subcarriers, where user k is in the nth adjacent subcarrier and the nth adjacent subcarrier. Corner domain support similarity on each subcarrier Represented as: in, This indicates that user k is in the nth adjacent subcarrier and the nth... Corner domain support similarity between subcarriers and These represent the values ​​of user k on the nth subcarrier and the nth subcarrier, respectively. Dominant angular domain support set on each subcarrier.

[0046] As an example, let the g-th shared window be denoted as... The shared window consists of several consecutive subcarriers. The shared window... The number of subcarriers included is defined as the shared window length of the shared window, and denoted as . ,Right now: in, Indicates a shared window The number of subcarriers contained therein This indicates the length of the current shared window, and g represents the shared window index.

[0047] In the shared window Within, the dominant angular domain support set of user k in the g-th shared window. satisfy: in, Let k be the support set on the nth subcarrier; This represents the dominant angular domain support set of user k within the g-th shared window, i.e., the shared support set of user k within the g-th shared window.

[0048] User k's dominant domain support set within the g-th shared window Used to characterize the position of the dominant angular domain component of user k within the corresponding shared window.

[0049] Furthermore, the sparsity of user k is set to 1. satisfy: in, This indicates the number of elements in the set.

[0050] The partitioning of the shared window is determined through an adaptive method that supports similarity perception: for user k, when the corner domain supports similarity... Within the preset support similarity threshold When the subcarriers are within the range, the corresponding subcarriers are assigned to the same shared window. Starting with the currently ungrouped subcarrier, it is used as the starting subcarrier of the current shared window, and adjacent subcarriers are added sequentially. After each candidate subcarrier is added, the current shared window is calculated. The average support similarity within the range, wherein the average support similarity is expressed as: in, This indicates that user k is in the shared window. Average support similarity within, To share window length, Indicates that user k is in the adjacent... n The subcarrier and the first Angular domain support similarity on each subcarrier.

[0051] Preset support similarity threshold When the following conditions are met: and: At that time, that is, the average support similarity within the current shared window. Greater than or equal to the support similarity threshold And shared window length Less than the maximum shared window length When the two conditions are not simultaneously met, i.e., when the average support similarity is... Less than the support similarity threshold Or share window length Greater than or equal to the maximum shared window length When the time is reached, stop expanding and determine the current shared window, and designate the current window as a shared window. Then, repeat the above process starting from the next ungrouped subcarrier until all subcarriers are grouped, obtaining the length of the shared window after the shared window division. .

[0052] As an example, when the corner domain support similarity is within a preset support similarity threshold... When the range is met, the corresponding subcarrier is determined to satisfy the support sharing condition. The support similarity threshold... The range of values ​​for can be: .

[0053] Furthermore, the shared window length constraint refers to setting upper and lower boundary limits for the adaptively partitioned subcarrier shared window, including the minimum shared window length. and maximum shared window length Minimum shared window length Used to limit the minimum length and maximum length of a shared window. This is used to avoid support mismatch caused by an excessively large shared window. Specifically, the length of the shared window is determined when expanding the window and deciding whether to add more adjacent subcarriers. satisfy .

[0054] As an example, the minimum shared window length can be taken as... The maximum shared window length can be set according to the number of subcarriers and the channel frequency domain correlation, for example, taking... .

[0055] Using the above-mentioned shared window partitioning method, when the support similarity between adjacent subcarriers is high, a larger shared window is automatically formed; when the support similarity between adjacent subcarriers is low, the shared window is automatically reduced.

[0056] It should be noted that the support sharing strategy in this invention refers to the sharing between adjacent subcarriers for the same user, rather than requiring different users to share the same support set. Different users can have different propagation directions and channel structures, therefore, each user maintains its own shared support set. This approach can utilize the support consistency of a single user within a local frequency domain while preserving the channel differences between multiple users.

[0057] In step S4, please refer to Figure 1 In step S4, the feedback budget satisfies: in, User k represents the feedback budget for the scale description overhead on the nth subcarrier. This indicates the overhead of the supporting description. Indicate the cost of scale description. This represents the cost of coefficient quantization used for complex coefficient quantization.

[0058] As an example, let the grid size of the corner domain dictionary A be E, and the sparsity of user k be... Since each support index needs to be represented from E corner domain grids, each support index can be represented using... Each bit is used for representation. Therefore, the support description overhead for user k is... It can be represented as: Where E represents the grid size of the corner domain dictionary, Indicates the sparsity of user k. This indicates the rounding up operation.

[0059] In step S5, please refer to Figure 1 In step S5, the support description information includes a shared support set for users within a shared window, and the dominant angular domain support set for the same user within locally adjacent subcarrier ranges has approximately the same consistency. Based on this characteristic, within each shared window, all subcarriers reuse the same shared support set. That is, only one support description information needs to be fed back, and the support description information is reused for multiple subcarriers within the window; the scale description information is a shared dynamic range scale within the shared window. Taking advantage of the fact that the complex coefficient amplitude range of the same user is similar within the shared window, a shared dynamic range scale is extracted within each shared window, and the scale description information is reused for multiple subcarriers within the corresponding shared window.

[0060] As an example, for the g-th shared window The shared dynamic range scale of user k can be expressed as: in, This represents the scale of the shared dynamic range of user k in the g-th shared window. This indicates that user k is on the nth subcarrier. The complex coefficients corresponding to each support position and These represent the operations of extracting the real part and the imaginary part, respectively.

[0061] As an example, the shared dynamic range scale The quantization range is determined by the complex coefficients of all subcarriers within the shared window at the shared support location, covering the main value ranges of the real and imaginary parts of the complex coefficients within that window. Therefore, when quantizing the complex coefficients of different subcarriers within each shared window, a unified shared dynamic range scale can be used as the quantization range.

[0062] By extracting a shared dynamic range scale within each shared window and reusing the shared dynamic range scale on multiple subcarriers within that shared window, scale description information only needs to be fed back once within each shared window, instead of repeatedly feeding back scale information on each subcarrier. This further reduces the repetitive fixed overhead caused by dynamic range scale description, allowing more feedback budget to be used for the quantization of the complex coefficients themselves.

[0063] This invention leverages the near-consistency of support information within a shared window, feeding back support description information only once within each shared window and reusing this information across multiple subcarriers within that window. Simultaneously, scale description information can also be reused within the shared window. In this way, support description and scale description overheads are no longer borne individually by each subcarrier, but are shared by multiple subcarriers within the shared window. When the support similarity between adjacent subcarriers is high, a larger shared window is automatically formed to improve the amortization benefits of support description and scale description overheads; when the support similarity between adjacent subcarriers is low, the shared window is automatically reduced to mitigate the impact of support mismatch on complex coefficient quantization and cascaded channel reconstruction.

[0064] It should be noted that the method for extracting the shared scale is not limited to the maximum amplitude method described above. In other embodiments, quantile amplitude, root mean square amplitude, or dynamic range estimate after amplitude limiting can also be used as the shared scale. As long as the scale can be multiplexed by multiple subcarriers within the shared window and used to support the quantization of position complex coefficients, it falls within the protection scope of this invention.

[0065] In step S6, please refer to Figure 1 In step S6, the number of effective coefficient quantization bits is expressed as: in, Let represent the number of effective coefficient quantization bits for user k on the nth subcarrier, and b represent the feedback budget per user subcarrier. This indicates the number of bits used for scale description information. This indicates the length of the shared window after the shared window division is completed.

[0066] As shown in the above expressions, when the shared window length is large, the average support description overhead and scale description overhead borne by each subcarrier decreases. Therefore, under the same feedback budget b, this invention can reserve more effective bits for complex coefficient quantization. Increasing the number of effective coefficient quantization bits can improve the quantization accuracy of complex coefficients at support positions, thereby improving the performance of subsequent cascaded channel reconstruction.

[0067] In step S7, please refer to Figure 1 In step S7, after obtaining the number of effective coefficient quantization bits after cost amortization, the shared support set is... The complex coefficients on the carrier are quantized. For user k in the nth subcarrier within the g-th shared window, the complex coefficients at its support position are denoted as... ,in The complex coefficient can be decomposed into a real part and an imaginary part: in, This indicates the operation of taking the real part. This indicates the operation of taking the imaginary part. This indicates that user k is on the nth subcarrier. The complex coefficients corresponding to each support position.

[0068] In this invention, complex coefficient quantization employs a shared dynamic range scale. As a quantization range, based on the shared dynamic range scale The real and imaginary parts of the complex coefficients at each subcarrier support position within the shared window are quantized separately. The number of bits used for quantization is determined by the number of bits used for quantization of the effective coefficients. The decision is made because this invention increases the number of effective coefficient quantization bits by reusing support and scale information, thus achieving higher quantization accuracy for complex coefficients under the same feedback budget.

[0069] Based on the shared support set obtained from feedback Shared dynamic range scale And the quantized complex coefficients, to recover the angular domain sparse coefficient vector of user k on the nth subcarrier. Among them, the sparse coefficient vector of the corner domain In shared support set Positions outside the set can be set to zero, while positions within the shared support set are recovered from the quantized complex coefficients.

[0070] Furthermore, the cascaded channel is reconstructed using the corner domain dictionary A, and the reconstruction expression is: in, This represents the reconstructed sparse coefficient vector of the corner domain. Reconstruction cascaded channel for user k on the nth subcarrier.

[0071] As an example, when the number of effective coefficients quantized bits A value of zero indicates that the current feedback budget, after deducting amortized fixed costs, is insufficient to support complex coefficient quantization. In this case, the reconstruction coefficients at the corresponding support locations can be set to zero, or a preset low-complexity alternative reconstruction rule can be used. This setting ensures that the system still has a clear feedback and reconstruction process even under extremely low feedback budget conditions.

[0072] Through the aforementioned complex coefficient quantization and concatenated channel reconstruction process, this invention can transform the support information and scale information obtained by multiplexing within a shared window into reconstructed concatenated channels on each subcarrier. These reconstructed concatenated channels are further used for subsequent reconfigurable intelligent surface phase configuration and broadband orthogonal frequency division multiplexing data transmission.

[0073] In step S8, please refer to Figure 1 In step S8, after completing the cascaded channel reconstruction on each user's subcarrier, reconfigurable intelligent surface phase configuration is further performed based on the reconstructed cascaded channel.

[0074] Let the reconstructed set of multi-user, multi-subcarrier concatenated channels be: in, This represents the reconstructed concatenated channel for user k on the nth subcarrier, where K represents the number of users. Indicates the number of subcarriers.

[0075] Since reconfigurable smart surfaces typically employ a frequency-flat common reflection phase configuration within a coherent block, a common reconfigurable smart surface phase vector can be generated based on the reconfigurable concatenated channels on all users and all subcarriers. This process can be expressed as: in, This represents the phase vector of a reconfigurable smart surface. This represents a function for phase configuration based on reconstructed cascaded channels.

[0076] As an example, statistical aggregation can be performed on the reconfigured concatenated channels on multiple users and multiple subcarriers to obtain the equivalent channel characteristics for reflection phase design. Then, the phase of each reflection unit of the reconfigurable smart surface is updated according to these equivalent channel characteristics. If the reflection units of the reconfigurable smart surface use finite-resolution phase control, the continuous phase values ​​can be quantized to the nearest hardware phase codebook value to obtain a phase configuration that meets the actual hardware constraints.

[0077] Obtain the reconfigurable smart surface phase vector Then, the reconstructable smart surface phase vector Used for broadband orthogonal frequency division multiplexing (OFDM) transmission. For a cascaded channel of user k on the nth subcarrier. The configured equivalent receive gain and In other words, the more accurate the reconstructed cascaded channel is, the better the reconfigurable smart surface phase configuration generated based on it can match the real channel, thereby improving the system's transmission performance.

[0078] By reducing the feedback overhead caused by the reuse of support and scale information, the improvement of cascaded channel reconstruction accuracy is further transformed, and ultimately applied to reconfigurable intelligent surface phase configuration. This not only improves channel feedback and reconstruction performance, but also enhances the end-to-end communication performance of broadband orthogonal frequency division multiplexing systems.

[0079] In yet another specific embodiment, please refer to Figure 3The present invention also provides an overhead-sharing cascaded channel state information feedback system, including a system model construction module 31, a corner domain sparse representation module 32, a shared window partitioning module 33, a feedback budget decomposition module 34, a support and scale sharing module 35, an effective bit calculation module 36, a coefficient quantization and channel reconstruction module 37, and a phase configuration module 38. The system model construction module 31 constructs a reconfigurable intelligent surface communication system model and determines the system parameters; the corner domain sparse representation module 32 maps the cascaded channels of each user on each subcarrier to a corner domain dictionary to obtain the support set on each subcarrier; the shared window partitioning module 33 filters the dominant corner domain support sets on each subcarrier based on the support sets on each subcarrier; for the same user, the corner domain support similarity is calculated according to the dominant corner domain support sets between adjacent subcarriers, and the shared window is partitioned for adjacent subcarriers according to the corner domain support similarity and the shared window length constraint, and the shared support set of the user is determined in each shared window to obtain the completed shared window. The shared window length after partitioning; after completing the shared window partitioning, the feedback budget decomposition module 34 decomposes the feedback budget of each user on each subcarrier into support description overhead, scale description overhead, and coefficient quantization overhead; the support and scale sharing module 35 performs one-time feedback of support description information and scale description information within each shared window, and reuses it for multiple subcarriers within the shared window; the effective bit calculation module 36 calculates the effective coefficient quantization bits after overhead amortization based on the shared window length after partitioning, support description overhead, and scale description overhead; the coefficient quantization and channel reconstruction module 37 uses the effective coefficient quantization bits to quantize the complex coefficients in the angular domain support description information, and reconstructs the cascaded channel on each subcarrier based on the quantization result; the phase configuration module 38 performs reconfigurable smart surface phase configuration based on the reconstructed cascaded channel to obtain a reconfigurable smart surface phase vector, and uses the reconfigurable smart surface phase vector for data transmission in the broadband orthogonal frequency division multiplexing system.

[0080] In another specific embodiment, to verify the effectiveness of the overhead-sharing cascaded channel state information feedback method described in this invention, this embodiment sets up a reconfigurable smart surface-assisted broadband orthogonal frequency division multiplexing (OFDM) communication scenario and compares the method of this invention with the traditional subcarrier-by-subcarrier feedback method. The simulation system includes multiple users, multiple OFDM subcarriers, and multiple reconfigurable smart surface reflection units.

[0081] In simulations, traditional subcarrier-by-subcarrier feedback methods independently feed back support description information, scale description information, and complex coefficient quantization information on each subcarrier. The method of this invention, however, reuses support description information and scale description information within a shared window between adjacent subcarriers and quantizes the complex coefficients based on the effective number of quantization bits after overhead amortization. To comprehensively evaluate the performance of this invention, this embodiment uses cascaded channel reconstruction error and system average reachability as the main evaluation metrics.

[0082] Setting up a reconfigurable smart surface communication system model includes: With K subcarriers and K users, the cascaded channel reconstruction error is represented by the normalized mean square error, which is expressed as follows: in, This represents the true angular domain coefficient matrix for all users on the nth subcarrier. This represents the corresponding reconstructed corner field coefficient matrix. This represents the Frobenius norm.

[0083] Please see Figure 4 The figure shows waveforms comparing the cascaded channel reconstruction error under different feedback budgets. In the figure, the RD (Rate-Distortion) reference curve represents the theoretical lower bound of the corresponding channel source under lossy quantization, indicating the minimum normalized mean square error achievable with a given feedback bit budget. It is used to measure the difference between the proposed algorithm and the theoretical optimal limit. RD reference curve 1 and RD reference curve 2 correspond to the rate-distortion lower bounds under two different prior assumptions. When the shared window length... When the value is 1, it coincides with the curve of traditional subcarrier-by-subcarrier feedback. It can be seen that the overhead-amortized cascaded channel state information feedback method of this invention achieves a lower cascaded channel reconstruction error compared to the traditional subcarrier-by-subcarrier feedback method. Especially in the low feedback budget region, the traditional method, due to repeated payment of support description overhead and scale description overhead, leaves fewer effective bits for complex coefficient quantization, thus resulting in a larger reconstruction error. This invention, through overhead amortization within a shared window, reserves more effective bits for complex coefficient quantization, thereby significantly improving the accuracy of cascaded channel reconstruction.

[0084] Furthermore, to verify whether the improved accuracy of cascaded channel reconstruction can be translated into a gain in communication performance, this embodiment performs reconfigurable intelligent surface phase configuration based on the reconstructed cascaded channel and calculates the system's average achievable rate on a real cascaded channel. For user k, its average achievable rate is... It can be represented as: in, This represents the reconfigurable smart surface phase vector obtained from the reconstructed cascaded channel. This represents the cascaded channel of user k on the nth subcarrier. This indicates the received signal-to-noise ratio.

[0085] System average achievable rate It can be represented as: Please see Figure 5 The graph shows the average achievable rate of the system under different feedback budgets. The dashed line represents the ideal CSI, which is the upper bound of the performance of the ideal channel state information (CSI). The dotted line represents the random RIS, which is the lower bound of the performance of the random reconfigurable intelligent surface (RIS). This is based on the shared window length. When the value is 1, it coincides with the curve of traditional subcarrier-by-subcarrier feedback. Traditional subcarrier-by-subcarrier feedback only occurs above the critical feedback budget. Only when this is achieved will the system's average reachability significantly improve. Under different feedback budgets, this invention can achieve a higher system average reachability than the traditional subcarrier-by-subcarrier feedback method. This result demonstrates that this invention can not only reduce cascaded channel reconstruction errors but also improve the reconfigurable smart surface phase configuration effect through more accurate reconstruction of cascaded channels, thereby improving the end-to-end transmission performance of broadband orthogonal frequency division multiplexing systems.

[0086] Please see Figure 6 , Figure 6 The waveforms show the average reachability of the system under different numbers of users, where (a) represents the fixed total feedback budget. The waveforms comparing the average reachable rate of the system at different times are shown in (b), which represents the fixed total feedback budget. The waveforms comparing the average reachable rate of the system are shown. The purple dashed line represents the upper bound of the ideal channel state information (CSI) performance, and the green dotted line represents the lower bound of the reconfigurable intelligent surface (RIS) performance. In scenarios with an increasing number of users, the method of this invention has better multi-user scalability compared to the traditional subcarrier-by-subcarrier feedback method. As the number of users increases, the fixed total feedback resources are shared by more users. The repetitive fixed overhead in the traditional subcarrier-by-subcarrier feedback method further compresses the effective bits available for complex coefficient quantization for each user, leading to a more significant decrease in the average reachable rate of the system. This invention alleviates the resource pressure under the fixed feedback budget by reusing support information and scale information within the shared window, thus maintaining good system performance in multi-user scenarios.

[0087] In yet another specific embodiment, the present invention also provides a communication device, please refer to [link to relevant documentation]. Figure 7 It includes a controller 71 and a memory 72. The memory 72 stores a computer program, which, when executed by the controller 71, implements the aforementioned overhead-sharing cascaded channel state information feedback method.

[0088] As an example, the controller 71 may employ a cascaded channel state information feedback system with shared overhead, as described above.

[0089] In yet another embodiment, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a controller, implements the above-described overhead-sharing cascaded channel state information feedback method.

[0090] In summary, this embodiment demonstrates that the method described in this invention can reduce the overhead of repeated fixed feedback under the same feedback budget, improve the accuracy of cascaded channel reconstruction, and improve the phase configuration effect of reconfigurable smart surfaces based on reconfigurable channels. It is applicable to reconfigurable smart surface communication systems under limited feedback conditions.

[0091] This invention provides a cascaded channel state information feedback method, system, and device with overhead amortization. By dividing a shared window and reusing support description information and scale description information within the shared window, the repetitive fixed overhead in the subcarrier-by-subcarrier feedback process is reduced. The reuse of support and scale information increases the effective coefficient quantization bits, allowing for higher quantization accuracy of complex coefficients under the same feedback budget. This invention is applicable to multi-user broadband communication scenarios. When the number of users increases or the total feedback budget is limited, the overhead amortization mechanism within the shared window can alleviate feedback resource pressure in multi-user scenarios and improve the system's scalability under limited feedback conditions.

[0092] Although the present invention has been disclosed above by way of embodiments, it is not intended to limit the present invention. Anyone skilled in the art can make some modifications and refinements without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

Claims

1. A cascaded channel state information feedback method with cost-sharing, applied to a reconfigurable smart surface communication system, characterized in that, Includes the following steps: Step S1: Construct a reconfigurable intelligent surface communication system model and determine the system parameters in the reconfigurable intelligent surface communication system model according to the deployment scenario; the system parameters include the corner domain dictionary, the number of users, and the number of subcarriers; Step S2: Map the cascaded channels of each user on each subcarrier to the corner domain dictionary to obtain the support set on each subcarrier; Step S3 involves dividing the shared window based on the support set on each subcarrier to obtain the length of the shared window after the division. This includes the following steps: Step S31: Based on the support set on each subcarrier, filter to obtain the dominant angular domain support set on each subcarrier; Step S32: For the same user, calculate the angular domain support similarity based on the dominant angular domain support set between adjacent subcarriers; Step S33: Based on the corner domain support similarity and shared window length constraints, the adjacent subcarriers are divided into shared windows to obtain the shared window length after the shared window division is completed; Step S4: After completing the shared window partitioning, decompose the feedback budget of each user on each subcarrier into support description overhead, scale description overhead, and coefficient quantization overhead. Step S5: Within each shared window, the supporting description information and scale description information are fed back in one go and reused for multiple subcarriers within the shared window.

2. The overhead-sharing cascaded channel state information feedback method as described in claim 1, characterized in that, In step S2, the cascaded channels of each user on each subcarrier are mapped into the corner domain dictionary, as follows: in, This represents the cascaded channel of user k on the nth subcarrier, where A represents the corner domain dictionary. Represents the angular domain sparse coefficient vector of user k on the nth subcarrier; The support set is then represented as: in, Let k be the support set for user k on the nth subcarrier.

3. The overhead-sharing cascaded channel state information feedback method as described in claim 1, characterized in that, In step S3, the corner domain support similarity is expressed as: in, This indicates that user k is in the nth adjacent subcarrier and the nth... Corner domain support similarity between subcarriers and These represent the values ​​of user k on the nth subcarrier and the nth subcarrier, respectively. Dominant angular domain support set on each subcarrier.

4. The overhead-sharing cascaded channel state information feedback method as described in claim 1, characterized in that, In step S3, the method for dividing the shared window is as follows: when the average support similarity within the current shared window is greater than or equal to a preset support similarity threshold and the shared window length is less than the maximum shared window length, the current shared window continues to expand backward; when the two conditions are not simultaneously met, that is, when the average support similarity is less than the support similarity threshold and the shared window length is greater than or equal to the maximum shared window length, the expansion stops and the current shared window is determined, and the current shared window is determined as a shared window; the above process is repeated from the next ungrouped subcarrier until all subcarriers are grouped; The shared window length constraint refers to setting upper and lower boundary limits for the adaptively partitioned subcarrier shared window, including a minimum shared window length and a maximum shared window length. When expanding the window and determining whether to add adjacent subcarriers, the shared window length satisfies the following: In the formula, To share window length, To minimize the shared window length, This is the maximum shared window length.

5. The overhead-sharing cascaded channel state information feedback method as described in claim 1, characterized in that, Also includes: Step S6: Calculate the number of effective coefficient quantization bits after cost amortization based on the shared window length, support description overhead, and scale description overhead after the shared window partitioning is completed. Step S7: Quantize the complex coefficients in the corner domain support description information using the effective coefficients quantization bit number, and reconstruct the cascaded channel on each subcarrier based on the quantization result; Step S8: Based on the reconstructed cascaded channel, perform reconfigurable smart surface phase configuration to obtain a reconfigurable smart surface phase vector, and use the reconfigurable smart surface phase vector for data transmission in the broadband orthogonal frequency division multiplexing system.

6. The overhead-sharing cascaded channel state information feedback method as described in claim 1, characterized in that, The feedback budget satisfies: in, Let user k represent the feedback budget for the scale description overhead on the nth subcarrier. This indicates the overhead of the supporting description. Indicate the cost of scale description. This represents the coefficient quantization overhead used for complex coefficient quantization.

7. The overhead-sharing cascaded channel state information feedback method as described in claim 1, characterized in that, The scale description information is the shared dynamic range scale within the shared window, expressed as: in, This represents the scale of the shared dynamic range of user k in the g-th shared window. This indicates that user k is on the nth subcarrier. The complex coefficients corresponding to each support position and These represent the operations of extracting the real part and the imaginary part, respectively.

8. The overhead-sharing cascaded channel state information feedback method as described in claim 5, characterized in that, The effective coefficient quantization bit number is expressed as: in, Let represent the number of effective coefficient quantization bits for user k on the nth subcarrier, and b represent the feedback budget per user subcarrier. This indicates the number of bits used for scale description information. This indicates the length of the shared window after the shared window division is completed.

9. The overhead-sharing cascaded channel state information feedback method as described in claim 5, characterized in that, After obtaining the effective coefficient quantization bits after cost amortization, the complex coefficients on the shared support set are quantized. The complex coefficients at the support locations are decomposed into real and imaginary parts, as follows: in, This indicates the operation of taking the real part. This indicates the operation of taking the imaginary part. This indicates that user k is on the nth subcarrier. The complex coefficients corresponding to each support position.

10. A cascaded channel state information feedback system with cost-sharing, characterized in that, include: The system model building module constructs a model of the reconfigurable intelligent surface communication system and determines the system parameters. The angular domain sparse representation module maps the cascaded channels of each user on each subcarrier to the angular domain dictionary, and obtains the support set on each subcarrier; The shared window partitioning module, based on the support set on each subcarrier, filters to obtain the dominant angular domain support set on each subcarrier; For the same user, the corner domain support similarity is calculated based on the dominant corner domain support set between adjacent subcarriers. Based on the corner domain support similarity and the shared window length constraint, the adjacent subcarriers are divided into shared windows. The shared support set of the user is determined within each shared window, and the shared window length after the shared window division is obtained. The feedback budget decomposition module, after completing the shared window partitioning, decomposes the feedback budget of each user on each subcarrier into support description overhead, scale description overhead, and coefficient quantization overhead. The support and scale sharing module provides one-time feedback of support description information and scale description information within each sharing window, and reuses them for multiple subcarriers within the sharing window.

11. The overhead-sharing cascaded channel state information feedback system as described in claim 10, characterized in that, Also includes: The effective bit calculation module calculates the number of effective coefficient quantization bits after cost amortization based on the length of the shared window after the shared window partitioning, the support description overhead, and the scale description overhead. The coefficient quantization and channel reconstruction module uses the effective coefficient quantization bit number to quantize the complex coefficients in the corner domain support description information, and reconstructs the cascaded channel on each subcarrier based on the quantization result; The phase configuration module performs reconfigurable smart surface phase configuration based on the reconstructed cascaded channel to obtain a reconfigurable smart surface phase vector, which is then used for data transmission in a broadband orthogonal frequency division multiplexing system.

12. A communication device, comprising a controller and a memory, wherein the memory stores a computer program, characterized in that, When the computer program is executed by the controller, it implements a cascaded channel state information feedback method for overhead sharing as described in any one of claims 1 to 9.

13. The communication device as described in claim 12, characterized in that, The controller employs a cascaded channel state information feedback system with cost-sharing as described in any one of claims 10 to 11.

14. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by the controller, implements an overhead-sharing cascaded channel state information feedback method as described in any one of claims 1 to 9.

Images