RIS auxiliary safety wireless communication system based on channel knowledge map and environment sensing beam selection method

Abstract

The invention relates to an RIS auxiliary safety wireless communication system based on a channel knowledge map and an environment perception beam selection method. The invention relates to an RIS auxiliary secure wireless communication environment sensing beam selection method based on a channel knowledge map, which reduces the training overhead on the premise of ensuring the RIS auxiliary communication secrecy rate: considering an RIS auxiliary secure communication system, and obtaining position information and beam pairs corresponding to users in a region by using an environment sensing technology; using an abstract function to link a beam pair in a given environment with a user position, and storing the beam pair and the user position in a CKM; the method comprises the following steps: for a downlink of an RIS assisted secure communication system, selecting a required optimal beam pair from a CKM under the condition that the positions of a BS and an RIS are fixed; an eavesdropper is considered in a communication environment, the selected beam pair is input into an alternating optimization AO algorithm, and the alternating optimization AO algorithm is adopted to carry out iterative optimization on the beam to obtain a final beam pair; the algorithm complexity and the number of iterations are analyzed, and the optimization rate is improved.

Description

Technical Field

[0001] The invention relates to a RIS wireless auxiliary communication system, and in particular to an environment-aware beam selection method for RIS-assisted secure wireless communication based on a channel knowledge map. Background Art

[0002] Reconfigurable smart surface (RIS)-assisted communication systems are a promising technology in secure wireless communication systems that can play a key role in improving the security of wireless communication systems by reducing the impact of eavesdroppers. By intelligently controlling the reflection of signals, RIS can create a favorable propagation environment for legitimate users while interfering or weakening the signal path of potential eavesdroppers. This selective signal enhancement and suppression helps ensure the security of communications between authorized parties and makes it more difficult for eavesdroppers to intercept or decipher transmitted information.

[0003] RIS has significant advantages in wireless communication, but the performance of RIS depends on accurate perfect channel state information (CSI). However, since RIS is a passive device, it does not have the ability to process or feedback channel information. Therefore, the strong dependence of RIS wireless communication on perfect CSI and the high cost of channel estimation are the main challenges that need to be faced in its practical application. In practice, on the one hand, perfect channel state information (CSI) is required to achieve the maximum communication rate, and on the other hand, it is very difficult to obtain accurate perfect CSI for RIS-assisted communication. Although various channel estimation methods have been proposed, they usually require high training overhead and increase with the increase of the number of RIS components.

[0004] At the same time, recent advances in positioning and sensing technologies have increased interest in utilizing location information and geo-location-based databases in wireless communication systems. One such concept is CKM, which leverages historical data and context-aware technology to efficiently build and update channel models, reducing the need for real-time CSI and reducing the complexity and cost of channel estimation. Summary of the invention

[0005] Aiming at the deficiencies of the prior art, the present invention proposes a new method for RIS-assisted communication by integrating an alternating optimization (AO) algorithm based on a channel knowledge map (CKM). The method does not require obtaining real-time CSI, reduces the difficulty of RIS-assisted communication in practice, and greatly reduces the training overhead while ensuring the confidentiality rate.

[0006] The technical solution adopted by the present invention is:

[0007] A RIS-assisted secure wireless communication environment-aware beam selection method based on channel knowledge map reduces the training overhead while ensuring the confidentiality rate of RIS-assisted communication through the following steps:

[0008] Step S1, consider a RIS-assisted secure communication system, use environmental sensing technology to obtain the location information and beam pairs corresponding to users in the area, use an abstract function to associate the beam pairs in a given environment with the user locations, and store them in the CKM;

[0009] Step S2, when the positions of the BS and the RIS are fixed, selecting the required optimal beam pair from the CKM;

[0010] Step S3, considering the eavesdropper in the communication environment, inputting the selected beam pair into the alternating optimization AO algorithm, and using the alternating optimization AO algorithm to iteratively optimize the beam to obtain the final beam pair;

[0011] Step S4, analyzing the algorithm complexity and the number of iterations to improve the optimization rate.

[0012] In step S1, environmental sensing technology is used to extract environmental features to obtain location and type features including obstacles and reflective surfaces. The CKM information is stored in a database and is only updated when there are significant changes in the environment. Such changes can be detected by using environmental perception nodes.

[0013] For the downlink of a RIS-assisted secure communication system, it is assumed that the BS has M >> 1 antennas and the RIS has N >> 1 passive reflective elements to assist the communication between the BS and the UE that may be moving in a specific area;

[0014] Will Represented as the BS active transmit beamforming vector

[0015] Where E = diag (e 1 ,...,e M ), where e m is the reflection coefficient of the mth reflection unit of RIS;

[0016] Based on CKM, select the optimal beam pair Design an active / passive beamforming scheme for optimizing environmental perception against eavesdroppers; consider the situation where there is an error in the user position, and use the alternating optimization AO algorithm to analyze the impact of the user position error on the accuracy of CKM; step S3 includes: S3.1, fix e to optimize f; S3.2, fix f to optimize e.

[0017] Consider the signal transmitted by BS as

[0018] x=fs+m t(1)

[0019] Where s is a confidential independent Gaussian data symbol, and is an independent Gaussian transmission distortion noise. The power of the distortion noise at each antenna is proportional to its transmission signal power, that is, where μ t ≥0 is the ratio of the transmission distortion noise power to the transmission signal power; the channels between BS and RIS, BS and legal users, and RIS and legal users are respectively express;

[0020] Then the downlink received signals at the legitimate user and the eavesdropper can be expressed as

[0021] y U =y U +m U (2)

[0022]

[0023] in n U and n E Both The additive Gaussian white noise of the distribution, the noise variance is and m U is the independent zero-mean Gaussian distortion noise at the legitimate user, following The distribution of μ r ≥0 is the ratio of distorted noise power to undistorted received signal power; They are the channel between BS and eavesdropper and the channel between IRS and eavesdropper respectively;

[0024] Assuming that the BS knows all the CSIs in the entire network, the confidentiality rate of the system can be expressed as

[0025] R=[R U -R E ] + , (4)

[0026] in R U is the communication rate of legitimate users, expressed as

[0027]

[0028] In the formula is the equivalent reflection coefficient vector, is the total power of the noise, G U is an equivalent channel, expressed as

[0029]

[0030] R E is the eavesdropper's communication rate, expressed as

[0031]

[0032] In the formula is the total power of the noise, G E is an equivalent channel, expressed as

[0033]

[0034] Ideally, when If perfect CSI is available on all channel coherent blocks, the communication rate can be written as

[0035]

[0036] (f m ,e m ) is the optimal beam pair selected from the channel knowledge map, that is: and are the beam sets in CKM;

[0037] Beam selection via CKM

[0038] When the BS and RIS are located at fixed positions, the equivalent cascade channel G U Mainly with the terminal position (denoted as Q U ) and the actual propagation environment (denoted as E U ) changes; therefore, G U It can be expressed as

[0039] G U =g 1 (Q U ,E U ) (10)

[0040] where g 1 is an unknown function that uses CKM to map any possible UE location and its channel knowledge specific to that location;

[0041] The optimal beam pair can be expressed as an abstract function of the terminal location and propagation environment:

[0042] (f m ,e m ) = g 2 (Q U ,E U ) (11)

[0043] For a given communication environment, the above formula can be further simplified to

[0044] (f m ,e m )≈g 3 (Q U ) (12)

[0045] g 3 (,.,) is an abstract function that relates the beam pairs in a given environment to the UE location.

[0046] In the RIS-assisted secure wireless communication environment-aware beam selection method based on channel knowledge map, in step S3, the alternating optimization AO algorithm is used to maximize the confidentiality rate R by jointly optimizing the beamforming vector at the BS and the passive beamforming at the RIS. In this work, this problem is formulated as

[0047]

[0048] Where (13b) is the transmit power constraint, the maximum transmit power P max At BS, (13c) imposes unit modulus on each term in e, setting The optimal value of problem (13) is always non-negative; if R U -R E <0, then set Make the value of the objective function 0;

[0049] Introduce auxiliary variable p = [p 1 ,p 2 ,p 3 ,p 4 ] T , by R U ≥p 1 -p 2 and R E ≤p 3 -p 4 , problem (13) can be restated as

[0050]

[0051] (14a)

[0052]

[0053] log 2 (Φ U (f,e))≤p 2 , (14e)

[0054]

[0055] log(Φ E (f,e))≥p4 . (14g)

[0056] It is easy to prove that problem (14) is equivalent to problem (13).

[0057] Step S3.1, fix e and optimize f;

[0058] When e is fixed, we further introduce an auxiliary variable k f =[k f1 ,k f2 ,k f3 ,k f4 ] T , so that the non-convex constraints (15), (16), (17) and (18) are equivalent to

[0059]

[0060] From (15), (16), (17) and (18), we can see that constraints (15a), (16b), (17b) and (18a) are convex, while (15b), (16a), (17a) and (18b) are non-convex; according to y H diag(xx H )y=x H diag(yy H )x, we can derive the following equivalent constraints:

[0061]

[0062] In the formula A 2 =0. and f n is the solution obtained at the nth iteration;

[0063] Finally, the subproblem of solving f is expressed as

[0064]

[0065] st(13b),(15a),(16b),(17b),(18a),(19). (20b)

[0066] Problem (20) is a SOCP and is solved using the CVX tool.

[0067] Step S3.2, fix f and optimize e:

[0068] To solve the non-convex unit modulus constraint The semidefinite relaxation (SDR) technique is used to update e; by defining a new variable E = ee H , constrain Replace with

[0069] After fixing f and introducing a new auxiliary variable k e =[k e1 ,k e2 ,k e3 ,k e4 ] T In the case of , the subproblem of solving e can be expressed as

[0070]

[0071] log 2 (k e1 )≥p 1 ,(21c)

[0072]

[0073] The SDR method is applied to further relax problem (21) into a convex SDP, and then solved using the CVX tool;

[0074] In order to ensure that the target value sequence generated by each iteration does not decrease, the following update method is used

[0075]

[0076] When e n+1 =e n When R(f n+2 ,e n+1 )>R(f n+1 ,e n+1 )=R(f n+1 ,e n ), the iterative process ends until convergence.

[0077] The present invention provides a RIS-assisted secure wireless communication environment perception beam selection method based on a channel knowledge map. In step S4, the complexity and number of iterations of the CKM and AO joint optimization algorithm are analyzed, and the optimal beam pair (f) formed by active / passive environmental perception based on CKM is obtained. m ,e m ), use the following formula (23) to solve the problem of maximum confidentiality rate R

[0078]

[0079] Thus, the best beam pair is optimized through the AO algorithm and the maximum confidentiality rate R is calculated;

[0080] The target value sequence {R(F n ,e n )} is non-decreasing, where

[0081]

[0082] In addition, due to the bounded transmit power constraint, {R(F n ,e n )} has a finite upper bound, which guarantees the convergence of the AO algorithm; the complexity of solving problem (23) is O(N 3 )+O(M 3.5 ).

[0083] Beneficial effects of the invention:

[0084] 1. The present invention integrates the alternating optimization (AO) algorithm based on the channel knowledge map (CKM) to propose a new method for RIS-assisted communication. The method combining the channel knowledge map and the alternating optimization AO algorithm is applied to RIS-assisted communication, which reduces the computational complexity. This method does not require the acquisition of real-time CSI, reduces the difficulty of RIS-assisted communication in practice, and greatly reduces the training overhead while ensuring the confidentiality rate.

[0085] 2. The RIS-assisted secure wireless communication beam selection method of the present invention is based on the channel knowledge map. The CKM is used to retrieve the wireless propagation characteristics in the environment, obtain the location information and beam pairs corresponding to the users in the area, and realize the RIS-assisted secure wireless communication environment-aware beam selection. The historical data and environment-aware technology are used to use abstract functions to link the beam pairs in a given environment with the user positions and store them in the CKM. The channel model can be effectively established and updated without the need for channel estimation, reducing the demand for real-time CSI, and reducing the computational complexity and cost of channel estimation in the RIS-assisted communication system.

[0086] 3. The present invention is a RIS-assisted secure wireless communication beam selection method based on a channel knowledge map. CKM is a method that does not require training and does not require channel estimation. CKM can be obtained by extracting environmental features using environmental sensing technology, such as the location and type of obstacles and reflective surfaces, and then mapping these features to wireless propagation features in the environment. CKM can be stored in a database and updated only when there are significant changes in the environment, which can be detected by using environmental sensing nodes. The complexity of CKM acquisition mainly depends on the complexity of the environmental sensing technology used and the frequency of CKM updates. However, once the CKM is acquired and stored, the complexity of using it for environmentally aware communications is low because the wireless propagation characteristics in the environment can be easily retrieved from the CKM. In RIS-assisted communication systems, traditional channel estimation requires huge computational overhead, and the use of CKM can greatly reduce computational complexity.

[0087] 4. The present invention considers a RIS-assisted secure communication system. Under the condition of determining the user's location, the required optimal beam pair is selected from the CKM, and the eavesdropper is considered in the communication environment. The selected beam pair is input into the alternating optimization AO algorithm, and the final beam pair is obtained by iterative optimization. Finally, the algorithm complexity and the number of iterations are analyzed to achieve an improvement in the optimization rate. The advantage of the present invention is that there is no need to obtain real-time CSI, which reduces the difficulty of RIS-assisted communication in practice, greatly reduces the training overhead while ensuring the confidentiality rate, and has strong robustness to errors related to the terminal position in practical applications. The regional communication rate visualization diagram is used to analyze the confidentiality rate, so that the communication rate of legitimate users and eavesdroppers in the area is more intuitive. BRIEF DESCRIPTION OF THE DRAWINGS

[0088] Figure 1 The figure shows a block diagram of the RIS-assisted safety communication system based on environment-aware beamforming of the present invention;

[0089] Figure 2 Shown are the confidentiality rates at different power levels based on CKM and random beam benchmark methods;

[0090] Figure 3 Shown are the regional distribution diagrams of confidentiality rates before (a) and after (b) active / passive beam optimization;

[0091] Figure 4 Shown are the regional distribution diagrams of the eavesdropper's communication rate before (a) and after (b) active / passive beam optimization;

[0092] Figure 5 The figure shows the effective interference range of RIS to eavesdroppers at the same power level;

[0093] Figure 6 The graph showing the relationship between confidentiality rate and transmission power under different hardware losses is shown;

[0094] Figure 7 Shown are the confidentiality rates of CKM with perfect CSI and user position error and the basic method at different powers;

[0095] Figure 8 Graph showing the relationship between confidentiality rate and the number of RIS components;

[0096] Fig. 9 Shown is a flow chart of the RIS-assisted secure communication beam selection method based on environment perception of the present invention. DETAILED DESCRIPTION

[0097] In order to make the technical concept and advantages of the invention more clearly understood, the technical solution of the invention is further described in detail below in conjunction with the accompanying drawings. It should be understood that the following embodiments are only used to explain and illustrate the preferred implementation methods of the invention and should not be regarded as and do not constitute a limitation on the scope of patent protection claimed by the invention.

[0098] Example 1

[0099] See also Fig. 9 The present invention proposes a RIS-assisted secure wireless communication environment-aware beam selection method based on a channel knowledge map, which reduces the training overhead while ensuring the confidentiality rate of RIS-assisted communication through the following steps:

[0100] Step S1, consider a RIS-assisted secure communication system, use environmental sensing technology to obtain the location information and beam pairs corresponding to users in the area, use an abstract function to associate the beam pairs in a given environment with the user locations, and store them in the CKM;

[0101] Step S2, when the positions of the BS and the RIS are fixed, selecting the required optimal beam pair from the CKM;

[0102] Step S3, considering the eavesdropper in the communication environment, inputting the selected beam pair into the alternating optimization AO algorithm, and using the alternating optimization AO algorithm to iteratively optimize the beam to obtain the final beam pair;

[0103] Step S4, analyzing the algorithm complexity and the number of iterations to improve the optimization rate.

[0104] The complexity of CKM acquisition depends mainly on the complexity of the environmental perception technology used and the frequency of CKM updates. However, once the CKM is acquired and stored, the complexity of using it for environmental perception communication is low because the wireless propagation characteristics in the environment can be easily retrieved from the CKM. In RIS-assisted communication systems, traditional channel estimation requires huge computational overhead. Using environmental sensing technology to extract environmental features to obtain location and type features including obstacles and reflective surfaces, and then mapping these features to wireless propagation features in the environment can greatly reduce the computational complexity. Therefore, it can be applied to the problem of interference with eavesdroppers.

[0105] Based on the above, the present invention integrates the existing alternating optimization AO algorithm on the basis of CKM and proposes a new method for RIS-assisted communication. This method does not require obtaining real-time CSI, reduces the difficulty of RIS-assisted communication in practice, and greatly reduces the training overhead while ensuring the confidentiality rate.

[0106] In step S1, environmental sensing technology is used to extract environmental features to obtain location and type features including obstacles and reflective surfaces. The CKM information is stored in a database and is only updated when there are significant changes in the environment. Such changes can be detected by using environmental perception nodes.

[0107] Example 2

[0108] like Figure 1 As shown in the figure, the present invention provides an environment-aware beam selection method for RIS-assisted secure wireless communication based on a channel knowledge map. Considering a downlink of a RIS-assisted secure communication system, a single-antenna eavesdropper intends to eavesdrop on a confidential signal sent by a BS (Basestation) to a single-antenna legitimate user. Assuming that the BS has M>>1 antennas and the RIS has N>>1 passive reflective elements to assist the communication between the BS and the UE (User Equipment, UE for short) that may move in a specific area. It is represented as the BS active transmit beamforming vector. E = diag (e 1 ,...,e M ), where e m is the reflection coefficient of the mth reflection unit of RIS. The object of the present invention is to select the optimal beam pair based on CKM. To design an active / passive beamforming scheme with optimized environment perception against eavesdroppers.

[0109] The present invention considers that the signal transmitted by the BS is

[0110] x=fs+m t , (1)

[0111] Where s is a confidential independent Gaussian data symbol, and is an independent Gaussian transmission distortion noise. The power of the distortion noise at each antenna is proportional to its transmission signal power, that is, where μ t ≥0 is the ratio of the transmission distortion noise power to the transmission signal power. The channels between BS and RIS, BS and legal users, and RIS and legal users are respectively The downlink received signals at the legitimate user and the eavesdropper can be expressed as

[0112] y U =y U +m U , (2)

[0113]

[0114] in n U and nE Both The additive white Gaussian noise (AWGN) of the distribution is and m U is the independent zero-mean Gaussian distortion noise at the legitimate user, following The distribution of μ r ≥0 is the ratio of the distorted noise power to the undistorted received signal power. are the channels between BS and eavesdropper and between IRS and eavesdropper, respectively. In this work, we assume that the BS is fully aware of all CSIs in the entire network.

[0115] The confidentiality rate of the system can be expressed as

[0116] R=[R U -R E ] + , (4)

[0117] in

[0118] R U is the communication rate of legitimate users, expressed as

[0119]

[0120] In the formula is the equivalent reflection coefficient vector, is the total power of the noise, G U is an equivalent channel, expressed as

[0121]

[0122] R E is the eavesdropper's communication rate, expressed as

[0123]

[0124] In the formula is the total power of the noise, G E is an equivalent channel, expressed as

[0125]

[0126] Ideally, when If perfect CSI is available on all channel coherent blocks, the communication rate can be written as

[0127]

[0128] (f m ,e m) is the optimal beam pair selected from the channel knowledge map, that is: and ε are the beam sets in CKM.

[0129] However, achieving the maximum communication rate requires perfect CSI, which is very difficult in practice for IRS-assisted communication. Although various channel estimation methods have been proposed, they usually require high training overhead and increase with the number of IRS elements (N>>1).

[0130] In order to solve the above problems, the present invention integrates the existing alternating optimization AO algorithm on the basis of CKM and proposes a new method for RIS-assisted communication. The method does not need to obtain real-time CSI, reduces the difficulty of RIS-assisted communication in practice, and greatly reduces the training overhead while ensuring the confidentiality rate.

[0131] 1Beam selection via CKM

[0132] For the RIS-assisted communication system under consideration, when the BS and RIS are located at fixed positions, the equivalent cascade channel G U Mainly with the terminal position (denoted as Q U ) and the actual propagation environment (denoted as E U ). Therefore, G U It can be expressed as

[0133] G U =g 1 (Q U ,E U ), (10)

[0134] In practice, due to g 1 is an unknown function, so describing it precisely is challenging. Fortunately, a new approach to this thorny problem is to use CKM, which aims to map any possible UE location and its channel knowledge specific to that location.

[0135] The optimal beam pair can be expressed as an abstract function of the terminal location and propagation environment:

[0136] (f m ,e m ) = g 2 (Q U ,E U ), (11)

[0137] In most practical scenarios, the main components of the radio propagation environment (such as the location, height, and dielectric properties of buildings) vary on a much larger time scale than the terminal location. Therefore, the variation of the wireless channel is mainly due to the mobility of the user. Therefore, for any given communication environment, the above equation can be further simplified to

[0138] (f m ,e m )≈g 3 (Q U ), (12)

[0139] g 3 (,.,) is an abstract function that relates the beam pairs in a given environment to the UE location.

[0140] CKM is a training-free method that does not require channel estimation. Instead, CKM can be obtained by extracting environmental features using environmental sensing technology, such as the location and type of obstacles and reflective surfaces, and then mapping these features to wireless propagation characteristics in the environment. CKM can be stored in a database and updated only when there are significant changes in the environment, which can be detected by using environmental sensing nodes. Regarding the complexity of CKM acquisition, it mainly depends on the complexity of the environmental sensing technology adopted and the frequency of CKM updates. However, once the CKM is acquired and stored, the complexity of using it for environmentally aware communications is low because the wireless propagation characteristics in the environment can be easily retrieved from the CKM. In RIS-assisted communication systems, traditional channel estimation requires huge computational overhead. The use of CKM can greatly reduce the computational complexity, and therefore, it can be applied to the problem of interference with eavesdroppers.

[0141] Since the current positioning technology is not perfect enough, the location information of the terminal user may not be accurate. Therefore, the present invention considers the situation where there is an error in the user's position and analyzes the impact of the user's position error on the accuracy of CKM.

[0142] 2 Application of alternating optimization AO algorithm

[0143] In this work, we propose to maximize the confidentiality rate R by jointly optimizing the beamforming vector at the BS and the passive beamforming at the RIS. This problem is formulated as

[0144]

[0145] Where (13b) is the transmit power constraint, the maximum transmit power P max At BS, (13c) imposes unit modulus on each term in e, setting We note that the optimal value of problem (13) is always non-negative. U -RE <0, then set Make the value of the objective function equal to 0.

[0146] We introduce an auxiliary variable p = [p 1 ,p 2 ,p 3 ,p 4 ] T , by R U ≥p 1 -p 2 and R E ≤p 3 -p 4 , problem (13) can be restated as

[0147]

[0148] log 2 (Φ U (f,e))≤p 2 , (14e)

[0149]

[0150] log(Φ E (f,e))≥p 4 .(14g)

[0151] It is easy to prove that problem (14) is equivalent to problem (13).

[0152] 1) Fix e and optimize f

[0153] When e is fixed, we further introduce an auxiliary variable k f =[k f1 ,k f2 ,k f3 ,k f4 ] T , so that the non-convex constraints (15), (16), (17) and (18) are equivalent to

[0154]

[0155] From (15), (16), (17) and (18), we can see that constraints (15a), (16b), (17b) and (18a) are convex, while (15b), (16a), (17a) and (18b) are non-convex. H diag(xx H )y=x H diag(yy H )x, we can derive the following equivalent constraints:

[0156]

[0157]

[0158] In the formula A 2 =0. and f n is the solution obtained at the nth iteration.

[0159] Finally, the subproblem of solving f is expressed as

[0160]

[0161] st(13b),(15a),(16b),(17b),(18a),(19). (20b)

[0162] Problem (20) is a SOCP and can be solved using the CVX tool.

[0163] 2) Fix f and optimize e

[0164] To solve the non-convex unit modulus constraint We use the semidefinite relaxation (SDR) technique to update e. In particular, by defining a new variable E = ee H , constrain Replace with After fixing f and introducing a new auxiliary variable k e =[k e1 ,k e2 ,k e3 ,k e4 ] T In the case of , the subproblem of solving e can be expressed as

[0165]

[0166] log 2 (k e1 )≥p 1 , (21c)

[0167]

[0168] The SDR method is applied to further relax the problem (21) into a convex SDP and then solved using the CVX tool. However, the SDR method generates a suboptimal solution There is no guarantee that the target value of problem (13) in the (n+1)th iteration is not less than that in the previous iteration. Therefore, in order to ensure that the target value sequence generated in each iteration does not decrease, we adopt the following update method

[0169]

[0170] Note that when e n+1 =e n When we still have R(f n+2 ,e n+1 )>R(f n+1 ,e n+1 )=R(f n+1 ,e n ), the iteration process ends until convergence. The reason is that in f n+1 and e n Based on this, we can get the optimal f for the next iteration. n+2 .

[0171] 3CKM and AO joint optimization and algorithm analysis

[0172] Through CKM, we can obtain the optimal beam pair (f m ,e m ), then the problem of solving the maximum confidentiality rate R can be expressed as

[0173]

[0174] The AO algorithm is then used to optimize the best beam pair and calculate the maximum confidentiality rate R. The corresponding algorithms are summarized in Table 1.

[0175] We declare that the target value sequence {R(F n ,e n )} is non-decreasing. Among them,

[0176]

[0177] In addition, due to the bounded transmit power constraint, {R(F n ,e n )} has a finite upper bound. Therefore, the convergence of the AO algorithm is guaranteed. The complexity of solving problem (23) is O(N 3 )+O(M 3.5 ).

[0178] Based on the concept of CKM and the location of user UE, we can achieve the maximum confidentiality rate R. The BS is located at (0m, 0m) and the RIS is located at (50m, 0m), where users are randomly distributed in an area of ​​60m×60m. We set the number of BS antennas to 4 and the RIS consists of 8×8 passive reflective units. Large-scale path loss is expressed as PL = -30-10αlog 10(d) dB, where α is the path loss exponent and d is the link length in meters. The path loss exponent of the channel associated with RIS is set to 2.2, while the path loss exponent of the channel for direct BS-UE and BS-EVE is set to 3.6. Small-scale fading follows a Rayleigh distribution, with a Rayleigh coefficient of 10 for the channel associated with RIS and 0 for the direct link.

[0179] Figure 2 The confidentiality rates of the CKM-based and random beam forming benchmark methods at different power levels are shown. It can be observed that higher power levels lead to higher confidentiality rates, and the confidentiality rate of the CKM-based method is generally better than that of the benchmark method. The figure also shows that the confidentiality rates of the two methods tend to converge after about 40 iterations, proving the convergence of the present invention.

[0180] Figure 3 The regional distribution of confidentiality rates before and after active / passive beam optimization is shown, where the user is located at (50m, 2m) and e and f are fixed to their optimal values ​​at this location. It can be seen that the confidentiality rate is higher at locations close to the user and BS before optimization, while it approaches 0 at other locations. After optimization, the confidentiality rate of the entire area is at a higher level due to the interference effect of RIS on the eavesdropper's communication rate.

[0181] Figure 4 The regional distribution of the eavesdropper's communication rate before and after active / passive beam optimization is shown. The eavesdropper is located at (45m, 2m). It can be seen that before optimization, the eavesdropper's communication rate in the entire area is high, but after optimization, due to the interference effect of RIS, the communication rate at the eavesdropper's location is significantly reduced.

[0182] Figure 5 The effective interference range of RIS on eavesdroppers at different power levels is shown. The figure shows the relationship between the communication rate and power of the eavesdropper within different radii with the eavesdropper's position (45m, 2m) as the center. It can be observed that the communication rate of the eavesdropper increases with the increase of radius. This is because RIS provides the best interference effect at the eavesdropper's position, but the interference effect gradually weakens with the increase of distance. In addition, the communication rate of the eavesdropper also increases with the increase of power, and the rate growth trend is faster within a larger radius. This shows that the interference effect of RIS on eavesdroppers is more significant at lower base station power than at higher base BS power levels.

[0183] Most studies on RIS-assisted secure communication assume that the legitimate user has perfect hardware, which is too idealistic in practice. Actual radio frequency (RF) components inevitably have hardware defects, including phase noise, quantization error, amplification noise, and nonlinearity. Therefore, we consider the legitimate user to have hardware losses. Since the BS cannot fully obtain the knowledge about the eavesdropper, we consider the worst case, that is, the eavesdropper has high-quality hardware with no residual hardware losses. Figure 6 The figure shows the relationship between the confidentiality rate and the maximum transmission power under different hardware loss conditions when N=4 and M=64. Figure 6 It can be seen that with the increase of transmission power, confidentiality first increases and then tends to stabilize. This is because the distortion noise caused by hardware loss is proportional to the power of the transmitted and received signals, resulting in an upper limit for the confidentiality rate as a function of the maximum transmission power. In addition, as the hardware loss increases, the confidentiality rate decreases as the power increases. It can be seen that the impact of hardware loss on the confidentiality rate becomes more obvious as the power increases. This shows that in actual RIS-assisted secure communications, hardware loss may lead to performance degradation and also affect the accuracy of CKM. The channel estimation of traditional RIS-assisted communication systems requires huge computational overhead, and the use of CKM can greatly reduce the computational complexity. Using a regional communication rate visualization diagram to analyze the confidentiality rate makes the communication rates of legitimate users and eavesdroppers in the area more intuitive.

[0184] Considering that the current positioning technology is not perfect enough, the location information of the terminal user may not be accurate. Therefore, the present invention explores the influence of the user location error on the accuracy of CKM. Figure 7 The confidentiality rates of CKM with perfect CSI and average user location errors of 0.5m and 1m and the baseline method based on random beam generation are shown at different powers, where the BS transmit power P varies from 20dBm to 45dBm. 50 users randomly distributed in a circle with a center of (50m, 2m) and a radius of 0.5m and 1m are taken as users with location errors. It can be seen that the confidentiality rate increases with the increase of power. As the average user location error increases, the confidentiality rate decreases but is still better than the baseline method. The results show that the proposed scheme is robust to the terminal user location error.

[0185] Figure 8 The confidentiality rates of CKM with perfect CSI, average user location error of 0.5m and 1m, and the baseline method based on random beam generation are shown for different numbers of RIS reflective elements, with BS power of 30dBm. It can be observed that the confidentiality rate increases with the number of RIS elements, which indicates that the interference effect against eavesdroppers increases with the increase in the number of RIS elements.

[0186] Table 1 CKM integrated AO algorithm process

[0187]

[0188]

[0189] Example 3

[0190] The present invention also proposes a RIS-assisted secure wireless communication system. The RIS-assisted secure wireless communication system uses the RIS-assisted secure wireless communication environment-aware beam selection method based on the channel knowledge map as described above, based on CKM, using historical data and environment perception technology to extract and obtain environmental features including the location and type of obstacles and reflective surfaces, and then maps these features to wireless propagation features in the environment; establishes and updates the channel model.

[0191] RIS has significant application advantages in wireless communications, but its strong dependence on perfect CSI and high-cost channel estimation are major challenges. The present invention adopts a method combining channel knowledge map and alternating optimization AO algorithm for RIS-assisted communication, and proposes a RIS-assisted secure wireless communication system and an environment-aware beam selection method based on channel knowledge map. CKM is a method that does not require training and does not require channel estimation, which reduces the computational complexity. The advantage of the present invention is that it does not require real-time CSI, which reduces the difficulty of RIS-assisted communication in practice, greatly reduces training overhead while ensuring confidentiality, and has strong robustness to errors related to terminal position in practical applications.

Claims

1. A RIS-assisted secure wireless communication environment-aware beam selection method based on a channel knowledge map, characterized in that: Through the following steps, the training cost of RIS-assisted communication can be reduced while ensuring the confidentiality rate: Step S1, consider a RIS-assisted secure communication system, use environmental sensing technology to obtain the location information and beam pairs corresponding to users in the area, use an abstract function to associate the beam pairs in a given environment with the user locations, and store them in the CKM; Step S2, when the positions of the BS and the RIS are fixed, selecting the required optimal beam pair from the CKM; Step S3, considering the eavesdropper in the communication environment, inputting the selected beam pair into the alternating optimization AO algorithm, and using the alternating optimization AO algorithm to iteratively optimize the beam to obtain the final beam pair; Step S4, analyzing the algorithm complexity and the number of iterations to improve the optimization rate.

2. According to claim 1, the RIS-assisted secure wireless communication environment-aware beam selection method based on channel knowledge map is characterized in that: In step S1, environmental sensing technology is used to extract environmental features to obtain location and type features including obstacles and reflective surfaces. The CKM information is stored in a database and is only updated when there are significant changes in the environment. Such changes can be detected by using environmental perception nodes.

3. The RIS-assisted secure wireless communication environment-aware beam selection method based on channel knowledge map according to claim 1 or 2, characterized in that: In step S3, the optimal beam pair is selected based on CKM. Design active / passive beamforming schemes to optimize environmental perception against eavesdroppers: Considering the situation where there is error in user position, the alternating optimization AO algorithm is used to analyze the impact of user position error on CKM accuracy; It includes: Step S3.1, fix e and optimize f; Step S3.2, fix f and optimize e.

4. The RIS-assisted secure wireless communication environment-aware beam selection method based on channel knowledge map according to claim 3 is characterized in that: For the downlink of a RIS-assisted secure communication system, it is assumed that the BS has M >> 1 antennas and the RIS has N >> 1 passive reflective elements to assist the communication between the BS and the UE that may be moving in a specific area; Will Represented as the BS active transmit beamforming vector Where E = diag(e1,...,e M ), where e m is the reflection coefficient of the mth reflection unit of RIS; Based on CKM, select the optimal beam pair Design an active / passive beamforming scheme to optimize environmental awareness against eavesdroppers; Consider the signal transmitted by BS as x=fs+m t (1) Where s is a confidential independent Gaussian data symbol, and is an independent Gaussian transmission distortion noise. The power of the distortion noise at each antenna is proportional to its transmission signal power, that is, where μ t ≥0 is the ratio of the transmission distortion noise power to the transmission signal power; the channels between BS and RIS, BS and legal users, and RIS and legal users are respectively express; Then the downlink received signals at the legitimate user and the eavesdropper can be expressed as and U =and U +m U (2) in n U and n E Both The additive Gaussian white noise of the distribution, the noise variance is and m U is the independent zero-mean Gaussian distortion noise at the legitimate user, following The distribution of μ r ≥0 is the ratio of distorted noise power to undistorted received signal power; They are the channel between BS and eavesdropper and the channel between IRS and eavesdropper respectively; Assuming that the BS knows all the CSIs in the entire network, the confidentiality rate of the system can be expressed as R=[R U -R E ] + , (4) in R U is the communication rate of legitimate users, expressed as In the formula is the equivalent reflection coefficient vector, is the total power of the noise, G U is an equivalent channel, expressed as R E is the communication rate of the eavesdropper, expressed as In the formula is the total power of the noise, G E is an equivalent channel, expressed as Ideally, when If perfect CSI is available on all channel coherent blocks, the communication rate can be written as (f m ,e m ) is the optimal beam pair selected from the channel knowledge map, that is: and ε are the beam sets in CKM; Beam selection via CKM When the BS and RIS are located at fixed positions, the equivalent cascade channel G U Mainly with the terminal position (denoted as Q U ) and the actual propagation environment (denoted as E U ) changes; therefore, G U It can be expressed as G U =g1(Q U ,E U ) (10) where g1 is an unknown function that uses CKM to map any possible UE location and its channel knowledge specific to that location; The optimal beam pair can be expressed as an abstract function of the terminal location and propagation environment (f m ,e m )=g2(Q U ,E U ) (11) For a given communication environment, the above formula can be further simplified as (f m ,e m )≈g3(Q U )(12) g3(,.,) is an abstract function that relates the beam pairs in a given environment to the UE location.

5. The RIS-assisted secure wireless communication environment-aware beam selection method based on channel knowledge map according to claim 4 is characterized in that: In step S3, the alternating optimization AO algorithm is used to maximize the confidentiality rate R by jointly optimizing the beamforming vector at the BS and the passive beamforming at the RIS. In this work, this problem is formulated as Where (13b) is the transmit power constraint, the maximum transmit power P max At BS, (13c) imposes unit modulus on each term in e, setting The optimal value of problem (13) is always non-negative; if R U -R E <0, then set Make the value of the objective function 0; Introduce auxiliary variables p = [p1, p2, p3, p4] T , by R U ≥p1-p2 and R E ≤p3-p4, problem (13) can be restated as log2(Φ U (f,e))≤p2,(14e) log(Φ E (f,e))≥p4.(14g) It is easy to prove that problem (14) is equivalent to problem (13).

6. The RIS-assisted secure wireless communication environment-aware beam selection method based on channel knowledge map according to claim 5 is characterized in that: Step S3.1, fix e and optimize f; When e is fixed, we further introduce an auxiliary variable k f =[k f1 ,k f2 ,k f3 ,k f4 ] T , so that the non-convex constraints (15), (16), (17) and (18) are equivalent to From (15), (16), (17) and (18), we can see that constraints (15a), (16b), (17b) and (18a) are convex, while (15b), (16a), (17a) and (18b) are non-convex; according to y H diag(xx H )y=x H diag(yy H )x, we can derive the following equivalent constraints: In the formula A2=0. and f n is the solution obtained at the nth iteration; Finally, the subproblem of solving f is expressed as st(13b),(15a),(16b),(17b),(18a),(19). (20b) Problem (20) is a SOCP and is solved using the CVX tool.

7. The RIS-assisted secure wireless communication environment-aware beam selection method based on channel knowledge map according to claim 6 is characterized in that: Step S3.2, fix f and optimize e: To solve the non-convex unit modulus constraint The semidefinite relaxation (SDR) technique is used to update e; by defining a new variable E = ee H , constrain Replace with {E≥0,rank(E)=1,diag(E)=I M+1 }; After fixing f and introducing a new auxiliary variable k e =[k e1 ,k e2 ,k e3 ,k e4 ] T In the case of , the subproblem of solving e can be expressed as s.t.E≥0,rank(E)=1,diag(E)=I M+1 , (21b) log2(k e1 )≥p1, (21c) The SDR method is applied to further relax problem (21) into a convex SDP, and then solved using the CVX tool; In order to ensure that the target value sequence generated by each iteration does not decrease, the following update method is used When e n+1 =e n When R(f n+2 ,e n+1 )>R(f n+1 ,e n+1 )=R(f n+1 ,e n ), the iterative process ends until convergence.

8. The RIS-assisted secure wireless communication environment-aware beamforming based on a channel knowledge map according to any one of claims 4 to 7 choose The method is characterized in that: In step S4, the complexity and number of iterations of the joint optimization algorithm of CKM and AO are analyzed, and the optimal beam pair (f m ,e m ), solve the problem of maximum confidentiality rate R Thus, the best beam pair is optimized through the AO algorithm and the maximum confidentiality rate R is calculated; The target value sequence {R(F n ,e n )} is non-decreasing, where In addition, due to the bounded transmit power constraint, {R(F n ,e n )} has a finite upper bound, which guarantees the convergence of the AO algorithm; the complexity of solving problem (23) is O(N 3 )+O(M 3.5 ).

9. A RIS-assisted secure wireless communication system, characterized in that: The RIS-assisted secure wireless communication system uses the RIS-assisted secure wireless communication environment-aware beam selection method based on the channel knowledge map described in claim 1, based on CKM, using historical data and environment-aware technology to extract and obtain environmental features including the location and type of obstacles and reflective surfaces, and then maps these features to wireless propagation features in the environment; Build and update channel models.

Images