Information acquisition method of tensor-based blocking IRS (Inverse Reference Signal) assisted MIMO (Multiple Input Multiple Output) system

Abstract

The invention aims to provide a tensor-based information acquisition method for a blocking IRS auxiliary MIMO system, and the method comprises the following steps: 1, setting i to be equal to 0, and carrying out the random initialization of a signal matrix # imgabs0 # and a second hop channel matrix # imgabs1 #; 2, i <-i + 1; the method comprises the following steps: step 1, obtaining an estimated value of # imgabs3 through a formula # imgabs2, and reconstructing a time-varying channel # imgabs4, step 4, obtaining an estimated value of # imgabs6 through a formula # imgabs5, wherein # imgabs7, step 5, estimating signal matrixes # imgabs8 and # imgabs9 by using a Kronecker product rearrangement algorithm, step 6, estimating factor matrixes # imgabs11 and # imgabs12 in # imgabs10 through a Khatri-Rao integral solution method, and step 7, repeating the steps until a convergence condition is reached. According to the tensor-oriented blocking IRS-assisted MIMO system information acquisition method disclosed by the invention, the performance of a receiving algorithm can still be ensured under the condition that the blocking IRS is not replaced.

Description

Technical Field

[0001] The invention belongs to the field of wireless communication and intelligent communication, and in particular relates to an information acquisition method of a blocked IRS-assisted MIMO system based on a tensor. Background Art

[0002] With the rise of the sixth-generation mobile communication technology research boom, the increase in hardware costs and the impact of obstacles are important issues that need to be urgently resolved in the communication system. To address this problem, the recently emerged Intelligent Reflecting Surface (IRS) has become a potentially efficient solution. However, due to the dynamics of the mobile user end, the channel exhibits a faster time-varying rate, and the traditional quasi-static channel estimation method has limitations in mobile scenarios. Secondly, in harsh environments, the long time used by the IRS will cause hardware damage, that is, array blocking, and the receiving algorithm will experience serious performance degradation. Therefore, the ability to ensure the performance of the receiving algorithm without replacing the blocked IRS is a problem that needs to be solved at present. Summary of the invention

[0003] The object of the present invention is to provide a tensor-based information acquisition method for a blocked IRS-assisted MIMO system, so as to solve the problem that the receiving algorithm will suffer from serious performance degradation when the array is blocked in the above-mentioned background technology.

[0004] The object of the present invention is to provide a tensor-based information acquisition method for a blocking IRS-assisted MIMO system, namely a time-varying nested Khatri-Rao product and Kronecker product PARAFAC Rearrangement Factorization Alternating Least Squares (TNKRKP-RFALS) receiving algorithm, comprising the following steps:

[0005] Step 1, let i = 0, for the signal matrix and the second hop channel matrix Perform random initialization;

[0006] Step 2, i←i+1;

[0007] Step 3, by formula get The estimated value of

[0008] Step 4, by formula get The estimated value of

[0009] Step 5: Use the Kronecker product rearrangement algorithm to estimate the signal matrix and

[0010] Step 6: Estimate by Khatri-Rao integral decomposition method The factor matrix in and

[0011] (1) From m = 1 to m = M

[0012]

[0013] (2) Apply SVD algorithm to (1), that is, We get (u1,σ1,p1), where σ1 is the maximum singular value;

[0014] (3)b m represents the mth column of the signal matrix B, h 2,m represents the mth column of the second channel matrix H2, where

[0015] (4) Reconstruct the signal matrix and the second channel matrix

[0016]

[0017] Step 7: Repeat the above steps until the convergence condition is reached.

[0018] Compared with the prior art, the present invention has the following beneficial effects:

[0019] Aiming at the IRS array blocking problem, this method proposes a novel TNKRKP-RFALS algorithm. The NKRKP model is constructed according to the structural characteristics of the received signal of the blocked IRS system. In addition, the Kronecker rearrangement and Khatri-Rao decomposition methods are used to estimate the channel, signal and IRS array blocking error matrix respectively. At the same time, the signal can be detected and the channel can be estimated, and the IRS array blocking error can be estimated in real time. Even if the IRS array is blocked, the performance of the receiving algorithm can be guaranteed, which can also reduce the number of times the blocked IRS is replaced, which is conducive to cost saving. In addition, the identifiability conditions of the proposed receiving algorithm are also analyzed and derived. The simulation results are compared with other algorithms, which verifies the accuracy of the performance estimation of the proposed receiving algorithm and is close to the performance boundary curve. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is the system model of the present invention;

[0021] Figure 2 Normalized mean square error curves for the TNKRKP-RFALS algorithm, the HOSVD-STI algorithm, the BALS algorithm, and the NP-ALS algorithm;

[0022] Figure 3 These are the bit error rate curves of the TNKRKP-RFALS algorithm, the HOSVD-STI algorithm, the BALS algorithm, and the NP-ALS algorithm. DETAILED DESCRIPTION

[0023] The present invention is described in detail below in conjunction with the various embodiments shown in the accompanying drawings, but it should be noted that these embodiments are not limitations of the present invention, and any equivalent transformations or substitutions in functions, methods, or structures made by ordinary technicians in the field based on these embodiments are all within the scope of protection of the present invention.

[0024] This embodiment provides a tensor-based information acquisition method for a blocked IRS-assisted MIMO system, including the following steps:

[0025] Step 1, let i = 0, for the signal matrix and the second hop channel matrix Perform random initialization;

[0026] Step 2, i←i+1;

[0027] Step 3, by formula get The estimated value of

[0028] Step 4, by formula get The estimated value of

[0029] Step 5: Use the Kronecker product rearrangement algorithm to estimate the signal matrix and

[0030] Step 6: Estimate by Khatri-Rao integral decomposition method The factor matrix in and

[0031] (1) From m = 1 to m = M

[0032]

[0033] (2) Apply SVD algorithm to (1), that is, We get (u1,σ1,p1), where σ1 is the maximum singular value;

[0034] (3)b m represents the mth column of the signal matrix B, h 2,m represents the mth column of the second channel matrix H2, where

[0035] (4) Reconstruct the signal matrix and the second channel matrix

[0036]

[0037] Step 7: Repeat the above steps until the convergence condition is reached.

[0038] Compared with other inventions, the present invention has obvious advantages including:

[0039] (1) The TNKRKP-RFALS receiving algorithm can ensure the performance of the receiving algorithm even if the IRS array is blocked. This can also reduce the number of times the blocked IRS is replaced, which is beneficial to cost saving.

[0040] (2) The TNKRKP-RFALS receiving algorithm is close to the ZF receiving algorithm. Since the ZF receiving algorithm has a near-perfect estimation performance, it can be used as a basic boundary standard for performance reference. The TNKRKP-RFALS receiving algorithm proposed in the present invention has a higher estimation accuracy.

[0041] (3) Compared with the latest IRS-assisted communication system channel estimation method, the TNKRKP-RFALS channel estimation method not only takes time variation into account but also has better estimation effect.

[0042] In order to verify the performance of the proposed algorithm, a system model is established to analyze the method. It should be noted that once an item is defined in one figure, it does not need to be further defined and explained in the subsequent figures.

[0043] Please refer to Figure 1 , Figure 1 This is a system model diagram of the present invention, that is, the IRS is affected by adverse environmental factors such as frozen snowflakes and accumulated dust.

[0044] Step 1: Randomly initialize the signal matrix S and the channel matrix H2.

[0045] The base station and the intelligent reflector are fixed, and the user end is in a mobile state. The source node sends the symbol to the destination node through F IRS-assisted buildings. In this system, the base station and each IRS unit are equipped with Nb Antennas and different M f (f=1,2,...,F) reflective elements, and the user is equipped with N U Antenna, symbol is L is the symbol length. The channel is assumed to be Rayleigh fading with reciprocity characteristics, and the interference between IRSs and the direct link between the user and the base station are ignored to simplify our problem. is the channel matrix from the t-th time slot user to the f-th IRS-assisted building, is the channel matrix from the fth IRS-assisted building to the base station. The locations of the IRS-assisted buildings and base stations are fixed, while the users are in a mobile state. Therefore, the second hop channel H 2,f The change is greater than the first hop channel H 1,f,t Therefore, assuming that the second hop channel H 2,f The channel coherence time is the first hop channel H 1,f,t T times. In order to estimate the channel state information of all channels, a time domain scheme is proposed. Each time slot is divided into K sub-time slots. The user transmits the same symbol in each sub-time slot. The reflection matrix of IRS changes with the sub-time slot. Each reflective element of IRS can adjust its phase shift and amplitude reflection coefficient separately through the intelligent controller. The reflection matrix of the intelligent reflection surface is It can also be expressed as in

[0046]

[0047] Where 0≤α m,t ≤1,0≤θ m,t ≤2π, and α m,t and θ m,t denote the amplitude attenuation and phase shift disturbance affecting the mth IRS reflection unit element at the tth time block, respectively, and the model takes into account various blockages as follows:

[0048] (1)α m,t ≠0,θ m,t ≠0 indicates amplitude absorption and phase shift disturbance caused by dust particles suspended and accumulated on the mth IRS unit element, or blocking caused by hardware damage in the electronic circuits constituting the IRS.

[0049] (2)α m,t =1,θ m,t ≠0 indicates phase noise disturbance caused by low-resolution phase shift or phase error caused by long-term harsh environment channel estimation.

[0050] (3)αm,t =0 indicates maximum absorption, ie the mth IRS reflection unit element is completely blocked.

[0051] (4)α m,t =1,θ m,t =0 indicates IRS without blocking effect.

[0052] Therefore, b m,t It is expressed as a random variable, that is, the mth IRS reflection unit is expressed as P B (P B ∈[0,1]) encounters a blocking probability, making 1-P B The probability that b m,t = 1. Continuously monitoring the IRS and promptly detecting anomalies or taking targeted corrective measures, such as real-time estimation of the blocking error matrix, is very important to maintain the effective operation of the communication system. In this estimation process, we assume that among the M reflection units on the IRS, there are M B =MP B A reflector unit is blocked.

[0053] When the IRS array blocking is taken into account, the received signal at the receiving end of the user in the kth sub-time slot in the tth time block can be expressed as:

[0054]

[0055] in Represents the Gaussian white noise matrix of the receiving end. In order to facilitate analysis in the derivation process, the noise-free case is considered. The IRS blocking reflection unit matrix of the k-th sub-time slot of the transmitting end user is expressed as:

[0056]

[0057] in, is the IRS blocking matrix, is the IRS reflection unit matrix, which is a vector representation of the received signal

[0058]

[0059] in is the k-th row combination matrix, and the vectorized representation of the first hop channel is defined as The front slices of the vectorized signal received by the data receiving terminal in K sub-time slots are horizontally expanded and arranged to represent express:

[0060]

[0061] in It is a combination matrix. The signal matrix received by the data receiving terminal in T time blocks is stacked along the direction of the subscript t to obtain a multi-dimensional matrix receiving signal Therefore, the multi-dimensional matrix signal received at the data receiving terminal can be constructed into an embedded Khatri-Rao combined Kronecker product PARAFAC tensor model (referred to as embedded mixed product PARAFAC), which is characterized as follows:

[0062]

[0063] The matrix expansion of the embedded mixed product PARAFAC tensor model is as follows:

[0064]

[0065] Step 2: Perform pseudo-inverse operation on the modulo 3 expansion, and use the least squares algorithm to Make an estimate,

[0066]

[0067] Step 3: Perform pseudo-inverse operation on the module 1 expansion and use the least squares algorithm to find Make an estimate,

[0068]

[0069] in

[0070] Step 4: According to the Kronecker product rearrangement algorithm, let The least squares method is used to fit the following cost function to estimate and It is expressed as follows:

[0071]

[0072]

[0073] Let Q = B⊙H2, for the matrix The matrix rearrangement method of Kronecker product is performed to estimate the matrix factors S and Q corresponding to the Kronecker product.

[0074] Step 5: Perform Khatri-Rao integral decomposition on the matrix Q to estimate the matrix factors B and H2 corresponding to the Khatri-Rao product. The Khatri-Rao integral decomposition method is as follows:

[0075]

[0076] The normalized mean square error curve and bit error rate diagram of the algorithm are shown in Figure 2 , Figure 3 As shown, it can be seen that the algorithm is superior to the traditional receiving algorithm, and the effectiveness of the algorithm is verified by simulation results.

[0077] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the present invention can be implemented in other specific forms without departing from the spirit or essential features of the present invention. Therefore, the embodiments should be considered exemplary and non-restrictive in all respects, and the scope of the present invention is defined by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims be included in the present invention.

Claims

1. A tensor-based information acquisition method for a blocked IRS-assisted MIMO system, the system model of which is: The system model is a tensor-based time-varying channel estimation model for a multi-user multi-antenna system assisted by a blocked IRS. In order to ensure the performance of the receiving algorithm under the condition of a blocked IRS, a novel TNKP-RALS algorithm is adopted to construct the NKRKP model according to the structural characteristics of the received signal of the blocked IRS system. In addition, the Kronecker rearrangement and Khatri-Rao decomposition methods are used to estimate the channel, signal and IRS array blocking error matrix respectively. At the same time, it can detect the signal and estimate the channel, and estimate the IRS array blocking error in real time.

2. According to the system model described in claim 1, the characteristics of the wireless intelligent communication system information acquisition algorithm assisted by the blocked IRS are as follows: the user terminal in the actual scenario is in a mobile state, the channel presents a faster time-varying rate, and the traditional quasi-static channel estimation method has limitations in mobile scenarios. Secondly, in harsh environments, the long time used by the IRS will cause hardware damage, that is, array blocking, and the receiving algorithm will experience serious performance degradation. The TNKP-RALS algorithm can still guarantee the performance of the receiving algorithm without replacing the blocked IRS.

3. The wireless communication system information acquisition algorithm based on tensor blocking IRS as claimed in claim 2, comprising the steps of: Step 1, let i = 0, for the signal matrix and the second hop channel matrix Perform random initialization; Step 2, i←i+1; Step 3, by formula get The estimated value of Step 4, by formula get The estimated value of Step 5: Use the Kronecker product rearrangement algorithm to estimate the signal matrix and Step 6: Estimate by Khatri-Rao integral decomposition method The factor matrix in and Step 7: Repeat the above steps until the convergence condition is reached.

4. The Khatri-Rao integral decomposition algorithm in the wireless communication system information acquisition algorithm mentioned in claim 3 comprises the following steps: Step 1, Rearrange each column from m=1 to m=M Step 2: Perform the SVD algorithm, that is We get (u1,σ1,p1), where σ1 is the maximum singular value; Step 3, b m represents the mth column of the signal matrix B, h 2,m represents the mth column of the second channel matrix H2, where Step 4: Reconstruct the signal matrix and the second channel matrix 5. A wireless intelligent communication system between a tensor-based blocking IRS and a base station, characterized in that: User signal sending device, base station signal receiving device and channel estimation algorithm; wherein the channel estimation algorithm includes 4-QAM modulation on the sending device, 4-QAM demodulation on the signal receiving device and corresponding decoding, and the specific steps of the algorithm are as shown in steps 1-7 of the TNKP-RALS algorithm in claim 3.

6. An IRS-assisted wireless intelligent communication system between a mobile device and a base station, characterized in that: A mobile device with signal transmission function, a base station with signal reception capability and information acquisition algorithm capability, and a channel estimation algorithm; wherein, due to long-term use of IRS, IRS hardware damage is caused, and a NKPKP multi-dimensional matrix model is constructed according to a blocked IRS system model. The proposed receiving algorithm does not require the antenna array to be manually removed from the building and allows real-time array diagnosis. The specific steps of the algorithm are shown in steps 1-7 of the TNKP-RALS algorithm in claim 3.

7. A wireless intelligent signal receiver simulation program product based on blocking IRS assistance, comprising a computer program, which realizes the system model in claim 1 when being executed.

8. A computer program product, comprising a computer program, which implements the information acquisition algorithm in claims 3 and 4 when executed.

9. A wireless intelligent communication electronic device, characterized in that: A device with a signal sending function, a device with a signal receiving function, a processor, a memory and a wireless communication intelligent channel estimation algorithm program as claimed in claims 3 and 4; wherein the program is stored in the memory and is configured to be executed by the processor, and the computer program includes a computer program for executing the channel estimation algorithm as claimed in claims 3 and 4.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the channel estimation algorithm in claims 3 and 4 or the receiver simulation program in claim 7 when executed by a processor.

Images