Channel estimation method and device for ris-assisted millimeter wave system and storage medium

Abstract

The application provides a channel estimation method, device and storage medium of a RIS-assisted millimeter wave system. The method is applied to a network device and includes: receiving a pilot signal reflected by a smart reflecting surface (RIS); a beam pattern of the pilot signal is determined based on coherence of a sensing matrix; and channel estimation is performed based on the pilot signal by using a compressed sensing algorithm. The channel estimation method, device and storage medium of the RIS-assisted millimeter wave system provided in the application can determine a beam pattern based on coherence of a sensing matrix, thereby enabling channel estimation based on a compressed sensing algorithm and significantly reducing pilot overhead of channel estimation of a RIS-assisted millimeter wave multiple-input multiple-output system.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a channel estimation method, apparatus and storage medium for a RIS-assisted millimeter-wave system. Background Technology

[0002] Reconfigurable Intelligent Surfaces (RIS) can be applied to millimeter-wave (mmWave) multiple-input multiple-output (MIMO) systems to assist signal transmission, solving the problems of high hardware cost and high power consumption in mmWave MIMO systems. However, because RIS has a large number of passive components lacking signal processing capabilities, traditional channel estimation (CE) methods used in MIMO systems are not applicable to RIS-assisted mmWave MIMO systems.

[0003] Channel estimation for RIS-assisted millimeter-wave systems requires channel estimation at the receiver of a cascaded channel consisting of the channel from the transmitter to the RIS and the channel from the RIS to the receiver. However, existing channel estimation methods, such as the least-squares channel estimation method with binary reflection control, suffer from large pilot overhead. Summary of the Invention

[0004] This application provides a channel estimation method, apparatus, and storage medium for a RIS-assisted millimeter-wave system to solve the technical problem of large pilot overhead in channel estimation in the prior art.

[0005] In a first aspect, embodiments of this application provide a channel estimation method for a RIS-assisted millimeter-wave system, applied to network devices, comprising:

[0006] The pilot signal reflected by the intelligent reflective surface RIS is received; the beam pattern of the pilot signal is determined based on the coherence of the sensing matrix.

[0007] Channel estimation is performed using compressed sensing algorithms based on the pilot signals.

[0008] In some embodiments, the beam pattern of the pilot signal is determined based on the minimum coherence of the sensing matrix.

[0009] In some embodiments, channel estimation is performed based on the pilot signal using a compressed sensing algorithm, including:

[0010] The sparse vector is determined based on the pilot signal and the sensing matrix;

[0011] Based on the sparse vectors, the cascaded channel matrix is ​​recovered using a compressed sensing algorithm to obtain the channel estimation result.

[0012] In some embodiments, determining the sparse vector based on the pilot signal sensing matrix includes:

[0013] The sensing matrix is ​​determined based on the transmission beam matrix, the receiving beam matrix, and the dictionary matrices corresponding to the network devices and terminals, respectively.

[0014] The vectorized representation of the pilot signal is determined based on the sensing matrix;

[0015] The sparse vector is determined based on the vectorized representation of the pilot signal and the pilot signal.

[0016] In some embodiments, the pilot signal is a RIS-based reflection vector.

[0017] In some embodiments, each reflection vector is a column vector among the candidate reflection vectors that is different from the other column vectors;

[0018] The candidate reflection vector is determined based on the dictionary matrix associated with the RIS; the dictionary matrix associated with the RIS consists of the turning vectors of a pre-fixed number of grids of the target number; the target number is the number of elements corresponding to the uniform planar array equipped with the RIS.

[0019] Secondly, embodiments of this application provide a channel estimation method for a RIS-assisted millimeter-wave system, applied to a terminal, including:

[0020] The beam pattern of the pilot signal is determined based on the coherence of the sensing matrix.

[0021] The pilot signal is transmitted to the RIS based on the beam pattern; the pilot signal is used for channel estimation.

[0022] In some embodiments, determining the beam pattern of the pilot signal based on the coherence of the sensing matrix includes:

[0023] Determine the Kronecker product of the transmission beam matrix and the receiving beam matrix that minimizes the coherence of the sensing matrix;

[0024] The transmission beam matrix and the receiving beam matrix are calculated based on the Kronecker product of the transmission beam matrix and the receiving beam matrix to obtain the beam pattern of the pilot signal.

[0025] Thirdly, embodiments of this application provide a channel estimation apparatus for a RIS-assisted millimeter-wave system, comprising:

[0026] A receiving module is used to receive the pilot signal reflected by RIS; the beam pattern of the pilot signal is determined based on the coherence of the sensing matrix.

[0027] The channel estimation module is used to perform channel estimation based on the pilot signal using a compressed sensing algorithm.

[0028] In some embodiments, the beam pattern of the pilot signal is determined based on the minimum coherence of the sensing matrix.

[0029] In some embodiments, the channel estimation module includes:

[0030] The first determining unit is used to determine the sparse vector based on the pilot signal and the sensing matrix;

[0031] The recovery unit is used to recover the cascaded channel matrix based on the sparse vector using a compressed sensing algorithm, thereby obtaining the channel estimation result.

[0032] In some embodiments, the first determining unit includes:

[0033] The first determining subunit is used to determine the sensing matrix based on the transmission beam matrix, the receiving beam matrix, and the dictionary matrices corresponding to the network devices and terminals, respectively.

[0034] The second determining subunit is used to determine the vectorized representation of the pilot signal based on the sensing matrix;

[0035] The third determining subunit is used to determine a sparse vector based on the vectorized representation of the pilot signal and the pilot signal.

[0036] In some embodiments, the pilot signal is a RIS-based reflection vector.

[0037] In some embodiments, each reflection vector is a column vector among the candidate reflection vectors that is different from the other column vectors;

[0038] The candidate reflection vector is determined based on the dictionary matrix associated with the RIS; the dictionary matrix associated with the RIS consists of the turning vectors of a pre-fixed number of grids of the target number; the target number is the number of elements corresponding to the uniform planar array equipped with the RIS.

[0039] Fourthly, embodiments of this application provide a channel estimation apparatus for a RIS-assisted millimeter-wave system, comprising:

[0040] The determination module is used to determine the beam pattern of the pilot signal based on the coherence of the sensing matrix;

[0041] A transmitting module is used to transmit the pilot signal to the RIS based on the beam pattern; the pilot signal is used for channel estimation.

[0042] In some embodiments, the determining module includes:

[0043] The second determining unit is used to determine the Kronecker product of the transmission beam matrix and the receiving beam matrix when the coherence of the sensing matrix is ​​minimized.

[0044] The calculation unit is used to calculate the transmission beam matrix and the receiving beam matrix based on the Kronecker product of the transmission beam matrix and the receiving beam matrix, so as to obtain the beam pattern of the pilot signal.

[0045] Fifthly, embodiments of this application provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the channel estimation method for a RIS-assisted millimeter-wave system as described in the first or second aspect above.

[0046] In a sixth aspect, embodiments of this application also provide a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the channel estimation method for the RIS-assisted millimeter-wave system as described in the first or second aspect above.

[0047] In a seventh aspect, embodiments of this application also provide a computer program product, including a computer program that, when executed by a processor, implements the channel estimation method for a RIS-assisted millimeter-wave system as described in the first or second aspect above.

[0048] The channel estimation method, apparatus, and storage medium for RIS-assisted millimeter-wave systems provided in this application receive pilot signals reflected by the RIS. The beam pattern of the pilot signals is designed based on the coherence of the sensing matrix. Channel estimation is performed using compressed sensing algorithms based on the pilot signals, enabling high-performance channel estimation with fewer pilot signals and saving pilot overhead. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 This is one of the flowcharts illustrating the channel estimation method for a RIS-assisted millimeter-wave system provided in the embodiments of this application;

[0051] Figure 2This is a schematic diagram of the pilot block transmission process in an example scenario provided in the embodiments of this application;

[0052] Figure 3 This is a schematic diagram illustrating the derivation process of the received signal representation in the example scenario provided in the embodiments of this application;

[0053] Figure 4 This is the second flowchart illustrating the channel estimation method for a RIS-assisted millimeter-wave system provided in this application embodiment;

[0054] Figure 5 This is one of the structural schematic diagrams of a channel estimation device for a RIS-assisted millimeter-wave system provided in the embodiments of this application;

[0055] Figure 6 This is a second schematic diagram of the structure of a channel estimation device for a RIS-assisted millimeter-wave system provided in an embodiment of this application;

[0056] Figure 7 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0057] A reconfigurable intelligent surface (RIS), also known as an intelligent reflecting surface (IRS), is an electromagnetic metasurface composed of passive reflecting units that can passively reflect signals through programmable surface control. Existing techniques, at the cost of extremely high computational complexity, have proposed a least-squares (LS) channel estimation method with binary reflection control.

[0058] To reduce pilot overhead and computational complexity, compressed sensing (CS) algorithms are employed in Multiple-Input Single-Output (MISO) systems. However, CS-based channel estimation methods exhibit poor performance. For MISO systems, an optimized channel estimator with a closed-form solution is proposed using the typical mean square error (MSE) criterion. However, existing solutions only consider a single antenna at the receiver and do not account for the high dimensionality of MIMO systems. Therefore, methods for reducing pilot overhead in MISO systems cannot be directly applied to MIMO systems, resulting in high pilot overhead for channel estimation in RIS-assisted mmWave MIMO systems.

[0059] Based on the above-mentioned technical problems, this application proposes a channel estimation method for a RIS-assisted millimeter-wave system. The method determines the beam pattern of the pilot signal based on the coherence of the sensing matrix and transmits the pilot signal based on the beam pattern. This enables channel estimation to be completed using a compressed sensing algorithm based on the pilot signal, thereby reducing pilot overhead.

[0060] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0061] Figure 1 This is one of the flowcharts illustrating the channel estimation method for a RIS-assisted millimeter-wave system provided in this application embodiment, such as... Figure 1 As shown, this application provides a channel estimation method for a RIS-assisted millimeter-wave system, the execution entity of which can be a network device, such as a base station. The method includes:

[0062] Step 101: Receive the pilot signal reflected by the intelligent reflective surface RIS; the beam pattern of the pilot signal is determined based on the coherence of the sensing matrix.

[0063] Specifically, the method is applied to channel estimation scenarios in RIS-assisted mmWave MIMO systems. A terminal or user terminal acts as the transmitter of the pilot signal, and a network device, such as a base station, acts as the receiver. After determining the beam pattern of the pilot signal, the terminal transmits the pilot signal to a smart reflector based on the determined beam pattern. The smart reflector receives the pilot signal transmitted by the user terminal, processes it, and then reflects / transmits the pilot signal to the network device.

[0064] In this embodiment, the terminal determines the beam pattern of the pilot signal based on the coherence of the sensing matrix. For example, the terminal determines the beam pattern of the pilot signal corresponding to a certain level of coherence of the sensing matrix, and in particular, determines the beam pattern of the pilot signal corresponding to the minimum coherence of the sensing matrix.

[0065] For example, the terminal calculates the coherence of the sensing matrix corresponding to the preset beam pattern, then adjusts the beam pattern, and recalculates the coherence of the corresponding sensing matrix using the adjusted beam pattern. The beam pattern is continuously adjusted and optimized within a preset number of optimizations until a lower level of sensing matrix coherence is obtained. The beam pattern corresponding to the lower level of sensing matrix coherence is the final beam pattern of the determined pilot signal.

[0066] For example, the terminal first determines the Kronecker product of the transmission beam matrix and the receiving beam matrix when the coherence of the sensing matrix is ​​minimized, and then calculates the transmission beam matrix and the receiving beam matrix based on the Kronecker product of the transmission beam matrix and the receiving beam matrix to obtain the beam pattern of the pilot signal.

[0067] For example, the relationship between the coherence of the sensing matrix and the digital beam representation (determined by the Kronecker product of the transmit and receive beam matrices) is determined. The total coherence is minimized when the digital beam representation satisfies the target condition. Then, the optimal digital beam representation is solved based on this minimum coherence value. The transmit and receive beam matrices are determined based on this optimal digital beam representation. That is, the transmit and receive beam matrices are approximately solved based on the Kronecker product of the transmit and receive beam matrices, thereby obtaining the final design of the pilot signal beam pattern.

[0068] Optionally, the RIS is based on the reflection of pilot signals by reflection vectors. There are one or more reflection vectors, each reflection vector corresponds to multiple transmitted pilot blocks, and each pilot block corresponds to multiple pilots / beams.

[0069] Step 102: Perform channel estimation using compressed sensing algorithm based on the pilot signal.

[0070] Specifically, the network device receives the pilot signal reflected by the RIS (Receiving Signal), i.e., the received signal, and uses a compressed sensing algorithm to reconstruct the cascaded channel, thus completing channel estimation. The cascaded channel is calculated from the channel from the RIS to the base station (BS) and the channel from the user terminal to the RIS.

[0071] The channel estimation method for RIS-assisted millimeter-wave systems provided in this application involves a terminal designing the beam pattern of pilot signals based on the coherence of the sensing matrix, and a network device performing channel estimation using a compressed sensing algorithm based on the received pilot signals. This improves the performance of channel estimation and saves pilot overhead.

[0072] In some embodiments, the beam pattern of the pilot signal is determined based on the minimum coherence of the sensing matrix.

[0073] Specifically, the beam pattern of the pilot signal is determined based on the minimum coherence of the sensing matrix. When the network side performs channel estimation based on this pilot signal, since the coherence of the sensing matrix corresponding to the pilot signal reaches its minimum, the number of pilot signals required to reconstruct the cascaded channel using the compressed sensing algorithm is reduced.

[0074] Optionally, the terminal first determines the Kronecker product of the transmission beam matrix and the receiving beam matrix when the coherence of the sensing matrix is ​​minimized, and then calculates the transmission beam matrix and the receiving beam matrix based on the Kronecker product of the transmission beam matrix and the receiving beam matrix to obtain the beam pattern of the pilot signal.

[0075] For example, the relationship between the coherence of the sensing matrix and the digital beam representation is determined, the total coherence is minimized when the digital beam representation satisfies the target condition, and then the optimal digital beam representation is solved based on this minimum coherence value. The transmission beam matrix and the receiving beam matrix are determined according to the optimal digital beam representation, that is, the transmission beam matrix and the receiving beam matrix are approximately solved based on the Kronecker product of the transmission beam matrix and the receiving beam matrix, thereby obtaining the final design of the pilot signal beam pattern.

[0076] The channel estimation method for RIS-assisted millimeter-wave systems provided in this application determines the beam pattern of the pilot signal based on the minimum coherence of the sensing matrix, thereby maximizing the performance of channel estimation based on the pilot signal under this beam pattern and minimizing the pilot overhead required to achieve channel estimation.

[0077] In some embodiments, channel estimation is performed based on the pilot signal using a compressed sensing algorithm, including:

[0078] The sparse vector is determined based on the pilot signal and the sensing matrix;

[0079] Based on the sparse vectors, the cascaded channel matrix is ​​recovered using a compressed sensing algorithm to obtain the channel estimation result.

[0080] Specifically, after receiving the pilot signal, the network device calculates a sparse vector based on the pilot signal, the sensing matrix, and the noise matrix. Based on the sparse vector, a compressed sensing algorithm is used to recover the cascaded channel matrix, i.e., to reconstruct the cascaded channel, thus obtaining the channel estimation result.

[0081] Optionally, when the beam pattern of the pilot signal is determined by minimizing the coherence of the sensing matrix, the sensing matrix calculated by the network device using the pilot signal has very low coherence. Therefore, a sparse vector is determined based on the pilot signal and the sensing matrix, and the performance of recovering the cascaded channel based on the sparse vector is maximized. Thus, high performance of channel estimation can be guaranteed with less pilot overhead.

[0082] The channel estimation method for RIS-assisted millimeter-wave systems provided in this application transforms channel estimation into a sparse vector recovery problem by determining sparse vectors and using compressed sensing algorithms to recover the cascaded channel matrix. The application of compressed sensing algorithms reduces computational complexity.

[0083] In some embodiments, determining the sparse vector based on the pilot signal sensing matrix includes:

[0084] The sensing matrix is ​​determined based on the transmission beam matrix, the receiving beam matrix, and the dictionary matrices corresponding to the network devices and terminals, respectively.

[0085] The vectorized representation of the pilot signal is determined based on the sensing matrix;

[0086] The sparse vector is determined based on the vectorized representation of the pilot signal and the pilot signal.

[0087] Specifically, the sensing matrix is ​​calculated from the transmission beam matrix, the receiving beam matrix, the dictionary matrix corresponding to the network device, and the dictionary matrix corresponding to the terminal. Based on the sensing matrix, the vectorized representation of the pilot signal (i.e., the vectorized representation of the received signal) can be optimized or simplified. After the network device receives the pilot signal reflected by the RIS, i.e., the received signal, a sparse vector is calculated based on the received signal using the optimized vectorized representation of the received signal.

[0088] For example, defining a perception matrix Where F k，t W is the matrix representation of the transmission beam of the t-th pilot block with the k-th reflection vector. k，t Let T be the matrix representation of the received beam of the t-th pilot block of the k-th reflection vector, where T represents the transpose operation and H represents the conjugate operation. Indicates the Kronecker product; Indicates the user's A dictionary matrix consisting of the turning vectors of a pre-fixed grid. The BS indicates that it is made by A dictionary matrix consisting of the turning vectors of a pre-fixed grid, wherein, and All are positive integers, and * indicates the operation of solving the adjoint matrix. Based on the optimization of the sensing matrix, the vectorized representation of the received signal is obtained as follows:

[0089]

[0090] Among them, y k，t λ is the vectorized representation of the received signal in the t-th pilot block of the k-th reflection vector; k Let n represent the sparse vector corresponding to the k-th reflection vector. k，t Let $\mathbf{k}$ be the vector representation of the Gaussian white noise corresponding to the $t$-th pilot block of the $k$-th reflection vector. The sparse vector is obtained based on the received signal and the above equation.

[0091] The channel estimation method for RIS-assisted millimeter-wave systems provided in this application simplifies the vectorized representation of pilot signals using a sensing matrix, thereby reducing the computational complexity of channel estimation.

[0092] In some embodiments, the pilot signal is based on reflection vector reflection.

[0093] Each reflection vector is a column vector among the candidate reflection vectors that is different from the other column vectors;

[0094] The candidate reflection vector is determined based on the dictionary matrix associated with the RIS; the dictionary matrix associated with the RIS consists of the turning vectors of a pre-fixed number of grids of the target number; the target number is the number of elements corresponding to the uniform planar array equipped with the RIS.

[0095] Specifically, after receiving the pilot signal from the terminal, the RIS reflects the pilot signal based on the reflection vector. The column vector that is different from other column vectors refers to a candidate reflection vector that does not have a matching column vector. The candidate reflection vectors are determined based on the dictionary matrix associated with the RIS.

[0096] For example, the uniform planar array equipped with RIS has M elements, and the dictionary matrix of RIS, consisting of the orientation vectors of M pre-fixed grids, is V. M Based on V M Sure in, This represents the KR product. Then, the first M rows of D are determined to obtain... Will The inverse of is taken as the candidate reflection vector.

[0097] The channel estimation method for RIS-assisted millimeter-wave systems provided in this application embodiment designs a beam pattern based on the coherence of the sensing matrix. After the network device receives the pilot signal reflected by the RIS based on the beam pattern, it performs channel estimation using a compressed sensing algorithm based on the pilot signal, thus saving pilot overhead.

[0098] The channel estimation methods for RIS-assisted millimeter-wave systems provided in the above embodiments are further illustrated below with specific examples:

[0099] Establish a RIS-assisted millimeter-wave MIMO system model, with the transmitter being the user terminal or terminal and the receiver being the base station. Consider a single-user MIMO system with a carrier wavelength of λ, where the base station (BS) is equipped with N... r A uniform linear array (ULA) of N antennas, with the user end equipped with N tThe RIS is a uniform linear array of antennas, equipped with a uniform planar array (UPA) of M elements, where M is a positive integer and M = M x ×M y M x M represents the number of reflection cells in the RIS direction. y This represents the number of reflection cells of RIS in the y-direction.

[0100] A narrowband geometric channel model is used to characterize the RIS-to-BS channel and the user-to-RIS channel. Specifically, the matrix representation of the RIS-to-BS channel is as follows:

[0101]

[0102] Where G is the matrix representation of the channel from RIS to BS; M is the number of elements equipped in RIS and M = M x ×M y N r L1 indicates the number of antennas equipped on the BS; L1 indicates the number of paths between the BS and the RIS. Represents the complex gain of the transmission path; This represents the arrival angle (AoA) of the l1st path at BS; and These represent the azimuth and elevation angles of the departure angle (AoD) of the l1st path at RIS, respectively. Indicates the angle of arrival at BS. The varying normalized array steering vector; Indicates the azimuth angle at RIS. and elevation angle The varying normalized array steering vector; H denotes the conjugate operation. Similarly, the matrix representation of the user-to-RIS channel is:

[0103]

[0104] Where R is the matrix representation of the channel from the user to the RIS; M is the number of elements equipped in the RIS and M = M x ×M y N t L1 represents the number of antennas equipped at the user terminal; L2 represents the number of paths between the RIS and the user. This represents the complex gain of the l2th transmission path; This represents the departure angle (AoD) of the l2nd path at the user's location; and These represent the azimuth and elevation angles of the arrival angle (AoA) of the l2nd path at RIS, respectively. Indicates the azimuth angle at RIS. and elevation angle The varying normalized array steering vector; Indicates the angle at which the user leaves. The varying normalized array steering vector.

[0105] The normalized array steering vector of UPA can be expressed as:

[0106]

[0107] in, This represents the normalized array steering vector of the UPA; and These represent the azimuth and elevation angles of the transmission path at point AoA on the RIS, respectively; M represents the number of elements equipped on the RIS, and M = M x ×M y M x M represents the number of reflective units in the RIS along the x-direction. y The number of reflecting elements of RIS in the y-direction is represented by ; i is an integer; κ = 2πd, where d represents the antenna spacing, and d takes the value of λ / 2, where λ is the carrier wavelength; x1 represents the first intermediate vector, and x1 = [0, 1, 2, ..., M]. x -1] T An element in x1 multiplied by d represents the reflective unit spacing of a RIS in the x-direction; x2 represents the second intermediate vector, and x2 = [0, 1, 2, ..., M]. y -1] T In x2, an element multiplied by d represents the spacing of the reflective cells of a RIS in the y-direction.

[0108] The normalized array steering vector of ULA at BS can be expressed as:

[0109]

[0110] Where, a(θ) r ) represents the normalized array steering vector of ULA at BS; θ r Indicates the angle of arrival of the transmission path at BS; N r The number of antennas equipped on the BS is indicated; i is an integer; κ = 2πd, where d represents the antenna spacing, and d takes the value of λ / 2, where λ is the carrier wavelength; x3 represents the third intermediate vector, and x3 = [0, 1, 2, ..., N]. r -1] T In x3, an element multiplied by d represents an antenna spacing.

[0111] The normalized array steering vector of the ULA at the user site can be expressed as:

[0112]

[0113] Where, a(θ) u ) represents the normalized array steering vector of the ULA at the user location; θ u N represents the angle of arrival of the transmission path at the user; t This represents the number of antennas equipped at the user terminal; i is an integer; κ = 2πd, where d represents the antenna spacing, and d takes the value of λ / 2, where λ is the carrier wavelength; x4 represents the fourth intermediate vector, and x4 = [0, 1, 2, ..., N]. t -1] T In x4, an element multiplied by d represents an antenna spacing.

[0114] G and R can be decomposed into:

[0115]

[0116] in, The BS indicates that it is made by A dictionary matrix consisting of steering vectors of a pre-fixed grid; RIS indicates that it is composed of A dictionary matrix consisting of the turning vectors of a pre-fixed grid; Indicates the user's A dictionary matrix consisting of the turning vectors of a pre-fixed grid; This is the matrix representation of the L1 sparse beam spatial channel corresponding to G. Let be the matrix representation of the L2 sparse beam spatial channel corresponding to R; H denotes the conjugate operation.

[0117] Assumption (This situation can easily be extended to the general case), where M x M represents the number of reflection cells in the RIS direction. y Let represent the number of reflection cells of RIS in the y-direction. All spatial angles lie on a uniform grid from -1 to 1. Therefore, the matrix representation of the uplink cascaded channel is defined as:

[0118]

[0119] Where H is the matrix representation of the uplink concatenated channel; G is the matrix representation of the channel from RIS to BS; and diag() represents the function to calculate the diagonal matrix. Let R represent the phase shift vector at RIS; R is the matrix representation of the channel from the user to RIS. Substituting (6) into (7) yields:

[0120]

[0121] Where H is the matrix representation of the uplink concatenated channel; The BS indicates that it is made by A dictionary matrix consisting of the steering vectors of a pre-fixed grid; Γ is the matrix representation of the L1 sparse beam space channel corresponding to G; V M RIS indicates that it is composed of A dictionary matrix consisting of the steering vectors of a pre-fixed grid; H represents the conjugate operation; Ψ represents the phase shift vector at RIS; ∑ is the matrix representation of the L2 sparse beam space channel corresponding to R; Indicates the user's A dictionary matrix consisting of the turning vectors of a pre-fixed grid;

[0122] Figure 2 This is a schematic diagram of the pilot block transmission process in an example scenario provided in the embodiments of this application, such as... Figure 2 As shown, the transmission has K different RIS reflection vector designs. Channel estimation (CE) and data transmission (DT) are performed within the coherence time (the period during which the channel information does not change significantly). This corresponds to the design of a reflection vector Ψ. k There is T k One pilot block is transmitted, and the corresponding cascaded channel is... Within the t-th pilot block, One beam is formed at BS. A beam is formed at the user's location, where t = 1, 2, ..., T k , so A single pilot signal can be transmitted within a pilot block. Precoding matrix. (corresponding to the transmitter) and the combined matrix (Corresponding to the receiving end) together represent the beamform design of the k-th reflection vector design / the t-th pilot block in the reflection vector. Therefore, the received signal y k,t,p (i.e., the received signal) corresponds to the p-th transmission beam f in the t-th pilot block within the k-th reflection vector of the user. k，t，p ,in, The received signal can be represented as:

[0123]

[0124] Among them, y k,t,p It is a matrix representation of the received signal transmitted by the p-th transmission beam in the t-th pilot block within the k-th reflection vector; H is the combination matrix of the t-th pilot block in the k-th reflection vector (the matrix representation of the receiving beam);k f is the matrix representation of the uplink concatenated channel corresponding to the k-th reflection vector; k，t，p The matrix representation of the p-th transmission beam in the t-th pilot block within the k-th reflection vector; s k，t，p Let be the matrix representation of the pilot signal transmitted by the p-th transmission beam in the t-th pilot block within the k-th reflection vector, and |s k，t，p |=1;W k This represents the assemblage matrix corresponding to the k-th reflection vector; This is a matrix representation of Gaussian white noise. All [noise] is collected within the pilot block. The received signal can be represented as: (The pilot signal is transmitted, and the signal is received as follows:)

[0125]

[0126] in, Let be the matrix representation of the received signal in the t-th pilot block of the k-th reflection vector. This is the matrix representation of the transmission beam in the t-th pilot block within the k-th reflection vector; H represents the noise matrix; k Let be the matrix representation of the uplink cascaded channel corresponding to the k-th reflection vector. The number of measurements designed for the k-th reflection vector is... Among them, T k This represents the number of pilot blocks within the k-th reflection vector; This indicates the number of beams formed at the user's location; This indicates the number of beams formed at BS.

[0127] The reflection vector is Ψ k The structured matrix representation of the cascaded channel is as follows:

[0128]

[0129] Among them, vec(H k ) is the reflection vector Ψ k The structured matrix representation of cascaded channels; Indicates the user's A dictionary matrix consisting of the turning vectors of a pre-fixed grid; * indicates solving the adjoint matrix operation; The BS indicates that it is made by A dictionary matrix consisting of the turning vectors of a pre-fixed grid; This is the matrix representation of the L1 sparse beam spatial channel corresponding to G; Here is the matrix representation of the L2 sparse beam spatial channel corresponding to R; T denotes the transpose operation; Ψ k This represents the k-th reflection vector; It is the result of the KR (Khatri-Rao) product. The mixed product property of the Kronecker product is utilized in process (*). (11) can be simplified to:

[0130]

[0131] Among them, vec(H k ) is the reflection vector Ψ k The structured matrix representation of cascaded channels; Indicates the user's A dictionary matrix consisting of the turning vectors of a pre-fixed grid; The BS indicates that it is made by A dictionary matrix consisting of the turning vectors of a pre-fixed grid; It is the M before D G Row-matrix representation. It is a merged version of J, and its expression is as follows:

[0132]

[0133] Among them, s i It is the set of indices in D that have the same row as the i-th row; Based on expression (12), the received signal in (10) can be further represented in vector form as follows:

[0134]

[0135] In process (a), the property of the mixed product of the Kronecker product was used, and H is the vectorized representation of the received signal transmitted by all transmitted beams in the t-th pilot block within the k-th reflection vector; k F is the matrix representation of the uplink concatenated channel corresponding to the k-th reflection vector; k，t W is the matrix representation of the transmitted beam in the t-th pilot block within the k-th reflection vector; k，t Let be the combination matrix of the t-th pilot block in the k-th reflection vector; This is the matrix representation of the Gaussian white noise corresponding to the t-th pilot block within the k-th reflection vector. Process (b) utilizes the result of expression (12). In process (c), the mixed product property of the Kronecker product is utilized again to combine... get Considering The sparsity of the channel allows for channel estimation using compressed sensing algorithms. However, The enormous dimensionality of the array still results in high computational complexity, hindering practical applications. To reduce... The huge computational complexity allows us to further simplify expression (14) to obtain an optimized expression:

[0136]

[0137] Among them, y k，t F is the vectorized representation of the received signal transmitted by all transmission beams in the t-th pilot block within the k-th reflection vector; k，t W is the matrix representation of the transmitted beam in the t-th pilot block within the k-th reflection vector; k，t It is the combination matrix of the t-th pilot block in the k-th reflection vector (the matrix representation of the receiving beam); Indicates the user's A dictionary matrix consisting of the turning vectors of a pre-fixed grid; The BS indicates that it is made by A dictionary matrix consisting of the turning vectors of a pre-fixed grid; λ k Indicates merger The result after; n k，t Let be the matrix representation of the Gaussian white noise corresponding to the t-th pilot block within the k-th reflection vector;

[0138] Figure 3 This is a schematic diagram illustrating the derivation process of the received signal representation in an example scenario provided in this application embodiment. The derivation process of process (d) above is as follows: Figure 3 As shown, assume M a =4, in expression (14) (c) λ of (d) in expression (15) k After the merger It can be represented as:

[0139]

[0140] in, It is the M before D G Row matrix representation; Ψ k This represents the k-th reflection vector; Indicates merger The result afterward; This indicates the number of pre-defined grid turning vectors used by the user; This represents the number of steering vectors in the pre-fixed grid of the BS. Assume the number of pilot blocks T within the k-th reflection vector is... k =T, the number of measurements Q for the k-th reflection vector k =Q, where k = 1, 2, ...; K. In the process (e) of expression (15), by defining The equations are simplified. The matrix representation of the received signals obtained by collecting all signals received by pilot block T in expression (15) is as follows:

[0141] y k =Q k λ k +n k (17)

[0142] in, Let λ be the matrix representation of all received signals in pilot block T; k Indicates merger The result afterward; when When the phase shift vector is sparse, channel estimation can be performed based on the CS algorithm. After the channel estimation process involving K reflection combinations, data transmission occurs during the remaining coherence time. The phase shift vector Ψ used satisfies the following expression:

[0143]

[0144] in, It is the M before D G The matrix representation of rows; Ψ represents the reflection vector or phase shift vector used at RIS; K represents the number of reflection vectors at RIS; β k Ψ represents the coefficient corresponding to the k-th reflection vector; k Let represent the k-th reflection vector. The matrix representation of the uplink concatenated channel used for data transmission is as follows:

[0145]

[0146] Where H is the matrix representation of the uplink concatenated channel; Indicates the user's A dictionary matrix consisting of the turning vectors of a pre-fixed grid; The BS indicates that it is made by A dictionary matrix consisting of the turning vectors of a pre-fixed grid; The coefficients need to be estimated at the CE stage. It can be determined by the phase shift vector Ψ in expression (18). K reflection vectors Ψ k The reflection mode or phase shift vector Ψ used in data transmission can be represented as a linear mapping by expression (18).

[0147] Generally, reflection modes can be designed to maximize the received power to increase the received signal-to-noise ratio or minimize differences in λ. kThe coherence between them. Obviously, not all RIS reflection vectors Ξ that satisfy expression (18) will correspond to a sparse A. Since the CS algorithm can only be applied while ensuring the sparsity of λ, the Ξ of the K reflection vectors can be designed to select The K unique columns are as follows:

[0148]

[0149] Where Ξ represents the reflection vector; It contains M candidate reflection mode designs (i.e., candidate reflection vectors); express The inverse of. According to expression (16), With only one non-zero element, we know that Λ is very sparse. Generally, and At that time, in order to ensure that the phase shift characteristics satisfy the following equation in, yes The false reversal. When hour, Need to meet Similarly, when hour, Need to meet

[0150] According to expression (20), when When K = M, further simplification of expression (19) yields the matrix representation of the uplink cascaded channel as follows:

[0151]

[0152] Where H is the matrix representation of the uplink concatenated channel; Indicates the user's A dictionary matrix consisting of the turning vectors of a pre-fixed grid; The BS indicates that it is made by A dictionary matrix consisting of the orientation vectors of a pre-fixed grid; Λ represents a sparse matrix; It is the M before D G The matrix representation of the row; Ψ represents the phase shift vector at RIS.

[0153] Similarly, the above expression for the channel also applies to the case where K < M.

[0154] Based on CS theory, reduce the perceptual matrix Q kThe coherence can improve the performance of sparse signal recovery. According to the improved channel estimation formula of expression (17), the beam pattern design is based on minimizing the coherence (total coherence) of the sensing matrix, which is defined as:

[0155]

[0156] Where, μ t (Q k The vector representation of the coherence of the sensing matrix is ​​the sum of the inner products of the q1-th and q2-th columns of the sensing matrix. When digital beamforming occurs, ||Z k，t (:,n)||2=1,where n=1,2,…,Q。 This is the Kronecker product of the transmitted and received beams. Therefore, expression (22) can be written as:

[0157]

[0158] in, It can be seen that when When and when hour, The total coherence is minimized. V N Satisfy the following expression:

[0159]

[0160] in, This indicates the number of pre-defined grid turning vectors used by the user; N represents the number of pre-fixed grid turning vectors in the BS; t Indicates the number of antennas on the ULA equipped at the user end; N r Indicates the number of antennas on the ULA equipped with the BS; I represents the identity matrix.

[0161] Using expression (24) for substitution, expression (23) can be converted to:

[0162]

[0163] Then, Singular Value Decomposition (SVD) can be used to find the optimal Z. k ,For example:

[0164] When Q≤N t N r Solving expression (25) yields:

[0165]

[0166] in, and It is a unitary matrix.

[0167] Through Z k The design can be based on W is approximately obtained k，t and F k，t Preferably, it is assumed that... W k，t and F k，t It can be designed as:

[0168]

[0169] Among them, W k，t F is the matrix representation of the received beam of the t-th pilot block in the k-th reflection vector; k，t This is the matrix representation of the transmission beam in the t-th pilot block within the k-th reflection vector; This indicates the number of pre-defined grid turning vectors used by the user; N represents the number of pre-fixed grid turning vectors in the BS; t Indicates the number of antennas equipped on the user terminal; N r Indicates the number of antennas equipped on the BS; This represents the floor function that returns the largest integer less than or equal to x; I represents the identity matrix, and O represents the zero matrix. Optionally, the designed beam pattern can be implemented as hybrid beamforming, which combines digital and analog methods to reduce the number of RF links and reduce hardware complexity.

[0170] After designing the beam pattern by minimizing the coherence of the sensing matrix and designing the reflection pattern to ensure the sparsity of the cascaded channels, the terminal designs the beam pattern (W) based on the optimized beam pattern. k,t and F k,t By combining the transmitted pilot signal, the pilot signal is reflected at the RIS via the reflection mode (Ξ). After the network side receives the pilot signal, it performs channel estimation based on the compressed sensing algorithm, thus realizing channel estimation of the RIS-assisted millimeter-wave MIMO system. This significantly reduces the computational complexity of channel estimation and the pilot overhead.

[0171] Figure 4 This is a second schematic flowchart of the channel estimation method for a RIS-assisted millimeter-wave system provided in this application embodiment, as shown below. Figure 4 As shown in the figure, this application provides a channel estimation method for a RIS-assisted millimeter-wave system, the execution entity of which can be a terminal or a user terminal. The method includes:

[0172] Step 401: Determine the beam pattern of the pilot signal based on the coherence of the sensing matrix.

[0173] Step 402: Send the pilot signal to the RIS based on the beam pattern; the pilot signal is used for channel estimation.

[0174] In some embodiments, determining the beam pattern of the pilot signal based on the coherence of the sensing matrix includes:

[0175] Determine the Kronecker product of the transmission beam matrix and the receiving beam matrix corresponding to the minimum coherence of the sensing matrix; calculate the transmission beam matrix and the receiving beam matrix based on the Kronecker product of the transmission beam matrix and the receiving beam matrix to obtain the beam pattern of the pilot signal.

[0176] Specifically, the channel estimation method for a RIS-assisted millimeter-wave system provided in this application embodiment can refer to the aforementioned embodiment of the channel estimation method for a RIS-assisted millimeter-wave system where the execution subject is a network device. The parts identical to those in the corresponding method embodiments described above will not be described in detail here. The terminal determines the beam pattern of the pilot signal based on the coherence of the sensing matrix and transmits the pilot signal based on the beam pattern. This allows channel estimation to be achieved with fewer pilot signals and achieves better channel estimation performance, reducing the pilot overhead of the RIS-assisted millimeter-wave system's channel estimation.

[0177] Figure 5 This is a schematic diagram of the structure of a channel estimation device for a RIS-assisted millimeter-wave system provided in an embodiment of this application, as shown below. Figure 5 As shown in the figure, this application provides a channel estimation device for a RIS-assisted millimeter-wave system, including a receiving module 501 and a channel estimation module 502.

[0178] The receiving module 501 is used to receive the pilot signal reflected by RIS; the beam pattern of the pilot signal is determined based on the coherence of the sensing matrix. The channel estimation module 502 is used to perform channel estimation based on the pilot signal using a compressed sensing algorithm.

[0179] In some embodiments, the beam pattern of the pilot signal is determined based on the minimum coherence of the sensing matrix.

[0180] In some embodiments, the channel estimation module includes:

[0181] The first determining unit is used to determine a sparse vector based on the pilot signal and the sensing matrix; the recovery unit is used to recover the cascaded channel matrix based on the sparse vector using a compressed sensing algorithm to obtain the channel estimation result.

[0182] In some embodiments, the first determining unit includes:

[0183] The first determining subunit is used to determine the sensing matrix based on the transmission beam matrix, the receiving beam matrix, and the dictionary matrices corresponding to the network device and the terminal, respectively; the second determining subunit is used to determine the vectorized representation of the pilot signal based on the sensing matrix; the third determining subunit is used to determine the sparse vector based on the vectorized representation of the pilot signal and the pilot signal.

[0184] In some embodiments, the pilot signal is a RIS-based reflection vector.

[0185] In some embodiments, each reflection vector is a column vector among the candidate reflection vectors that is different from the other column vectors; the candidate reflection vectors are determined based on a dictionary matrix associated with the RIS; the dictionary matrix associated with the RIS consists of the orientation vectors of a pre-fixed number of grids; the target number is the number of elements corresponding to the uniform planar array equipped with the RIS.

[0186] Specifically, the channel estimation device for the RIS-assisted millimeter-wave system provided in this application embodiment can implement all the method steps of the channel estimation method embodiment for the RIS-assisted millimeter-wave system with the network device as the execution subject, and can achieve the same technical effect. Here, the parts that are the same as those in the method embodiment and the beneficial effects will not be described in detail.

[0187] Figure 6 This is a schematic diagram of the structure of a channel estimation device for a RIS-assisted millimeter-wave system provided in an embodiment of this application, as shown below. Figure 6 As shown in the figure, this application provides a channel estimation device for a RIS-assisted millimeter-wave system, including a determination module 601 and a transmission module 602.

[0188] The determining module 601 is used to determine the beam pattern of the pilot signal based on the coherence of the sensing matrix. The transmitting module 602 is used to transmit the pilot signal to the RIS based on the beam pattern; the pilot signal is used for channel estimation.

[0189] In some embodiments, the determining module includes:

[0190] The second determining unit is used to determine the Kronecker product of the transmission beam matrix and the receiving beam matrix corresponding to the minimum coherence of the sensing matrix; the calculation unit is used to calculate the transmission beam matrix and the receiving beam matrix based on the Kronecker product of the transmission beam matrix and the receiving beam matrix to obtain the beam pattern of the pilot signal.

[0191] Specifically, the channel estimation device for the RIS-assisted millimeter-wave system provided in this application embodiment can implement all the method steps implemented in the channel estimation method embodiment of the RIS-assisted millimeter-wave system with the terminal as the execution subject, and can achieve the same technical effect. Here, the parts that are the same as those in the method embodiment and the beneficial effects will not be described in detail.

[0192] It should be noted that the division of units / modules in the above embodiments of this application is illustrative and only represents one logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units described above can be implemented in hardware or as software functional units.

[0193] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application, such as... Figure 7 As shown, the electronic device may include: a processor 710, a communication interface 720, a memory 730, and a communication bus 740, wherein the processor 710, the communication interface 720, and the memory 730 communicate with each other through the communication bus 740. The processor 710 can call logical instructions in the memory 730 to execute a channel estimation method for a RIS-assisted millimeter-wave system, the method including:

[0194] Pilot signals are obtained by receiving pilot signals reflected by the intelligent reflective surface RIS; the beam pattern of the pilot signals is determined based on the coherence of the sensing matrix.

[0195] Channel estimation is performed using compressed sensing algorithms based on the pilot signals.

[0196] Specifically, the processor 710 can be a central processing unit (CPU), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a complex programmable logic device (CPLD). The processor can also adopt a multi-core architecture.

[0197] When the logical instructions in memory 730 can be implemented as software functional units and sold or used as independent products, they can be stored in a processor-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0198] In some embodiments, a computer program product is also provided, the computer program product including a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the channel estimation method for the RIS-assisted millimeter-wave system provided in the above method embodiments. The method includes: receiving pilot signals reflected by a smart reflector surface RIS; the beam pattern of the pilot signals is determined based on the coherence of a sensing matrix; and performing channel estimation based on the pilot signals using a compressed sensing algorithm.

[0199] Specifically, the computer program product provided in this application embodiment can implement all the method steps implemented in the above method embodiments and can achieve the same technical effect. Here, the parts that are the same as those in the method embodiments and the beneficial effects will not be described in detail.

[0200] In some embodiments, a computer-readable storage medium is also provided, the computer-readable storage medium storing a computer program for causing a computer to execute the channel estimation method for the RIS-assisted millimeter-wave system provided in the above method embodiments, the method comprising: receiving a pilot signal reflected by a smart reflector RIS; the beam pattern of the pilot signal being determined based on the coherence of a sensing matrix; and performing channel estimation based on the pilot signal using a compressed sensing algorithm.

[0201] Specifically, the computer-readable storage medium provided in the embodiments of this application can implement all the method steps implemented in the above method embodiments and can achieve the same technical effect. Here, the parts that are the same as those in the method embodiments and the beneficial effects will not be described in detail.

[0202] It should be noted that the computer-readable storage medium can be any available medium or data storage device that the processor can access, including but not limited to magnetic storage (e.g., floppy disk, hard disk, magnetic tape, magneto-optical disk (MO)), optical storage (e.g., CD, DVD, BD, HVD), and semiconductor storage (e.g., ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid-state drive (SSD)).

[0203] In this application, "determining B based on A" means that factor A must be considered when determining B. It is not limited to "B can be determined based solely on A," but also includes: "determining B based on A and C," "determining B based on A, C, and E," and "determining C based on A, and further determining B based on C," etc. It can also include using A as a condition for determining B, for example, "when A satisfies the first condition, B is determined using the first method"; or "when A satisfies the second condition, B is determined"; or "when A satisfies the third condition, B is determined based on the first parameter," etc. Of course, it can also be a condition that uses A as a factor in determining B, for example, "when A satisfies the first condition, C is determined using the first method, and B is further determined based on C," etc.

[0204] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (such as disk storage and optical storage) containing computer-usable program code.

[0205] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-executable instructions. These computer-executable instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0206] These processor-executable instructions may also be stored in a processor-readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the processor-readable memory produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The functions specified in one or more boxes. These processor-executable instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0207] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for channel estimation of a RIS-assisted millimeter wave system, the method comprising: The application is applied to a network device, comprising: Receiving a pilot signal reflected by an intelligent reflecting surface (RIS); the beam pattern of the pilot signal is determined based on the coherence of a sensing matrix; the pilot signal is reflected by the RIS based on reflection vectors, each reflection vector being a column vector different from other column vectors in candidate reflection vectors; the candidate reflection vectors are determined based on a dictionary matrix associated with the RIS; Performing channel estimation based on the pilot signal by using a compressed sensing algorithm.

2. The method of claim 1, wherein, The beam pattern of the pilot signal is determined based on the minimum coherence of the sensing matrix.

3. The method of claim 1, wherein, Performing channel estimation based on the pilot signal by using a compressed sensing algorithm, comprising: Determining a sparse vector based on the pilot signal and the sensing matrix; Recovering a cascaded channel matrix based on the sparse vector by using the compressed sensing algorithm to obtain a channel estimation result.

4. The method of claim 3, wherein, The determination of the sparse vector based on the pilot signal and the sensing matrix comprises: Determining the sensing matrix based on a transmission beam matrix, a receiving beam matrix and dictionary matrices corresponding to the network device and a terminal respectively; Determining a vectorized form of the pilot signal based on the sensing matrix; Determining the sparse vector based on the vectorized form of the pilot signal and the pilot signal.

5. The method of claim 1, wherein, The dictionary matrix associated with the RIS is composed of steering vectors of a target number of pre-fixed grids; the target number is the number of elements corresponding to a uniform planar array equipped by the RIS. 6.A method for channel estimation of a RIS-assisted millimeter wave system, characterized in that, The application is applied to a terminal, comprising: Determining a beam pattern of a pilot signal based on the coherence of a sensing matrix; Transmitting the pilot signal to an intelligent reflecting surface (RIS) based on the beam pattern; the pilot signal is used for channel estimation; the pilot signal is reflected by the RIS based on reflection vectors, each reflection vector being a column vector different from other column vectors in candidate reflection vectors; the candidate reflection vectors are determined based on a dictionary matrix associated with the RIS.

7. The method of claim 6, wherein, Determining a beam pattern of a pilot signal based on the coherence of a sensing matrix, comprising: Determining the Kronecker product of a transmission beam matrix and a receiving beam matrix corresponding to the minimum coherence of the sensing matrix; Calculating the transmission beam matrix and the receiving beam matrix based on the Kronecker product of the transmission beam matrix and the receiving beam matrix to obtain the beam pattern of the pilot signal. 8.A device for channel estimation of a RIS-assisted millimeter wave system, characterized in that, Comprising: A receiving module for receiving a pilot signal reflected by an intelligent reflecting surface (RIS); The beam pattern of the pilot signal is determined based on the coherence of a sensing matrix; The pilot signal is reflected by the RIS based on reflection vectors, each reflection vector being a column vector different from other column vectors in candidate reflection vectors; The candidate reflection vectors are determined based on a dictionary matrix associated with the RIS; A channel estimation module for performing channel estimation based on the pilot signal by using a compressed sensing algorithm. 9.A device for channel estimation of a RIS-assisted millimeter wave system, characterized in that, Comprising: A determining module for determining a beam pattern of a pilot signal based on the coherence of a sensing matrix; A transmitting module for transmitting the pilot signal to an intelligent reflecting surface (RIS) based on the beam pattern; The pilot signal is used for channel estimation; the pilot signal is reflected by the RIS based on reflection vectors, each reflection vector being a column vector different from other column vectors in candidate reflection vectors; the candidate reflection vectors are determined based on a dictionary matrix associated with the RIS. ​ 10. An electronic device comprising a memory, a processor, and a computer program stored on the memory and running on the processor, characterized in that, The processor implements the channel estimation method of the RIS-assisted millimeter wave system as claimed in any one of claims 1 to 7 when executing the program. 11.A non-transitory computer-readable storage medium having stored thereon a computer program. The computer program implements the channel estimation method of the RIS-assisted millimeter wave system as claimed in any one of claims 1 to 7 when executed by the processor.

Images