RIS-aided mobile antenna cascaded channel construction method based on STRCS compressive sensing

Abstract

This invention discloses a method for constructing RIS-assisted mobile antenna cascaded channels based on STRCS compressed sensing. The high-dimensional and complex TM-RIS-RM cascaded channel estimation problem is decomposed into TM-RIS link channels and RIS-RM link channels. The AoD, AoA, and path response matrices of the TM-RIS link and RIS-RM link are calculated separately, and the TM-RIS link channels and RIS-RM link channels are synthesized into a TM-RIS-RM cascaded channel. By utilizing the inherent sparsity of wireless channels in the angular domain and applying compressed sensing theory, high-precision reconstruction of complete cascaded channel state information at any location within a continuous area can be achieved with a finite number of channel measurements. Optimizing the additional measurement positions of the mobile antenna improves the numerical stability of the channel and enhances the accuracy of the path response matrix.

Description

Technical Field

[0001] This invention relates to a method for constructing a RIS-assisted mobile antenna cascaded channel based on STRCS compressed sensing, belonging to the field of wireless communication technology. Background Technology

[0002] Multiple-input multiple-output (MIMO) technology significantly improves the capacity and reliability of wireless systems by utilizing spatial degrees of freedom. However, traditional fixed-location antenna systems may face deep fading on specific time-frequency resources, and the number of antennas is limited by physical space and the complexity of the RF chain. In recent years, mobile antenna technology, by allowing antennas to move mechanically within a continuous area, can dynamically adapt to channel changes, demonstrating the potential to improve channel capacity and energy efficiency. Meanwhile, Reflective Surfaces (RIS), as artificial electromagnetic surfaces composed of a large number of low-cost passive reflective elements, can intelligently reconstruct the wireless propagation environment, becoming a key enabling technology for improving coverage and spectral efficiency.

[0003] Combining a Resonant Radio Interchange (RIS) system with a mobile antenna to form a RIS-assisted mobile antenna system holds promise for further pushing the performance limits of wireless communication. However, the efficient operation of this system heavily relies on accurate channel state information. Unlike traditional single mobile antenna systems or RIS systems, RIS-assisted mobile antenna systems face the challenge of TM-RIS-RM cascaded channel estimation. This cascaded channel has high parameter dimensionality; using traditional joint estimation methods or dense sampling strategies would result in unacceptable pilot overhead and computational complexity. Furthermore, RIS systems typically lack active sensing capabilities, only reflecting signals, further increasing the difficulty of channel estimation.

[0004] While intelligent algorithms such as deep reinforcement learning have been used to optimize decision-making in such complex systems, their effective application still relies on obtaining relatively accurate channel information. Therefore, there is an urgent need for a method that can achieve efficient channel estimation for cascaded RIS-assisted mobile antenna systems with low pilot overhead. Summary of the Invention

[0005] This invention provides a method for constructing a RIS-assisted mobile antenna cascaded channel based on STRCS compressed sensing, which solves the problems disclosed in the background art.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows:

[0007] A RIS-assisted mobile antenna cascaded channel construction method based on STRCS compressed sensing:

[0008] The TM-RIS-RM cascaded channel to be constructed is decomposed into TM-RIS link channel and RIS-RM link channel by using a pre-constructed TM-RIS-RM communication model.

[0009] Calculate the AoD and AoA for the TM-RIS link and the RIS-RM link respectively;

[0010] Calculate the path response matrix of the TM-RIS link based on its AoD and AoA.

[0011] Calculate the path response matrix of the RIS-RM link based on its AoD and AoA.

[0012] The TM-RIS link channel is constructed based on the AoD, AoA, and path response matrix of the TM-RIS link.

[0013] The RIS-RM link channel is constructed based on the AoD, AoA, and path response matrix of the RIS-RM link.

[0014] The TM-RIS link channel and the RIS-RM link channel are combined into a TM-RIS-RM cascaded channel.

[0015] Furthermore, methods for calculating the AoD of the TM-RIS link include:

[0016] Move TM to M different positions in sequence. Activate a single RIS reflection unit, the RIS reflection unit is located at... The signal vector reaching the RIS is represented as:

[0017] ;

[0018] Where P is the transmit power of TM, This is the AoD guidance matrix for the TM-RIS link. Includes the AoD parameters to be determined. This is the path response matrix of the TM-RIS link. Let be the field response vector of RIS. , It is additive white Gaussian noise;

[0019] Discretize the continuous angle domain into an overcomplete dictionary. The AoD computation is transformed into a sparse recovery problem, that is, finding a... Satisfy min Output using OMP algorithm and corresponding All AoD pairs set , The pitch angle of the AoD of the TM-RIS link. The azimuth angle of the AoD of the TM-RIS link. For the multipath component of TM.

[0020] Furthermore, the methods for calculating the AoA of the TM-RIS link include:

[0021] Fixed TM at position Activate F-1 RIS reflection units sequentially, at positions of The received signal vector is represented as:

[0022] ;

[0023] Where P is the transmit power of TM, This is the AoA steering matrix for the TM-RIS link. Includes the AoA parameter to be determined. This is the path response matrix of the TM-RIS link. Let be the field response vector of TM. , It is additive white Gaussian noise;

[0024] Discretize the continuous angle domain into an overcomplete dictionary. The AoA computation is transformed into a sparse recovery problem, i.e., finding a... Satisfy min Output using OMP algorithm and corresponding All AoA pairs set , The pitch angle of the AoA of the TM-RIS link. This refers to the azimuth angle of the AoA of the TM-RIS link. To reach the multipath component of RIS.

[0025] Furthermore, methods for calculating the AoD of RIS-RM links include:

[0026] Activating different numbers of RIS reflection units, the set of RIS reflection unit locations is: Fix RM to At that point, the received signal vector is represented as:

[0027] = ;

[0028] in, This is the AoD guidance matrix for the RIS-RM link. Includes the AoD parameters to be determined. This is the path response matrix of the RIS-RM link. Let be the field response vector of RM.

[0029] , It is additive white Gaussian noise;

[0030] Discretize the continuous angle domain into an overcomplete dictionary. The AoD computation is transformed into a sparse recovery problem, that is, finding a... Satisfy min Output using OMP algorithm and corresponding All AoD pairs set , The pitch angle of the AoD in the RIS-RM link. The azimuth angle of the AoD of the RIS-RM link. This represents the multipath component of RIS reflection.

[0031] Furthermore, methods for calculating the AoA of a RIS-RM link include:

[0032] Activate a single RIS reflection unit at position 1 Move RM to N different positions in sequence. The received signal vector is represented as:

[0033] ;

[0034] in, This is the AoA guidance matrix for the RIS-RM link. Includes the AoA parameter to be determined. This is the path response matrix of the RIS-RM link. Let be the field response vector of RIS.

[0035] , It is additive white Gaussian noise;

[0036] Discretize the continuous angle domain into an overcomplete dictionary. The AoA computation is transformed into a sparse recovery problem, i.e., finding a... Satisfy min The OMP algorithm was used to estimate and corresponding All AoA pairs set , The pitch angle of the AoA of the RIS-RM link. The azimuth angle of the AoA of the RIS-RM link. To reach the multipath components of RM.

[0037] Furthermore, the method for calculating the path response matrix of the TM-RIS link based on its AoD and AoA includes:

[0038] The AoD and AoA of the TM-RIS link are parameters respectively. and The path response matrix of the TM-RIS link is vectorized as follows: ;

[0039] ;

[0040] in, It integrates initial and additional measurement signals. For the perception matrix, These are the additional measurement locations for the TM and RIS reflector units to be optimized, respectively.

[0041] By solving optimization problems To determine the optimal location for additional measurements;

[0042] The path response matrix vector is calculated using the least squares algorithm:

[0043] ;

[0044] And reconstructed into the path response matrix of the TM-RIS link. .

[0045] Furthermore, methods for calculating the path response matrix of the RIS-RM link based on its AoD and AoA include:

[0046] The AoD and AoA of the RIS-RM link are parameters respectively. and The path response matrix of the RIS-RM link is vectorized as follows: ;

[0047] ;

[0048] in, It integrates initial and additional measurement signals. For the perception matrix, and These are the sets of additional measurement locations for the RIS and RM systems to be optimized, respectively.

[0049] By solving optimization problems To determine the optimal location for additional measurements;

[0050] The path response matrix vector is calculated using the least squares algorithm:

[0051] .

[0052] And reconstruct the path response matrix of the RIS-RM link. .

[0053] Furthermore, the TM-RIS link channel is as follows:

[0054] ;

[0055] in, The TM field response vector, This is the RIS end-field response vector. This is the path response matrix of the TM-RIS link.

[0056] Furthermore, the channel modeling of the RIS-RM link is as follows:

[0057] ;

[0058] in, This is the RIS end-field response vector. The RM field response vector, This is the path response matrix of the RIS-RM link.

[0059] Furthermore, the TM-RIS-RM cascaded channel is as follows:

[0060] ;

[0061] in, For TM-RIS link channel, For RIS-RM link channels, Here is the reflection phase shift matrix of RIS.

[0062] The beneficial effects achieved by this invention are as follows: This invention decomposes the high-dimensional and complex TM-RIS-RM cascaded channel estimation problem into serial estimation of two lower-dimensional sub-channels (TM-RIS and RIS-RM). By utilizing the inherent sparsity of wireless channels in the angular domain and applying compressed sensing theory, high-precision reconstruction of complete cascaded channel state information at any location within a continuous region can be achieved with a finite number of channel measurements. By optimizing the additional measurement positions of the mobile antenna, the numerical stability of the channel is improved, and the accuracy of the path response matrix is ​​increased. Attached Figure Description

[0063] Figure 1 This is a schematic diagram of the RIS-assisted mobile antenna communication system involved in this invention.

[0064] Figure 2 This is a simulation diagram illustrating the impact of different mobile antenna placements on the normalized mean square error of TM-RIS link channel construction in an embodiment of the present invention.

[0065] Figure 3This is a simulation diagram illustrating the impact of different mobile antenna placements on the normalized mean square error of RIS-RM link channel construction in an embodiment of the present invention.

[0066] Figure 4 This is a simulation diagram illustrating the impact of different mobile antenna placements on the normalized mean square error of cascaded channel construction in an embodiment of the present invention. Detailed Implementation

[0067] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention.

[0068] This invention proposes a RIS-assisted mobile antenna cascaded channel construction method based on STRCS compressed sensing, which specifically includes the following steps:

[0069] Step 1: System and Channel Model Construction

[0070] like Figure 1 As shown, consider a RIS-assisted mobile antenna communication system, with each of the transmitter and receiver configured to operate in a two-dimensional continuous region. and The transmitting mobile antenna is denoted as TM, and the receiving mobile antenna is denoted as RM; a RIS containing F reflector elements is deployed in the middle, and its area is... Assuming , and All regions are square regions with a side length of A. The system adopts time-division duplex (TDD) mode and is based on the channel reciprocity assumption, that is, the channel state information estimated in the uplink pilot stage can be directly used for downlink beamforming.

[0071] TM transmits a signal at position t, which is reflected by RIS and received by RM at position r. The received signal model is as follows:

[0072] ;

[0073] in, Let P be the cascaded channel to be constructed, P be the transmit power, d=1 be the pilot signal, and z be additive white Gaussian noise.

[0074] Step 2: Cascaded channel parameterization based on field response information:

[0075] Cascaded channels Decomposed into TM-RIS link channels With RIS-RM link channel The cascaded structure utilizes field response information for parametric modeling, transforming the infinite-dimensional continuous spatial channel estimation problem into the calculation of finite FRI parameters (AoD, AoA, path response matrix).

[0076] The TM-RIS link channel is modeled as follows:

[0077] ;

[0078] in, This is the response vector of the TM. Let be the field response vector of RIS, whose elements are composed of phase shifts at corresponding positions; The path response matrix of the TM-RIS link contains A path response matrix with complex coefficients. and These are the number of multipath components reaching TM and RIS, respectively.

[0079] Similarly, the RIS-RM link channel model is as follows:

[0080] ;

[0081] in, This is the RIS end-field response vector. The RM field response vector, The path response matrix of the RIS-RM link contains A path response matrix with complex coefficients. and These are the number of multipath components after RIS reflection and RM reflection, respectively.

[0082] The complete channel can be reconstructed by estimating the FRI parameters (i.e., AoD / AoA and the path response matrix).

[0083] Step 3: Cascaded channel estimation based on STRCS:

[0084] Step 301: Calculate AoD (Angle of Emission) / AoA (Angle of Incidence):

[0085] Methods for calculating the AoD of a TM-RIS link include:

[0086] Move TM to M different positions in sequence. Activate a single RIS reflection unit, the RIS reflection unit is located at... The signal vector reaching the RIS is represented as:

[0087] ;

[0088] Where P is the transmit power of TM, This is the AoD guidance matrix for the TM-RIS link. Includes the AoD parameters to be determined. This is the path response matrix of the TM-RIS link. Let be the field response vector of RIS. , It is additive white Gaussian noise;

[0089] Discretize the continuous angle domain into an overcomplete dictionary. The AoD computation is transformed into a sparse recovery problem, that is, finding a... Satisfy min Output using OMP algorithm and corresponding All AoD pairs set , The pitch angle of the AoD of the TM-RIS link. The azimuth angle of the AoD of the TM-RIS link. For the multipath component of TM.

[0090] Methods for calculating the AoA of a TM-RIS link include:

[0091] Fixed TM at position Activate F-1 RIS reflection units sequentially, at positions of The received signal vector is represented as:

[0092] ;

[0093] Where P is the transmit power of TM, This is the AoA steering matrix for the TM-RIS link. Includes the AoA parameter to be determined. This is the path response matrix of the TM-RIS link. Let be the field response vector of TM. , It is additive white Gaussian noise;

[0094] Discretize the continuous angle domain into an overcomplete dictionary. The AoA computation is transformed into a sparse recovery problem, i.e., finding a... Satisfy min Output using OMP algorithm and corresponding All AoA pairs set , The pitch angle of the AoA of the TM-RIS link. This refers to the azimuth angle of the AoA of the TM-RIS link. To reach the multipath component of RIS.

[0095] Methods for calculating the AoD of a RIS-RM link include:

[0096] Activating different numbers of RIS reflection units, the set of RIS reflection unit locations is: Fix RM to At that point, the received signal vector is represented as:

[0097] = ;

[0098] in, This is the AoD guidance matrix for the RIS-RM link. Includes the AoD parameters to be determined. This is the path response matrix of the RIS-RM link. Let be the field response vector of RM.

[0099] , It is additive white Gaussian noise;

[0100] Discretize the continuous angle domain into an overcomplete dictionary. The AoD computation is transformed into a sparse recovery problem, that is, finding a... Satisfy min Output using OMP algorithm and corresponding All AoD pairs set , The pitch angle of the AoD in the RIS-RM link. The azimuth angle of the AoD of the RIS-RM link. This represents the multipath component of RIS reflection.

[0101] Methods for calculating the AoA of a RIS-RM link include:

[0102] Activate a single RIS reflection unit at position 1 Move RM to N different positions in sequence. The received signal vector is represented as:

[0103] ;

[0104] in, This is the AoA guidance matrix for the RIS-RM link. Includes the AoA parameter to be determined. This is the path response matrix of the RIS-RM link. Let be the field response vector of RIS.

[0105] , It is additive white Gaussian noise;

[0106] Discretize the continuous angle domain into an overcomplete dictionary. The AoA computation is transformed into a sparse recovery problem, i.e., finding a... Satisfy min The OMP algorithm was used to estimate and corresponding All AoA pairs set , The pitch angle of the AoA of the RIS-RM link. The azimuth angle of the AoA of the RIS-RM link. To reach the multipath components of RM.

[0107] Taking the AoD calculation of the TM-RIS link as an example, the calculation process of the above OMP algorithm is as follows:

[0108] (1) Input The corresponding received signal vector y and the iteration number K are used to initialize the residual. =y, corresponding to Zero vector, orthogonal projection matrix Subspace index set ;

[0109] (2) For k=1,2,…,K;

[0110] a. Calculation The (k-1)th atom in With current residual Find the dot product with the residual. Most relevant column .

[0111] b. Add the j-th column to the corresponding set S, i.e. Update the orthogonal projection matrix. and residual ;

[0112] c. Calculate the corresponding ;

[0113] (3) Output the corresponding And S.

[0114] Step 302: Path response matrix estimation and mobile antenna location optimization:

[0115] The AoD and AoA of the TM-RIS link are parameters respectively. and The path response matrix of the TM-RIS link is vectorized as follows: ;

[0116] ;

[0117] in, It integrates initial and additional measurement signals. For the perception matrix, These are the sets of additional measurement locations for TM and RIS, respectively, to be optimized;

[0118] By solving optimization problems To determine the optimal location for additional measurements; to minimize the condition number of the sensing matrix. This improves the stability of the estimation.

[0119] Perform N channel measurements at the optimized additional measurement location, where N must satisfy... To ensure that the equation is solvable;

[0120] The path response matrix vector is calculated using the least squares algorithm:

[0121] ;

[0122] And reconstructed into the path response matrix of the TM-RIS link. .

[0123] The AoD and AoA of the RIS-RM link are parameters respectively. and The path response matrix of the RIS-RM link is vectorized as follows: ;

[0124] ;

[0125] in, It integrates initial and additional measurement signals. For the perception matrix, and These are the sets of additional measurement locations for the RIS and RM systems to be optimized, respectively.

[0126] By solving optimization problems To determine the optimal additional measurement location; after obtaining the optimized additional measurement location, perform N channel measurements;

[0127] The path response matrix vector is calculated using the least squares algorithm:

[0128] ;

[0129] And reconstruct the path response matrix of the RIS-RM link. .

[0130] Step 303: Cascaded Channel Reconstruction

[0131] Using all the estimated FRI parameters, reconstruct the two sub-channels:

[0132] ;

[0133] ;

[0134] Finally, the complete cascaded channel estimation is as follows:

[0135] ;

[0136] in, For TM-RIS link channel, For RIS-RM link channels, Here is the reflection phase shift matrix of RIS.

[0137] To verify the effectiveness of the method proposed in this invention, a simulation experiment was conducted. The system parameters were set as follows: wavelength Side length of the transmit / receive area The multipath components at the transmitting end, both sides of the RIS end, and the receiving end are: = = = =3, initial number of mobile antenna sampling points M=N=256, number of RIS reflector elements F=256, OMP algorithm grid size G=200, additional position measurement times K=16. Four mobile antenna position layouts—UPA (Uniform Planar Array), cross-shaped, square, and circular—were compared.

[0138] like Figure 2 As shown, in the construction of the TM-RIS link channel, the UPA layout achieves the lowest normalized mean square error under different signal-to-noise ratios, significantly outperforming traditional least squares and least mean square error methods. This indicates that uniform spatial sampling can effectively improve estimation accuracy.

[0139] like Figure 3 As shown, in the construction of the RIS-RM link channel, the trend of the normalized mean square error changing with the signal-to-noise ratio is similar to... Figure 2 Similarly, the UPA layout also maintains optimal performance, further validating the importance of optimizing mobile antenna layout.

[0140] like Figure 4As shown, in the complete cascaded channel construction, due to error accumulation, the overall normalized mean square error is higher than that of a single link. However, the advantages of the UPA layout are still obvious, with its normalized mean square error being much lower than that of traditional methods and other non-uniform layouts. This proves that the method proposed in this invention can effectively control the overall error of cascaded channel estimation.

[0141] The simulation results above show that the method proposed in this invention can achieve high-precision channel estimation of RIS-assisted mobile antenna communication systems with low pilot overhead through channel decoupling, sparse recovery and position optimization. Among them, the mobile antenna position configuration with uniform planar array layout can bring the best performance.

[0142] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

[0143] The above are merely embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of the claims of the present invention pending approval.

Claims

1. A method for constructing a RIS-assisted mobile antenna cascaded channel based on STRCS compressed sensing, characterized in that: The TM-RIS-RM cascaded channel to be constructed is decomposed into TM-RIS link channel and RIS-RM link channel by using a pre-constructed TM-RIS-RM communication model. Calculate the AoD and AoA for the TM-RIS link and the RIS-RM link respectively; Calculate the path response matrix of the TM-RIS link based on its AoD and AoA. Calculate the path response matrix of the RIS-RM link based on its AoD and AoA. The TM-RIS link channel is constructed based on the AoD, AoA, and path response matrix of the TM-RIS link. The RIS-RM link channel is constructed based on the AoD, AoA, and path response matrix of the RIS-RM link. The TM-RIS link channel and the RIS-RM link channel are combined into a TM-RIS-RM cascaded channel.

2. The RIS-assisted mobile antenna cascaded channel construction method based on STRCS compressed sensing according to claim 1, characterized in that, Methods for calculating the AoD of a TM-RIS link include: Move TM to M different positions in sequence. Activate a single RIS reflection unit, the RIS reflection unit is located at... The signal vector reaching the RIS is represented as: ； Where P is the transmit power of TM, This is the AoD guidance matrix for the TM-RIS link. Includes the AoD parameters to be determined. This is the path response matrix of the TM-RIS link. Let be the field response vector of RIS. , It is additive white Gaussian noise; Discretize the continuous angle domain into an overcomplete dictionary. The AoD computation is transformed into a sparse recovery problem, that is, finding a... Satisfy min Output using OMP algorithm and corresponding All AoD pairs set , The pitch angle of the AoD of the TM-RIS link. The azimuth angle of the AoD of the TM-RIS link. For the multipath component of TM.

3. The RIS-assisted mobile antenna cascaded channel construction method based on STRCS compressed sensing according to claim 1, characterized in that, Methods for calculating the AoA of a TM-RIS link include: Fixed TM at position Activate F-1 RIS reflection units sequentially, at positions of The received signal vector is represented as: ； Where P is the transmit power of TM, This is the AoA steering matrix for the TM-RIS link. Includes the AoA parameter to be determined. This is the path response matrix of the TM-RIS link. Let be the field response vector of TM. , It is additive white Gaussian noise; Discretize the continuous angle domain into an overcomplete dictionary. The AoA computation is transformed into a sparse recovery problem, i.e., finding a... Satisfy min Output using OMP algorithm and corresponding All AoA pairs set , The pitch angle of the AoA of the TM-RIS link. This refers to the azimuth angle of the AoA of the TM-RIS link. To reach the multipath component of RIS.

4. The RIS-assisted mobile antenna cascaded channel construction method based on STRCS compressed sensing according to claim 1, characterized in that, Methods for calculating the AoD of a RIS-RM link include: Activating different numbers of RIS reflection units, the set of RIS reflection unit locations is: Fix RM to At that point, the received signal vector is represented as: = ； in, This is the AoD guidance matrix for the RIS-RM link. Includes the AoD parameters to be determined. This is the path response matrix of the RIS-RM link. Let be the field response vector of RM. , It is additive white Gaussian noise; Discretize the continuous angle domain into an overcomplete dictionary. The AoD computation is transformed into a sparse recovery problem, that is, finding a... Satisfy min Output using OMP algorithm and corresponding All AoD pairs set , The pitch angle of the AoD in the RIS-RM link. The azimuth angle of the AoD of the RIS-RM link. This represents the multipath component of RIS reflection.

5. The RIS-assisted mobile antenna cascaded channel construction method based on STRCS compressed sensing according to claim 1, characterized in that, Methods for calculating the AoA of a RIS-RM link include: Activate a single RIS reflection unit at position 1 Move RM to N different positions in sequence. The received signal vector is represented as: ； in, This is the AoA guidance matrix for the RIS-RM link. Includes the AoA parameter to be determined. This is the path response matrix of the RIS-RM link. Let be the field response vector of RIS. , It is additive white Gaussian noise; Discretize the continuous angle domain into an overcomplete dictionary. The AoA computation is transformed into a sparse recovery problem, i.e., finding a... Satisfy min The OMP algorithm was used to estimate and corresponding All AoA pairs set , The pitch angle of the AoA of the RIS-RM link. The azimuth angle of the AoA of the RIS-RM link. To reach the multipath components of RM.

6. The RIS-assisted mobile antenna cascaded channel construction method based on STRCS compressed sensing according to claim 1, characterized in that, Methods for calculating the path response matrix of a TM-RIS link based on its AoD and AoA include: The AoD and AoA of the TM-RIS link are parameters respectively. and The path response matrix of the TM-RIS link is vectorized as follows: ; ； in, It integrates initial and additional measurement signals. For the perception matrix, These are the additional measurement locations for the TM and RIS reflector units to be optimized, respectively. By solving optimization problems To determine the optimal location for additional measurements; The path response matrix vector is calculated using the least squares algorithm: ； And reconstructed into the path response matrix of the TM-RIS link. .

7. The RIS-assisted mobile antenna cascaded channel construction method based on STRCS compressed sensing according to claim 1, characterized in that, Methods for calculating the path response matrix of a RIS-RM link based on its AoD and AoA include: The AoD and AoA of the RIS-RM link are parameters respectively. and The path response matrix of the RIS-RM link is vectorized as follows: ; ； in, It integrates initial and additional measurement signals. For the perception matrix, and These are the sets of additional measurement locations for the RIS and RM systems to be optimized, respectively. By solving optimization problems To determine the optimal location for additional measurements; The path response matrix vector is calculated using the least squares algorithm: ； And reconstruct the path response matrix of the RIS-RM link. .

8. The RIS-assisted mobile antenna cascaded channel construction method based on STRCS compressed sensing according to claim 1, characterized in that, The TM-RIS link channel is: ； in, The TM field response vector, Let be the RIS field response vector. This is the path response matrix of the TM-RIS link.

9. The RIS-assisted mobile antenna cascaded channel construction method based on STRCS compressed sensing according to claim 1, characterized in that, The channel modeling for the RIS-RM link is as follows: ； in, Let be the RIS field response vector. The RM field response vector, This is the path response matrix of the RIS-RM link.

10. The RIS-assisted mobile antenna cascaded channel construction method based on STRCS compressed sensing according to claim 1, characterized in that, The TM-RIS-RM cascaded channel is: ； in, For TM-RIS link channel, For RIS-RM link channels, Here is the reflection phase shift matrix of RIS.

Images