A method for separation channel reconstruction in MIMO system assisted by reconfigurable intelligent surface

By estimating channel parameters in stages in RIS-assisted MIMO-OFDM system, using different configurations of pilot signals and Newton's orthogonal matching tracking algorithm, the problems of channel estimation difficulty and pilot overhead are solved, and high-accuracy channels are achieved Improve communication efficiency of reconstruction and multi-user systems.

CN115865575BActive Publication Date: 2024-11-29SOUTHEAST UNIV
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202211509806.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-11-29
Publication Date
2024-11-29
Estimated Expiration
2042-11-29

AI Technical Summary

Technical Problem

In RIS auxiliary MIMO-OFDM system, channel estimation is difficult and pilot overhead is high, resulting in difficulty in obtaining channel parameters, especially in multi-user systems, which affects communication efficiency and accuracy.

Method used

The user sends pilot signals for two time periods, the first time period occupies all subcarriers, and designs the reconstructed intelligent surface RIS reflection coefficient; the second time period occupies part of the subcarriers, sets the random RIS reflection coefficient, and the base station extracts it. Channel parameters, estimate direction angle, delay and gain information in stages, and use Newton's orthogonal matching tracking algorithm to estimate and reconstruct channel parameters.

Benefits of technology

Highly accurate channel estimation and separate channel reconstruction under low pilot overhead are realized, reducing the difficulty of channel estimation in the RIS system, suitable for multi-user scenarios, and improving communication efficiency and user positioning accuracy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115865575B_ABST
    Figure CN115865575B_ABST
Patent Text Reader

Abstract

The invention discloses a method for reconstructing separation channels of a reconfigurable intelligent surface-assisted MIMO system, comprising the following steps: a user sends pilots in two time periods, a pilot at a single moment occupying all subcarriers is used in the first time period, a pilot at continuous moments occupying only part of the subcarriers is used in the second time period, and a random RIS reflection coefficient is set; through the RIS reflection signal, the base station extracts the azimuth, delay and gain information of all propagation paths in the separation channel of the target user; the extraction process of the separation channel parameters is divided into two stages, the base station first completes the extraction of the angle and delay of the UE-RIS segment according to the known LoS channel of the RIS-BS segment, and then completes the extraction of the angle and delay of the remaining NLoS paths of the RIS-BS segment according to the previous results, and estimates the gains respectively, and finally completes the estimation of all channel parameters; the base station completes the separation channel reconstruction using the parameters. The invention realizes the RIS cascade channel data transmission under low signal-to-noise ratio conditions.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of wireless communication, and in particular relates to a separation channel reconstruction method based on a RIS (Reconfigurable Intelligent Surface) assisted MIMO-OFDM (Multiple Input Multiple Output Orthogonal Frequency Division Multiplexing) system. Background Art

[0002] Reconfigurable smart metasurfaces have recently been considered as a potential solution to meet the needs of future wireless networks because of their fast configuration and low power consumption. RIS can be used to improve the coverage and spectrum efficiency of wireless communications by deploying a large number of nearly passive reflective elements to programmably configure the wireless propagation environment. It has been used in existing wireless communication systems and has been deeply studied and applied in indoor positioning, environmental perception, transceivers, etc. In addition, companies in the wireless industry have begun to test prototype systems based on RIS in 5G networks with base stations and commercial user equipment.

[0003] However, channel estimation in RIS-assisted wireless systems is more difficult than channel estimation in traditional systems because the reconfigured channel model actually adds a cascade channel with a large number of units as the core. In addition, RIS usually works in a nearly passive state and has no signal processing capabilities. These problems greatly increase the difficulty of obtaining perfect channel state information (CSI) channel estimation. The focus of the research is on solutions to more specific pilot overhead problems. Because the overhead of training pilots is closely related to the number of RIS elements, RIS reflectors with a large number of units will inevitably lead to a significant increase in the pilot overhead of multi-user systems, which is unacceptable in existing communication systems.

[0004] Based on sparse channel estimation, the channel parameter estimation method of the RIS system means that not only the channel matrix needs to be obtained, but also a large number of channel parameters related to each user, such as fading coefficient, direction of arrival (DOA) and delay, need to be obtained according to the channel matrix. This can realize the estimation of the separated channel state information, thereby providing important guidance for performing precoding, channel reconstruction, coherent detection, environmental perception, user positioning and other tasks. However, in current research, parameter estimation for the RIS end is currently rare, mainly because the design and calculation difficulties of its channel estimation are high. However, its importance cannot be ignored. For example, the accurate acquisition of the arrival angle and transmission angle of the RIS channel are of great significance for RIS reflection control and user positioning. In addition, due to the limitation of RIS complexity, there are still few studies on multi-carrier situations.

[0005] In summary, how to obtain CSI of RIS cascaded channels with high accuracy with small pilot overhead and how to achieve effective estimation of separation channels in RIS system have become difficult problems faced by RIS-assisted MIMO-OFDM system. Summary of the invention

[0006] Technical problem: To solve the above problems, the present invention provides a separation channel reconstruction method based on RIS-assisted MIMO-OFDM system, aiming to achieve more accurate channel estimation under the condition of lower pilot overhead and to achieve separation channel reconstruction with higher accuracy.

[0007] Technical solution: A reconfigurable intelligent surface-assisted MIMO system separation channel reconstruction method of the present invention comprises the following steps:

[0008] Step 1: The user sends pilots in two time periods. In the first time period, a single-time pilot that occupies all subcarriers is used, and the reconfigurable smart surface RIS reflection coefficient is designed to ensure performance. In the second time period, a continuous-time pilot that occupies only some subcarriers is used, and a random RIS reflection coefficient is set.

[0009] Step 2: The reflected signal reaches the base station through RIS, and the base station extracts the azimuth, delay and gain information of all propagation paths in the target user channel; the channel parameter extraction process is divided into two stages. The base station first completes the extraction of the angle and delay of the user-reconfigurable smart surface UE-RIS segment based on the known line-of-sight LoS channel of the reconfigurable smart surface-base station RIS-BS segment;

[0010] Step 3: Based on the results in step 2, the base station extracts the angle and delay of the non-line-of-sight NLoS path in the RIS-BS segment, estimates the gains respectively, and finally completes the estimation of all channel parameters;

[0011] Step 4: The base station completes the separation channel reconstruction using the parameters for user data transmission of the RIS cascade channel.

[0012] in:

[0013] The step 1 is specifically as follows:

[0014] Step 1.1, the user sends the pilot signal in two time periods and adopts different RIS reflection coefficient configurations;

[0015] Step 1.2, for the first time period, the pilot signal sent by the user has a single moment but occupies all subcarriers, and the RIS reflection coefficient is set to ensure reliable channel estimation. Specifically, the design of a discrete Fourier transform DFT matrix is ​​adopted, where the reflection coefficient of the nth RIS unit is Where, N is the number of RIS units;

[0016] Step 1.3, for the second time period, the pilot signal sent by the user has a setting of multiple consecutive moments but occupies part of the subcarriers, and a random RIS reflection coefficient is set.

[0017] The step 2 is specifically as follows:

[0018] Step 2.1, the base station reads the received pilot signal of the first time period, designs a dictionary based on the known LoS channel delay and angle parameters of the RIS-BS segment, and uses the Newton orthogonal matching pursuit algorithm NOMP to complete the delay estimation of each path in the UE-RIS segment, and updates the remaining signal; the base station executes this step repeatedly until the path energy is less than the set threshold, obtains the estimated number of paths and enters the next step;

[0019] Step 2.2, the base station reads the received pilot signal of the second time period, calculates the target matrix according to the known LoS channel delay and angle of the RIS-BS segment and the estimated delay of each path of the UE-RIS segment, designs the dictionary, and uses the NOMP algorithm to complete the angle estimation of the UE-RIS segment;

[0020] Step 2.3, the base station constructs a measurement matrix based on the delay parameters and angle parameters of each path of the UE-RIS segment obtained in steps 2.1 and 2.2, re-estimates the path gain of the UE-RIS segment, and updates the remaining pilot signals of the second time period.

[0021] The step 3 is specifically as follows:

[0022] Step 3.1, the base station obtains the remaining received pilot signals in the first time period, sorts the UE-RIS segment paths according to gain strength, selects the path parameters with the strongest gain, designs a dictionary, uses the NOMP algorithm to estimate the receiving angle and delay of the RIS-BS segment, updates the remaining signals, and calculates the path energy of the RIS-BS segment;

[0023] Step 3.2, the base station reads the remaining received pilot signals in the second time period, calculates the target matrix based on the parameters of the path with the strongest gain in the UE-RIS segment obtained by sorting and the estimated receiving angle and delay parameters of each path in the RIS-BS segment, designs a dictionary, and uses the NOMP algorithm to complete the estimation of the transmission angle of each path in the RIS-BS segment;

[0024] Step 3.3) The base station constructs a measurement matrix based on the transmission angle, reception angle and delay parameters of the RIS-BS segment obtained in steps 3.1 and 3.2, re-estimates the path gain of each path of the RIS-BS segment, and updates the residual signal of the second time period;

[0025] Step 3.4) The base station executes steps 3.1 to 3.3 in a loop until the path energy of the RIS-BS segment obtained in step 3.1 is less than the set threshold Γ, and the estimation ends.

[0026] Beneficial effects: The present invention adopts the above technical solution, which can produce the following technical effects:

[0027] 1. Most of the existing RIS-assisted mobile communication system channel estimation schemes are based on cascade channel estimation. The scheme proposed in the present invention is based on sparse channels and develops a RIS-assisted MIMO OFDM system separation channel parameter estimation algorithm, which can accurately obtain the angle, delay, and gain information of each path of the separation channel and effectively restore the separation channels on both sides of the RIS.

[0028] 2. The scheme proposed in the present invention also proposes a corresponding pilot protocol and a configuration method of the RIS phase shift matrix for the channel model of the RIS-assisted MIMO OFDM system, which effectively solves the problem of excessive pilot overhead in the channel estimation of the RIS-assisted system and can be further expanded by integrating multiple users. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A pilot protocol diagram based on a RIS-assisted MIMO-OFDM system provided by an embodiment of the present invention;

[0030] Figure 2 A flow chart of a channel parameter estimation algorithm based on a RIS-assisted MIMO-OFDM system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0031] In the following description, specific details such as a specific pilot protocol are provided for illustration rather than limitation to provide a thorough understanding of the present invention. However, it should be clear to those skilled in the art that the present invention may also be implemented in other embodiments without these specific details.

[0032] A reconfigurable intelligent surface-assisted MIMO system separation channel reconstruction method of the present invention comprises the following steps:

[0033] Step 1: The user sends pilots in two time periods. In the first time period, a single-time pilot that occupies all subcarriers is used, and the reconfigurable smart surface RIS reflection coefficient is designed to ensure performance. In the second time period, a continuous-time pilot that occupies only some subcarriers is used, and a random RIS reflection coefficient is set.

[0034] Step 2: The reflected signal reaches the base station through RIS, and the base station extracts the azimuth, delay and gain information of all propagation paths in the target user channel; the channel parameter extraction process is divided into two stages. The base station first completes the extraction of the angle and delay of the user-reconfigurable smart surface UE-RIS segment based on the known line-of-sight LoS channel of the reconfigurable smart surface-base station RIS-BS segment;

[0035] Step 3: Based on the results in step 2, the base station extracts the angle and delay of the non-line-of-sight NLoS path in the RIS-BS segment, estimates the gains respectively, and finally completes the estimation of all channel parameters;

[0036] Step 4: The base station completes the separation channel reconstruction using the parameters for user data transmission of the RIS cascade channel.

[0037] In order to make the purpose, technical solution and advantages of the present invention clearer, the following will be described in detail with reference to the accompanying drawings. The embodiment of the present invention provides a separation channel reconstruction method based on a RIS-assisted MIMO-OFDM system, wherein the base station (BS) and the RIS are both uniform linear arrays (ULA), and are respectively equipped with M antennas and N reflection units. A single antenna user receives N s The subcarriers transmit the signal to the base station via RIS, occupying T time slots in total.

[0038] The received signal of the base station is expressed as in Z=[z1 … z T ] are respectively the carrier coefficient matrix, RIS coefficient matrix and noise coefficient matrix. is the combined gain, is the cascade gain, is the phase shift vector, is the phase shift coefficient of RIS at the tth moment, z t is the noise vector at the tth moment, L ur and L rb are the number of UE-RIS paths and the number of RIS-BS paths, and the subscripts correspond to l1 and l2 respectively. is the base station angle vector, d is the base station antenna spacing, and λ is the wavelength of the carrier; set is the OFDM delay vector, Δf is the subcarrier spacing; set is the RIS angle vector, and r is the RIS element spacing. is the gain, receiving angle, delay, and transmitting angle of the qth RIS-BS path, is the gain, receiving angle and delay of the pth UE-RIS path. In an example, the base station receiving signal at the tth time can be expressed as

[0039]

[0040] Step 1: The user sends pilots in two time periods. In the first time period, a single-time multi-carrier pilot is used, and a special RIS reflection coefficient is designed to ensure performance. In the second time period, a continuous-time few-carrier pilot is used, and a random RIS reflection coefficient is set.

[0041] 1.1) For the first time period, if Figure 1 As shown in the dotted square on the left, the pilot signal sent by the user has a single moment but occupies all subcarriers, and a special RIS reflection coefficient is set to ensure reliable channel estimation. Specifically, a DFT-like matrix design is adopted, where the reflection coefficient of the nth RIS unit is

[0042]

[0043] and similar settings;

[0044] 1.2) For the second time period, the pilot signal sent by the user has a setting of multiple consecutive moments but occupies part of the subcarriers, such as Figure 1 As shown in the diagonal square on the right, the preset N u subcarriers, and the actual upper limit of the number of paths of the UE-RIS subchannel Related. It is believed that when This is the minimum requirement to complete the estimation, otherwise it may cause path separation errors and estimation errors. The RIS unit sets random reflection coefficients and similar configurations.

[0045] Step 2: The reflected signal reaches the base station through RIS, and the base station extracts the directional angle, delay and gain information of all propagation paths in the target user channel; the channel parameter extraction process is divided into two stages. The first stage completes the channel parameter estimation based on LoS. The base station completes the extraction of the corresponding UE-RIS segment angle and delay based on the known LoS channel of the RIS-BS segment, and estimates the path gain of the UE-RIS segment. The specific steps are as follows:

[0046] 2.1) The base station reads the received signal y in the first time period 1,res =y1, based on the known LoS channel delay and angle parameters of the RIS-BS segment, with an oversampling rate of β τ Configure the delay sampling vector And design the dictionary

[0047]

[0048] And use Newton's orthogonal matching pursuit algorithm (NOMP), with D1 as the dictionary and y 1,res is the input signal to be estimated, and the delay estimate of the l1th path in the UE-RIS segment is obtained. and the combined gain estimate And update the remaining signal y 1,res ,in is the base station receiving angle and delay of the known RB-RIS LoS path, y 1,res is the residual signal of the first time period set in step 1. The base station executes this step repeatedly until the path energy is less than the set threshold Γ, that is,

[0049]

[0050] The loop ends when , and the total number of estimated UE-RIS paths is obtained. And go to step 2.2.

[0051] 2.2) The base station reads the received signal Y in the second time period 2:T , based on the known LoS channel delay and angle of the RIS-BS segment and the delay of the UE-RIS segment obtained in step 2.1, calculate the phase shift vector in For using N u The target matrix is ​​calculated as

[0052]

[0053] Oversampling rate Configure the angle sampling vector And design sub-dictionaries Consider the RIS phase shift matrix C from 2 to T 2:T , Design Dictionary

[0054]

[0055] Will As the input signal to be estimated by the NOMP algorithm, D2 is used as the dictionary to obtain the estimated receiving angle of the l1th path in the UE-RIS segment. and the cascade gain estimate Total Cycle This process is repeated until the paths of all UE-RIS segments are estimated.

[0056] 2.3) The base station constructs the measurement matrix based on all the estimated parameters obtained in 2.1 and 2.2 in Re-estimate the path gain of the UE-RIS segment

[0057]

[0058] And update the remaining signal of the second time period in is the received signal of the second time period Merge the reorganized results by column to get the updated residual signal Y′ res After that, reverse operation to get Y 2:T,res , where res represents the residual signal.

[0059] Step 3: In the second stage, the channel parameter estimation based on NLoS is completed. The base station extracts the NLoS path angle and delay of the RIS-BS segment according to the estimation results in step 2, and estimates the path gain of the RIS-BS segment.

[0060] 3.1) The base station obtains the remaining received signal y in the first time period 1,res , extract the parameters of the path with the strongest gain in the UE-RIS segment in step 2, the oversampling rate β θ Configure the delay sampling vector Design Dictionary

[0061]

[0062] Using the NOMP algorithm, D3 is used as the dictionary and y is used as the 1,res is the input signal to be estimated, and the delay estimate of the l2th path in the RIS-BS segment is obtained. Angle estimate and the combined gain estimate And update the remaining signal y 1,res .

[0063] If the estimated gain satisfies

[0064]

[0065] Then proceed to steps 3.2 and 3.3 to continue estimating other parameters, otherwise stop the entire step 3 and the estimation ends.

[0066] 3.2) The base station reads the remaining received signal Y in the second time period 2:T,res , using the estimated parameters of the path with the strongest gain in the UE-RIS segment (the paths are sorted by the strength of the path gain by default, so the index 1 is selected to represent the path with the strongest gain) and the parameters obtained in 3.1, calculate the phase shift vector Calculate the target matrix

[0067]

[0068] Oversampling rate Configure the angle sampling vector Design RIS Perspective Dictionary And build a dictionary based on the phase shift matrix of RIS

[0069]

[0070] Will As the input signal to be estimated by the NOMP algorithm, using D4 as the dictionary, the estimated value of the transmission angle of the l2th path in the RIS-BS segment is obtained. and the cascade gain estimate

[0071] 3.3) The base station constructs the measurement matrix W based on the parameters obtained in 3.1 and 3.2 to satisfy Re-estimate the path gain of the RIS-BS segment

[0072]

[0073] And update the remaining signal of the second time period Where Y′ is the received signal Y in the second time period 2:T,res Merge and reorganize the results by column, and get the updated signal Y′ res Reverse operation to get the updated Y 2:T,res .

[0074] Step 4: The base station uses the estimated parameters to complete the reconstruction of the cascade channel:

[0075]

[0076] Reconstruction of UE-RIS channel:

[0077]

[0078] And reconstruction of RIS-BS channel:

[0079]

[0080] It is also used for user data transmission and other related technologies in RIS cascade channels.

[0081] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for reconfigurable intelligent surface-assisted MIMO system separation channel reconstruction, characterized in that: The method comprises the following steps: Step 1: The user sends pilots in two time periods. In the first time period, a single-time pilot that occupies all subcarriers is used, and the reconfigurable smart surface RIS reflection coefficient is designed to ensure performance. In the second time period, a continuous-time pilot that occupies only some subcarriers is used, and a random RIS reflection coefficient is set. Step 2: The reflected signal reaches the base station through RIS, and the base station extracts the azimuth, delay and gain information of all propagation paths in the target user channel; the channel parameter extraction process is divided into two stages. The base station first completes the extraction of the angle and delay of the user-reconfigurable smart surface UE-RIS segment based on the known line-of-sight LoS channel of the reconfigurable smart surface-base station RIS-BS segment; Step 3: Based on the results in step 2, the base station extracts the angle and delay of the non-line-of-sight NLoS path in the RIS-BS segment, estimates the gains respectively, and finally completes the estimation of all channel parameters; Step 4: The base station uses the parameters to complete the separation channel reconstruction for user data transmission of the RIS cascade channel; in: The step 1 is specifically as follows: Step 1.1, the user sends the pilot signal in two time periods and adopts different RIS reflection coefficient configurations; Step 1.2, for the first time period, the pilot signal sent by the user has a single moment but occupies all subcarriers, and the RIS reflection coefficient is set to ensure reliable channel estimation. Specifically, the design of a discrete Fourier transform DFT matrix is ​​adopted, where the reflection coefficient of the nth RIS unit is Where, N is the number of RIS units; Step 1.3, for the second time period, the pilot signal sent by the user has a setting of multiple consecutive moments but occupying part of the subcarriers, and a random RIS reflection coefficient is set; The step 2 is specifically as follows: Step 2.1, the base station reads the received pilot signal of the first time period, designs a dictionary based on the known LoS channel delay and angle parameters of the RIS-BS segment, and uses the Newton orthogonal matching pursuit algorithm NOMP to complete the delay estimation of each path in the UE-RIS segment, and updates the remaining signal; the base station executes this step repeatedly until the path energy is less than the set threshold, obtains the estimated number of paths and enters the next step; Step 2.2, the base station reads the received pilot signal of the second time period, calculates the target matrix according to the known LoS channel delay and angle of the RIS-BS segment and the estimated delay of each path of the UE-RIS segment, designs the dictionary, and uses the NOMP algorithm to complete the angle estimation of the UE-RIS segment; Step 2.3, the base station constructs a measurement matrix according to the delay parameters and angle parameters of each path of the UE-RIS segment obtained in step 2.1 and step 2.2, re-estimates the path gain of the UE-RIS segment, and updates the remaining pilot signals of the second time period; The step 3 is specifically as follows: Step 3.1, the base station obtains the remaining received pilot signals in the first time period, sorts the UE-RIS segment paths according to gain strength, selects the path parameters with the strongest gain, designs a dictionary, uses the NOMP algorithm to estimate the receiving angle and delay of the RIS-BS segment, updates the remaining signals, and calculates the path energy of the RIS-BS segment; Step 3.2, the base station reads the remaining received pilot signals in the second time period, calculates the target matrix based on the parameters of the path with the strongest gain in the UE-RIS segment obtained by sorting and the estimated receiving angle and delay parameters of each path in the RIS-BS segment, designs a dictionary, and uses the NOMP algorithm to complete the estimation of the transmission angle of each path in the RIS-BS segment; Step 3.3) The base station constructs a measurement matrix based on the transmission angle, reception angle and delay parameters of the RIS-BS segment obtained in steps 3.1 and 3.2, re-estimates the path gain of each path of the RIS-BS segment, and updates the residual signal of the second time period; Step 3.4) The base station executes steps 3.1 to 3.3 in a loop until the path energy of the RIS-BS segment obtained in step 3.1 is less than the set threshold Γ, and the estimation ends.

Citation Information

Patent Citations

  • Optimized sparse antenna activation reconfigurable intelligent surface auxiliary communication method

    CN112564752A

  • Channel estimation method of intelligent reconfigurable surface-assisted millimeter wave MIMO system

    CN114745232A