Channel estimation method for intelligent reflector-assisted multi-user millimeter wave MIMO (Multiple Input Multiple Output) system based on compressed sensing

By transforming the channel estimation problem of multi-user cascaded channels into solving the channel estimation problem of cascaded factors and typical user cascaded channels, and using covariance matrix feature decomposition and angle domain projection signal processing, a two-stage channel estimation algorithm is designed to solve the problem of solving the problem of cascaded factors and typical user cascaded channels. The problems of high CSI acquisition overhead and error propagation in the prior art are achieved, and lower pilot overhead and higher channel estimation accuracy are achieved.

CN119051694APending Publication Date: 2024-11-29KUNMING UNIV OF SCI & TECH
View PDF 0 Cites 3 Cited by

Patent Information

Application Number
CN202411192848.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-28
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

In the existing IRS-assisted millimeter wave MIMO system channel estimation technology based on compression sensing, acquiring channel state information (CSI) requires a large pilot overhead and high algorithm complexity. In estimating other user-levels using typical user cascade channel information. Error propagation problems may occur when connecting channel information.

Method used

By establishing a channel model and received signal model, the channel estimation problem of multi-user cascaded channels is transformed into solving the channel estimation problem of cascaded factors and typical user cascaded channels. The covariance matrix feature decomposition and angle domain projection signal processing are used to design Two-stage channel estimation algorithm, including preliminary channel estimation and reparameterized CSI estimation, optimizes cascading factors to reduce pilot overhead and algorithm complexity.

Benefits of technology

While maintaining estimation accuracy, pilot overhead and algorithm complexity are reduced, the requirement of prior conditions is reduced, and error propagation problems are effectively solved, achieving lower pilot overhead and higher channel estimation accuracy .

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119051694A_ABST
    Figure CN119051694A_ABST
Patent Text Reader

Abstract

The invention relates to a channel estimation method of an intelligent reflector assisted multi-user millimeter wave MIMO system based on compressed sensing, and belongs to the technical field of wireless communication. The method comprises the following steps: establishing a channel model and a receiving signal model by utilizing the sparsity and correlation of a multi-user cascade channel in a millimeter wave MIMO system, calculating the number of scattering paths between a base station and an IRS and a public subspace of the multi-user cascade channel, projecting a receiving signal matrix to the subspace to simplify the channel estimation model, and estimating the multi-user cascade channel according to the received signal matrix. A low-complexity preliminary channel estimation method based on a simplified model is designed, and finally, a preliminary channel estimation result is optimized through an alternating iterative optimization algorithm to obtain accurate channel estimation information. According to the method, the cascade channel is estimated based on the compressed sensing technology, and compared with a traditional channel estimation technology based on compressed sensing, under the condition that the accuracy is kept, the pilot frequency overhead is smaller, and the algorithm complexity is lower.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to a channel estimation method for an intelligent reflecting surface-assisted multi-user millimeter wave MIMO system based on compressed sensing, and belongs to the technical field of wireless communications. Background Art

[0002] Since millimeter-wave signals have a short wavelength and cannot be transmitted over long distances, hybrid beamforming of a large-scale MIMO system is needed to obtain large beam gains and thus improve the quality of the millimeter-wave communication system.

[0003] Reconfigurable smart surface (IRS) is a promising technology that can achieve high spectral efficiency and high energy efficiency in wireless communication systems. IRS is usually composed of a large number of passive elements, each of which can produce phase shift and passive reflection of the incident signal. This feature enables IRS to establish additional communication links between base stations and users. Due to the poor penetration of millimeter-wave signals, reconfiguring IRS elements according to the surrounding environment can provide high beamforming gain for multi-user millimeter-wave MIMO systems, increase the power of received signals, and thus improve transmission quality.

[0004] The performance gain provided by IRS depends largely on the accuracy of channel state information (CSI). However, since the reflective element of IRS is a passive device and lacks the ability to transmit, receive and process pilot signals, the acquisition of CSI requires a high pilot overhead.

[0005] The cascaded channels in the IRS-assisted millimeter-wave MIMO communication system are highly sparse in the angle domain, and compressed sensing technology can recover sparse signals with lower pilot overhead. Therefore, the channel estimation problem of the cascaded channels can be transformed into a sparse signal recovery problem of compressed sensing and can be effectively solved by existing compressed sensing technology. Summary of the invention

[0006] The technical problem to be solved by the present invention is to address the shortcomings of the current IRS-assisted millimeter-wave MIMO system channel estimation technology based on compressed sensing. Acquiring CSI requires a large pilot overhead and the algorithm complexity is high. A channel estimation method for an IRS-assisted millimeter-wave MIMO system based on compressed sensing is provided. While maintaining the estimation accuracy, the pilot overhead and the algorithm complexity are reduced, and the "error propagation" problem that may arise when estimating the cascade channel information of other users using the cascade channel information of typical users is better solved.

[0007] The technical solution of the present invention is: a channel estimation method for a multi-user millimeter wave MIMO system assisted by an intelligent reflector based on compressed sensing, the method comprising:

[0008] Step 1: Establish the channel model and the received signal model, and transform the channel estimation problem of multiple user cascade channels into solving the cascade factor α k Channel estimation problem for the concatenated channel G1 with a typical user.

[0009] Step 2: Estimate the number of channel scattering paths between the base station and the IRS, solve the covariance matrix of the received signal model, and perform eigendecomposition on the covariance matrix to find the common column subspace of the multi-user cascade channel.

[0010] Step 3: Calculate the projection signal of the received signal matrix on the common column subspace.

[0011] Step 4: Based on the obtained projection signal, a preliminary channel estimation algorithm is designed. The algorithm is divided into two stages. The first stage estimates the cascade channel estimation information of a typical user. The second stage estimates the cascade channel estimation information of other users based on the cascade channel information obtained in the first stage by re-parameterizing the CSI.

[0012] Step 5: Based on the preliminary channel estimation information of all users, establish the cascade factor α k Angle domain projection channel with a typical user The alternating iterative optimization problem is solved and the channel estimation information of the reconfigurable smart surface assisted multi-user millimeter wave MIMO system is obtained.

[0013] The channel model is a classic narrowband SV channel model, including the channel H between the base station (BS) and the intelligent reflecting surface (IRS) and the channel h between the intelligent reflecting surface (IRS) and the user. k .

[0014] A uniform linear array (ULA) with M antennas is constructed at the base station (BS), and a uniform planar array (UPA) with N antennas is constructed at the intelligent reflecting surface (IRS), where N = N1 × N2, where N1 represents the number of horizontal antennas and N2 represents the number of vertical antennas. The channel between the BS and the IRS is It is expressed as:

[0015]

[0016] Among them, L H represents the number of scattering paths of channel H, and the complex gain on the lth path in channel H is α l express, is the UPA array steering vector at the base station, is the ULA array steering vector at the IRS, expressed as:

[0017]

[0018] Where n1 = [0, 1, ..., N1-1], n2 = [0, 1, ..., N2-1], d represents the distance between adjacent antennas, λ represents the wavelength of the carrier, and Indicates the azimuth and elevation angles of AOA at IRS, It represents the azimuth of AOD at BS. The angle between AOA and AOD is usually in the range of [-π / 2,π / 2].

[0019] Channel between intelligent reflecting surface (IRS) and user (UE) It is expressed as:

[0020]

[0021] Among them, J k represents the channel h between the Kth user and the IRS k The number of scattering paths, β k,j Indicates channel h k The complex gain of the lth path is, and represents the azimuth and elevation angles of the AOD at the IRS, is the UPA array steering vector at the IRS.

[0022] The cascade channel between the base station and the user is expressed as:

[0023]

[0024] According to the scaling property of the cascade channel, the cascade channel is expressed as the cascade factor α k The product of a typical user cascade channel is:

[0025]

[0026] in

[0027]

[0028] That is, the channel estimation problem of multiple user cascade channels is transformed into solving the cascade factor α k Channel estimation problem for the concatenated channel G1 with a typical user.

[0029] Constructing the received signal model is specifically as follows:

[0030] First, the signal is transmitted from the user end to the base station through the channel h k , reflection from the intelligent reflecting surface IRS element and channel H, completing a signal transmission.

[0031] The time to complete all signal transmission is the communication time

[0032] Then, the communication time is divided into Q subframes, each subframe contains T time slots, and each subframe performs a signal transmission. Assuming that the pilot signal transmitted by all users is 1, the signal sent by user k received at the base station in a subframe q is It is expressed as:

[0033]

[0034] in, is a noise vector, all elements of which are independent and identically distributed in a Gaussian normal distribution is the reflection coefficient of the IRS element in subframe q, and v in different time slots q In the same subframe, q is the same.

[0035] The signals of Q subframes received at the base station are combined into a received signal matrix As follows:

[0036]

[0037] in, w q,k The noise matrix is ​​concatenated into

[0038] Then the pilot signals sent by all users in Q subframes received at the base station are:

[0039]

[0040] Y is the received signal model.

[0041] The Step 2 is specifically as follows:

[0042] Step 2.1: Estimate the number of scattering paths of the H channel based on the minimum length description criterion (MDL)

[0043] Step 2.2: Calculate the covariance matrix of the received signal

[0044] Step 2.3: Find the covariance matrix Eigenvalue decomposition of

[0045] Step 2.4: The common column subspace of the multi-user channel is the front of the matrix D The subspace D spanned by the linear combination of || .

[0046] The Step 3 is specifically as follows:

[0047] Cascade channel G k Re-expressed using the virtual angle domain:

[0048]

[0049] in, and is the dictionary matrix in the angle domain, and the angle resolutions are G r With G t , angular resolution indicates the degree of angle quantization. The higher the angular resolution, the better it can simulate the actual angle.

[0050] is the angle domain sparse matrix of the cascaded channel.

[0051] At this time, the received signal is projected:

[0052]

[0053] Among them, equation (a) is established because At this time, the noise of the received signal after projection Compare the noise of the received signal before projection The noise power is significantly reduced.

[0054] The Step 4 is specifically as follows:

[0055] Step 4.1: Select the signal with the largest energy from the projection signals of the received signals of K users and record it as Its cascade channel is denoted as G1.

[0056] Step 4.2: Signal Matrix Doing the conjugate transpose calculation, we get Also because for and Divide the columns into blocks and get

[0057] Step 4.3: Based on equation use The OMP algorithm is Recover from

[0058] Step 4.4: Estimated by Step 4.3 Get cascade channel

[0059] Step 4.5: Estimation information based on the estimated cascade channel Compute the re-parameterized concatenated channel estimation information for other users.

[0060]

[0061] in, The channel between the kth user's reparameterized base station and IRS is: is the re-parameterized channel between user k and IRS.

[0062] The Step 5 is specifically as follows:

[0063] Step 5.1: Re-parameterize the cascaded channel estimation information of all users obtained in Step 4 The following optimization problem is established.

[0064]

[0065] Step 5.2: Initialize parameters Cascade Factor Set the convergence accuracy ε and the outer loop iteration counter r=0.

[0066] Step 5.3: Settings The inner loop iteration counter t=0.

[0067] Step 5.4: For the given Iterative calculation based on the formula And set t=t+1 until the inner loop iteration converges.

[0068] Step 5.5: Set r = r + 1, update For a given Iterative calculation based on the formula until the outer iteration converges.

[0069] Step 5.6: Output optimization results and The result of channel estimation is:

[0070] The beneficial effects of the present invention are as follows: the present invention estimates cascade channels based on compressed sensing technology. Compared with traditional channel estimation technology based on compressed sensing, the present invention has smaller pilot overhead and lower algorithm complexity while maintaining accuracy, and requires fewer prior conditions, and better solves the "error propagation" problem that may arise when using typical user cascade channel information to estimate other users' cascade channel information. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 It is a schematic diagram of a smart reflector-assisted multi-user millimeter-wave MIMO system based on compressed sensing of the present invention;

[0072] Figure 2 It is a flow chart of a channel estimation method for cascaded channels of an IRS-assisted multi-user millimeter wave MIMO system based on compressed sensing of the present invention.

[0073] Figure 3 It is a simulation diagram of the relationship between the normalized mean square error and the pilot overhead of each channel estimation method. DETAILED DESCRIPTION

[0074] The present invention will be further described below in conjunction with the accompanying drawings and specific implementation methods.

[0075] Example 1: Figure 1 As shown in the figure, a base station with M = 128 antennas communicates with K = 3 single-antenna users. When one of the users communicates, the direct link is blocked because the millimeter wave signal cannot penetrate the wall. Therefore, an IRS with N = 128 reflective elements is deployed to reflect the signal to construct a reliable link to maintain communication services. The other users have two communication links, H and h k The present invention proposes to connect the cascade channel to the communication channel between the IRS and the base station and between the IRS and the user. The estimation method is as follows: Figure 2 As shown, the specific steps are as follows:

[0076] Step 1: Establish the channel model and the received signal model, and transform the channel estimation problem of multiple user cascade channels into solving the cascade factor α k Channel estimation problem for the concatenated channel G1 with a typical user.

[0077] Step 2: Estimate the number of channel scattering paths between the base station and the IRS, solve the covariance matrix of the received signal model, and perform eigendecomposition on the covariance matrix to find the common column subspace of the multi-user cascade channel.

[0078] Step 3: Calculate the projection signal of the received signal matrix on the common column subspace.

[0079] Step 4: Based on the obtained projection signal, a preliminary channel estimation algorithm is designed. The algorithm is divided into two stages. The first stage estimates the cascade channel estimation information of a typical user. The second stage estimates the cascade channel estimation information of other users based on the cascade channel information obtained in the first stage by re-parameterizing the CSI.

[0080] Step 5: Based on the preliminary channel estimation information of all users, establish the cascade factor α k Angle domain projection channel with a typical user The alternating iterative optimization problem is solved and the channel estimation information of the reconfigurable smart surface assisted multi-user millimeter wave MIMO system is obtained.

[0081] The channel model is a classic narrowband SV channel model, including the channel H between the base station (BS) and the intelligent reflecting surface (IRS) and the channel h between the intelligent reflecting surface (IRS) and the user. k .

[0082] A uniform linear array (ULA) with M antennas is constructed at the base station (BS), and a uniform planar array (UPA) with N antennas is constructed at the intelligent reflecting surface (IRS), where N = N1 × N2, where N1 represents the number of horizontal antennas and N2 represents the number of vertical antennas. The channel between the BS and the IRS is It is expressed as:

[0083]

[0084] Among them, L H represents the number of scattering paths of channel H, and the complex gain on the lth path in channel H is α l express, is the UPA array steering vector at the base station, is the ULA array steering vector at the IRS, expressed as:

[0085]

[0086] Where n1 = [0, 1, ..., N1-1], n2 = [0, 1, ..., N2-1], d represents the distance between adjacent antennas, λ represents the wavelength of the carrier, and Indicates the azimuth and elevation angles of AOA at IRS, It represents the azimuth of AOD at BS. The angle between AOA and AOD is usually in the range of [-π / 2,π / 2].

[0087] Channel between intelligent reflecting surface (IRS) and user (UE) It is expressed as:

[0088]

[0089] Among them, J k represents the channel h between the Kth user and the IRS k The number of scattering paths, β k,j Indicates channel h k The complex gain of the lth path is, and represents the azimuth and elevation angles of the AOD at the IRS, is the UPA array steering vector at the IRS.

[0090] The cascade channel between the base station and the user is expressed as:

[0091]

[0092] According to the scaling property of the cascade channel, the cascade channel is expressed as the cascade factor α k The product of a typical user cascade channel is:

[0093]

[0094] in

[0095]

[0096] That is, the channel estimation problem of multiple user cascade channels is transformed into solving the cascade factor α k Channel estimation problem for the concatenated channel G1 with a typical user.

[0097] Constructing the received signal model is specifically as follows:

[0098] First, the signal is transmitted from the user end to the base station through the channel h k , reflection from the intelligent reflecting surface IRS element and channel H, completing a signal transmission.

[0099] The time to complete all signal transmission is the communication time

[0100] Then, the communication time is divided into Q subframes, each subframe contains T time slots, and each subframe performs a signal transmission. Assuming that the pilot signal transmitted by all users is 1, the signal sent by user k received at the base station in a subframe q is It is expressed as:

[0101]

[0102] in, is a noise vector, all elements of which are independent and identically distributed in a Gaussian normal distribution is the reflection coefficient of the IRS element in subframe q, and v in different time slots q In the same subframe, q is the same.

[0103] The signals of Q subframes received at the base station are combined into a received signal matrix As follows:

[0104]

[0105] in, w q,k The noise matrix is ​​concatenated into

[0106] Then the pilot signals sent by all users in Q subframes received at the base station are:

[0107]

[0108] Y is the received signal model.

[0109] The Step 2 is specifically as follows:

[0110] Step 2.1: Estimate the number of scattering paths of the H channel based on the minimum length description criterion (MDL)

[0111] Step 2.2: Calculate the covariance matrix of the received signal

[0112] Step 2.3: Find the covariance matrix Eigenvalue decomposition of

[0113] Step 2.4: The common column subspace of the multi-user channel is the front of the matrix D The subspace D spanned by the linear combination of || .

[0114] The Step 3 is specifically as follows:

[0115] Cascade channel G k Re-expressed using the virtual angle domain:

[0116]

[0117] in, and is the dictionary matrix in the angle domain, and the angle resolutions are G r With G t , angular resolution indicates the degree of angle quantization. The higher the angular resolution, the better it can simulate the actual angle.

[0118] is the angle domain sparse matrix of the cascaded channel.

[0119] At this time, the received signal is projected:

[0120]

[0121] Among them, equation (a) is established because At this time, the noise of the received signal after projection Compare the noise of the received signal before projection The noise power is significantly reduced.

[0122] The Step 4 is specifically as follows:

[0123] Step 4.1: Select the signal with the largest energy from the projection signals of the received signals of K users and record it as Its cascade channel is denoted as G1.

[0124] Step 4.2: Signal Matrix Doing the conjugate transpose calculation, we get Also because for and Divide the columns into blocks and get

[0125] Step 4.3: Based on equation use The OMP algorithm is Recover from

[0126] Step 4.4: Estimated by Step 4.3 Get cascade channel

[0127] Step 4.5: Estimation information based on the estimated cascade channel Compute the re-parameterized concatenated channel estimation information for other users.

[0128]

[0129] in, The channel between the kth user's reparameterized base station and IRS is: is the re-parameterized channel between user k and IRS.

[0130] The Step 5 is specifically as follows:

[0131] Step 5.1: Re-parameterize the cascaded channel estimation information of all users obtained in Step 4 The following optimization problem is established.

[0132]

[0133] Step 5.2: Initialize parameters Cascade Factor Set the convergence accuracy ε and the outer loop iteration counter r=0.

[0134] Step 5.3: Settings The inner loop iteration counter t=0.

[0135] Step 5.4: For the given Iterative calculation based on the formula And set t=t+1 until the inner loop iteration converges.

[0136] Step 5.5: Set r = r + 1, update For a given Iterative calculation based on the formula until the outer iteration converges.

[0137] Step 5.6: Output optimization results and The result of channel estimation is:

[0138] like Figure 3 As shown in the figure, it is observed that the performance of all methods improves with the increase of pilot overhead B. Among them, the S-Genie-aided LS algorithm estimates the cascade channel by assuming that the BS knows the exact angle of the IRS or user-side cascaded AOA and the BS-side AOD, and uses the solution of the LS estimator to project the received signal into the known common AOD subspace. This method serves as the performance lower bound of all other methods. It can be seen from the figure that under the same pilot overhead, the channel estimation accuracy of this method is better than that of the traditional algorithm, which proves that this method can maintain accuracy while lowering the pilot overhead than the traditional algorithm.

[0139] The specific implementation modes of the present invention are described in detail above in conjunction with the accompanying drawings, but the present invention is not limited to the above implementation modes, and various changes can be made within the knowledge scope of ordinary technicians in this field without departing from the purpose of the present invention.

Claims

1. A channel estimation method for a multi-user millimeter wave MIMO system assisted by an intelligent reflector based on compressed sensing, characterized in that , the method comprises: Step 1: Establish the channel model and the received signal model, and transform the channel estimation problem of multiple user cascade channels into solving the cascade factor α k The channel estimation problem of the cascaded channel G1 with a typical user; Step 2: Estimate the number of channel scattering paths between the base station and the IRS, solve the covariance matrix of the received signal model, and perform eigendecomposition on the covariance matrix to find the common column subspace of the multi-user cascade channel; Step 3: Calculate the projection signal of the received signal matrix on the common column subspace; Step 4: Based on the obtained projection signal, a preliminary channel estimation algorithm is designed. The algorithm is divided into two stages. The first stage estimates the cascade channel estimation information of a typical user. The second stage estimates the cascade channel estimation information of other users by re-parameterizing the CSI based on the cascade channel information obtained in the first stage. Step 5: Based on the preliminary channel estimation information of all users, establish the cascade factor α k Angle domain projection channel with a typical user The alternating iterative optimization problem is solved and the channel estimation information of the reconfigurable smart surface assisted multi-user millimeter wave MIMO system is obtained.

2. The channel estimation method for a multi-user millimeter wave MIMO system assisted by an intelligent reflector based on compressed sensing according to claim 1, characterized in that: The channel model is a classic narrowband SV channel model, including the channel H between the base station and the smart reflective surface and the channel h between the smart reflective surface and the user. k ; A uniform linear array of M antennas is constructed at the base station, and a uniform planar array of N antennas is constructed at the smart reflector. N = N1 × N2, where N1 represents the number of horizontal antennas and N2 represents the number of vertical antennas. Then the channel between BS and IRS is It is expressed as: Among them, L H represents the number of scattering paths of channel H, and the complex gain on the lth path in channel H is α l express, is the UPA array steering vector at the base station, is the ULA array steering vector at the IRS, expressed as: Where n1 = [0, 1, ..., N1-1], n2 = [0, 1, ..., N2-1], d represents the distance between adjacent antennas, λ represents the wavelength of the carrier, and Indicates the azimuth and elevation angles of AOA at IRS, It represents the azimuth of AOD at BS. The angle range between AOA and AOD is usually [-π / 2,π / 2]; Channel between smart reflective surface and user It is expressed as: Among them, J k represents the channel h between the Kth user and the IRS k The number of scattering paths, β k,j Indicates channel h k The complex gain of the lth path is, and represents the azimuth and elevation angles of the AOD at the IRS, is the UPA array steering vector at IRS; The cascade channel between the base station and the user is expressed as: According to the scaling property of the cascade channel, the cascade channel is expressed as the cascade factor α k The product of a typical user cascade channel is: in That is, the channel estimation problem of multiple user cascade channels is transformed into solving the cascade factor α k Channel estimation problem for the concatenated channel G1 with a typical user.

3. The channel estimation method for a multi-user millimeter wave MIMO system assisted by an intelligent reflector based on compressed sensing according to claim 2, characterized in that: Constructing the received signal model is specifically as follows: First, the signal is transmitted from the user end to the base station through the channel h k , the reflection of the intelligent reflective surface IRS element and channel H complete a signal transmission; Then, the communication time is divided into Q subframes, each subframe contains T time slots, and each subframe performs a signal transmission. Assuming that the pilot signal transmitted by all users is 1, the signal sent by user k received at the base station in a subframe q is It is expressed as: in, is a noise vector, all elements of which are independent and identically distributed in a Gaussian normal distribution is the reflection coefficient of the IRS element in subframe q, and v in different time slots q In the same subframe, q is the same; The signals of Q subframes received at the base station are combined into a received signal matrix As follows: in, w q,k The noise matrix is ​​concatenated into Then the pilot signals sent by all users in Q subframes received at the base station are: Y is the received signal model.

4. The channel estimation method for a multi-user millimeter wave MIMO system assisted by an intelligent reflector based on compressed sensing according to claim 1, characterized in that: The Step 2 is specifically as follows: Step 2.1: Estimate the number of scattering paths of the H channel based on the minimum length description criterion Step 2.2: Calculate the covariance matrix of the received signal Step 2.3: Find the covariance matrix Eigenvalue decomposition of Step 2.4: The common column subspace of the multi-user channel is the front of the matrix D The subspace D spanned by the linear combination of || .

5. The channel estimation method for a multi-user millimeter wave MIMO system assisted by an intelligent reflector based on compressed sensing according to claim 1, characterized in that: The Step 3 is specifically as follows: Cascade channel G k Re-expressed using the virtual angle domain: in, and is the dictionary matrix in the angle domain, and the angle resolutions are G r With G t ,The angular resolution indicates the degree of angle quantization. The higher the angular resolution, the better it can simulate the actual angle; is the angle domain sparse matrix of the cascaded channel; At this time, the received signal is projected: Among them, equation (a) is established because At this time, the noise of the received signal after projection Compare the noise of the received signal before projection The noise power is significantly reduced.

6. The channel estimation method for a multi-user millimeter wave MIMO system assisted by an intelligent reflector based on compressed sensing according to claim 1, characterized in that: The Step 4 is specifically as follows: Step 4.1: Select the signal with the largest energy from the projection signals of the received signals of K users and record it as Its cascade channel is denoted as G1; Step 4.2: Signal Matrix Doing the conjugate transpose calculation, we get Also because for and Divide the columns into blocks and get Step 4.3: Based on equation use The OMP algorithm starts from p1, p2, ..., Recover from h IRS,1 ,h IRS ,2,..., Step 4.4: h estimated by Step 4.3 IRS,1 ,h IRS,2 ,..., Get cascade channel Step 4.5: Estimation information based on the estimated cascade channel Calculate the re-parameterized cascade channel estimation information of other users; in, The channel between the kth user's reparameterized base station and IRS is: is the re-parameterized channel between user k and IRS.

7. The channel estimation method for a multi-user millimeter wave MIMO system assisted by an intelligent reflector based on compressed sensing according to claim 1, characterized in that: The Step 5 is specifically as follows: Step 5.1: Re-parameterize the cascaded channel estimation information of all users obtained in Step 4 The following optimization problem is established; Step 5.2: Initialize parameters Cascade Factor Set the convergence precision ε, and the outer loop iteration counter r = 0; Step 5.3: Settings Inner loop iteration counter t = 0; Step 5.4: For the given Iterative calculation based on the formula And set t=t+1 until the inner loop iteration converges; Step 5.5: Set r = r + 1, update For a given Iterative calculation based on the formula Until the outer iteration converges; Step 5.6: Output optimization results and The result of channel estimation is:

Citation Information

Cited By

  • Semi-active IRS system channel estimation method and system based on random and fixed sampling

    CN120455210A

  • A Channel Estimation Method and System for Semi-Active IRS Systems Based on Random and Fixed Sampling

    CN120455210B

  • User perception method based on multilayer intelligent omnidirectional metasurface

    CN121690289A