Rate-splitting-assisted near-field flux-sensing integrated hybrid beam forming method

Abstract

The invention discloses a rate splitting assisted near-field flux-sensing integrated hybrid beam forming method, and relates to the technical field of wireless communication and novel multiple access. The designed hybrid beam forming method comprises the following four steps: S1, constructing a near-field communication and sensing integrated network with a hybrid beam forming architecture, and establishing a near-field communication and near-field sensing channel model; s2, the base station sends required data to a communication user by using a rate splitting multiple access technology; s3, the base station collects echo signals reflected by the sensing target to carry out target detection; s4, establishing an optimization problem aiming at meeting the target sensing requirement and maximizing the communication rate, and designing a double-layer rotation optimization algorithm by using penalty dual decomposition, weight minimum mean square error and quadratic transformation methods to jointly process variables such as a receiving filter, digital beam forming, analog beam forming and common rate allocation; according to the invention, the communication and sensing rate of the communication and sensing integrated network can be effectively improved.

Description

Technical Field

[0001] The present invention belongs to the field of data communication in data centers, and in particular relates to a hybrid beamforming method for near-field synaesthesia integration assisted by rate splitting. Background Art

[0002] The next generation of wireless networks needs to support emerging applications such as autonomous driving, virtual reality, and smart healthcare. To support these applications, the network should have ultra-high data communication rates and high-precision target perception capabilities. Limited spectrum resources are a realistic pressure for building high-performance wireless networks. Integrated Sensing and Communications (ISAC) networks have become an important measure to actively cope with the scarcity of spectrum resources. ISAC realizes communication and perception functions simultaneously by sharing hardware platforms and radio resources, thereby effectively reducing hardware costs and greatly improving spectrum resource utilization. However, future ISACs need to deploy ultra-large-scale antenna arrays and use ultra-high frequencies to enhance communication capacity and perception accuracy. These advanced technologies fundamentally change the propagation characteristics of electromagnetic waves. Specifically, the electromagnetic radiation area at the transmission antenna end can be divided into far-field and near-field areas, which are modeled by plane waves and spherical waves for channels, respectively. Ultra-large-scale antenna arrays and ultra-high frequencies expand the coverage of the near-field area from the original several meters to tens of meters or even hundreds of meters, resulting in future communication channels dominated by spherical waves. Compared with plane waves, spherical waves introduce an additional distance dimension, so that the energy of the beam can be concentrated to a specific spatial position, effectively reducing interference. This feature provides a new idea for the design of ISAC, but the existing ISAC design is mainly based on far-field channels, which cannot match the actual wireless propagation environment. Therefore, it is urgent to design a near-field ISAC that can adapt to the spherical propagation characteristics. This design faces two technical challenges, which are summarized as follows:

[0003] 1) Co-channel interference is difficult to manage accurately: In ISAC, the coexistence of communication and perception functions leads to a complex wireless environment, so accurate co-channel interference management is extremely important. Existing ISACs mainly use Space Division Multiple Access (SDMA) and Non-orthogonal Multiple Access (NOMA) technologies to manage interference. SDMA treats all non-target signals as interference and directly decodes the target signal. When the network serves multiple users at the same time, excessive interference will cause the transmission rate to saturate. NOMA uses serial interference cancellation technology to fully decode all strong interference signals to enhance transmission performance. However, this technology requires strict decoding order, complex receiver design, and large channel gain differences. Therefore, it is urgent to design a new transmission scheme to accurately manage interference to enhance network performance.

[0004] 2) The deployment cost of the RF link is high: Near-field ISAC usually occurs in ultra-large-scale antenna array scenarios, and full digital beamforming requires each transmission antenna to be equipped with a dedicated RF link, which is difficult to achieve in near-field ISAC. Therefore, it is urgent to reduce the demand for RF links in ISAC.

[0005] The above challenges can be solved by rate splitting multiple access technology and hybrid beamforming technology. Rate splitting multiple access technology uses information splitting and serial interference cancellation technology to decode part of the interference signal while treating part of the interference as noise. By adjusting the percentage of information decoding, rate splitting multiple access technology can be automatically converted to SDMA and NOMA technology, that is, rate splitting multiple access technology includes two interference management methods, SDMA and NOMA. Hybrid beamforming connects each RF link to the antenna array, and no longer equips each antenna with a dedicated RF link, which can greatly reduce the number of RF links. Therefore, combining rate splitting and hybrid beamforming technology can effectively reduce the hardware cost of near-field ISAC and effectively improve performance, but there is no relevant research at present. In view of this, the present invention proposes a transmission scheme for a near-field ISAC network assisted by rate splitting, and designs a hybrid beamforming method using penalty dual decomposition, weighted minimum mean square error and quadratic transformation.

[0006] 1. Current status of research at home and abroad

[0007] In 2024, Zhengchuan Chen et al. published "Joint Rate Splitting and Beamforming Design for RSMA-RIS-Assisted ISAC System" in IEEE Wireless Communications Letters, using rate splitting multiple access technology to accurately manage co-channel interference of far-field ISAC to enhance target perception rate. However, this study used plane wave channels and all-digital beamforming, which cannot be applied to spherical wave channels and hybrid beamforming scenarios of near-field ISAC.

[0008] In 2024, Diluka Galappaththige et al. published "Near-Field ISAC: Beamforming for Multi-Target Detection" in IEEE Wireless Communications Letters, studying the all-digital beamforming method of near-field ISAC to maximize the signal-to-interference-noise ratio of target perception. However, this study has high requirements on the number of RF links and the SDMA used will significantly reduce the transmission performance.

[0009] In “Near-Field Integrated Sensing and Communications” published by Zhaolin Wang et al. in IEEE Communications Letters in 2023 and in “Near-Field Integrated Sensing, Positioning, and Communication: A Downlink and Uplink Framework” published by Haochen Li et al. in 2024 in IEEE Journal on Selected Areas in Communications, the hybrid beamforming design method of near-field ISAC was studied to reduce target perception errors. However, these two studies used SDMA to manage co-channel interference and could not ensure high-speed data transmission.

[0010] The above four types of ISACs have all studied rate splitting assisted ISAC and near-field ISAC separately. Currently, there is no research on near-field ISAC for rate splitting multiple access technology. Summary of the invention

[0011] The technical problem to be solved by the present invention is to provide a rate-splitting-assisted near-field synaesthesia-integrated hybrid beamforming method in view of the shortcomings of the background technology, which can effectively improve the communication and perception rates of the synaesthesia-integrated network.

[0012] The present invention adopts the following technical solutions to solve the above technical problems:

[0013] A rate-splitting-assisted near-field synaesthesia-integrated hybrid beamforming method, characterized in that it specifically comprises the following steps;

[0014] Step S1, constructing a near-field synaesthesia integrated network with a hybrid beamforming architecture, and establishing a channel model for near-field communication and near-field perception;

[0015] Step S2: The base station uses rate splitting multiple access technology to send required data to the communication user;

[0016] Step S3: The base station collects and senses the echo signal reflected by the target to perform target detection;

[0017] Step S4, establish an optimization problem with the goal of meeting the target perception requirements and maximizing the communication rate, and use the penalty dual decomposition, weighted minimum mean square error and quadratic transformation method to design a two-layer rotation optimization algorithm to jointly process the receiving filter, digital beamforming, analog beamforming and common rate allocation variables.

[0018] As a further preferred embodiment of a hybrid beamforming method for near-field synaesthesia integration assisted by rate splitting of the present invention, the near-field synaesthesia integration network comprises a base station, K>1 single-antenna communication users and M>1 sensing targets; the sets of communication users and sensing targets are respectively denoted as and The base station is equipped with N t Transmitting antennas and N r A uniform linear array of N receiving antennas f RF links (N f <N t ), where the distance between the connected antennas is denoted as d, then the boundary between the near-field and far-field regions is given by the Rayleigh distance Determine, among which, D in Rayleigh distance i =(N i -1)d and λ are the antenna aperture and signal wavelength respectively;

[0019] The coordinates of the midpoint of the transmission antenna array and the nth transmission antenna are recorded as (0, 0) and s n =(0, nd), where For a user or target with a distance r and an angle θ relative to the center of the transmission antenna array, its coordinates are r = (r cosθ, r sinθ), then the distance between the user or target and the nth transmission antenna is

[0020]

[0021] The corresponding near-field channel It can be modeled as in is the path loss; specifically, where f and c are the carrier frequency and the speed of light respectively; d is approximated using a second-order Taylor expansion n (r, θ) can obtain the spherical wave characteristics of the near-field channel, that is, d n (r, θ)≈r-δ n (r, θ), where δ n (r, θ) = ndsinθ - (nd) 2 cos 2 θ / 2r; Substitute it into The near-field channel between the user or target and the nth transmission antenna can be equivalent to: in Combining all the transmitting antennas, the total near-field channel of the user or target is

[0022]

[0023] Where a(r, θ) is the near-field array response vector of the transmitting antenna;

[0024] A multipath channel model is considered, which includes a line-of-sight path and Q non-line-of-sight paths introduced by Q scatterers; the distance and angle of the kth user are denoted by r and k and θ k The distance and angle of the qth scatterer are denoted as r k，q and θ k，q , then the channel Can be characterized as

[0025]

[0026] in and are the complex channel gains of the line-of-sight path and the qth non-line-of-sight path, respectively, is the distance between the kth user and the qth scatterer;

[0027] Target perception relies on the echo signal collected by the base station, that is, the base station needs to send a detection signal and collect the echo signal; the array response vector of the near-field channel established using equation (2), the perception channel matrix of the mth perception target It can be modeled as

[0028]

[0029] in, and are the channel gain and the coordinates of the mth perceived target; and are the near-field array response vectors of the receiving and transmitting antennas, respectively.

[0030] As a further preferred solution of the hybrid beamforming method of near-field synaesthesia integration assisted by rate splitting of the present invention, according to the coding and decoding principle of rate splitting multiple access technology, the information W of the kth user k Split into a common part W c，k and a private section W p，k ; Then all the common parts {W c，1 , ..., W c，K} is encoded as a public information stream x0, and the private part {W p，1 , ..., W p，K} is encoded as independent private information streams {s1, ..., s K}; The signal flow vector transmitted at the lth time index can be expressed as s(l)=[s0(l), s1(l), ..., s K (l)] T ,in is a discrete time index, L is the total number of time indices under a coherent processing interval; the signal stream s(l) passes through the hybrid beamforming matrix Linear precoding is performed, where is the analog beamforming matrix, is the digital beamforming matrix; in W, and are the digital beamforming vectors of the public information flow and the kth private information flow respectively; the signal transmitted by the base station at the lth time index can be expressed as

[0031]

[0032] The signal received by the kth user is

[0033]

[0034] in is additive Gaussian white noise; Combining equation (5) and equation (6), the power received by the kth user can be calculated as

[0035]

[0036] To reconstruct the required information, the kth user treats the private information stream as interference to decode the public information stream, so the signal-to-interference-to-noise ratio is At the same time, to ensure that all users can successfully decode the public information flow, the public rate should meet Where R c，k =log(1+γ c，k ); Since all users share the common information rate, Among them C c，k is the public information stream allocated to the kth user; after removing the interference of the public information stream using serial interference cancellation, the kth user starts decoding the private information stream, and its signal-to-interference-noise ratio and transmission rate are and R p，k =log(1+γ p，k ); Since the transmission rate of the kth user includes the public rate and the private rate, R k =C c，k +R p，k ;

[0037] The sensing target reflects the signal x(l) to the base station, so the echo signal received by the base station at the lth time index is

[0038]

[0039] where α m is the energy reflection coefficient of the mth sensing target, is additive white Gaussian noise; after receiving the echo signal, the base station uses the receiving filter Obtain the expected signal of the mth perceived target, and we can get

[0040]

[0041] From equation (9), the signal-to-interference-noise ratio of the mth perceived target can be obtained as

[0042]

[0043] in is the transmission covariance matrix; therefore, the rate of the mth sensor target is

[0044] Under the condition of satisfying the perceived rate, the communication rate of the worst user is maximized by optimizing the variables such as receiving filter, digital beamforming, analog beamforming and public rate allocation. Therefore, the optimization problem can be modeled as:

[0045]

[0046] Where U=[u1,...,u M ] and c=[C c，1 , ..., C c，K ]; P th and R th are the maximum transmission power threshold and target perceived rate requirement threshold of the base station respectively; (11b) and (11c) are the transmission power and perceived rate requirement limits respectively; (11d) and (11e) are the common rate allocation limits of the rate splitting technology; (11f) and (11g) are the module value constraints of the analog beamforming and receiving filters.

[0047] As a further preferred solution of the hybrid beamforming method of the present invention for near-field synaesthesia integration assisted by rate splitting, a hybrid beamforming optimization algorithm is designed to solve problem (11) by using penalized dual decomposition, weighted minimum mean square error and quadratic transformation method. The specific solution steps are as follows:

[0048] Step 1: Use penalty dual decomposition to reconstruct the problem (11). The specific process is as follows: Introduce the auxiliary matrix where p k =Fw k , refactoring equation (7) into

[0049]

[0050] Similarly, the covariance matrix R should be updated to R = PP H ; To eliminate non-smoothness, a non-negative auxiliary variable R is introduced s Refactoring problem (11) into

[0051]

[0052] in includes all optimization variables; in the optimization problem (13), the equation restriction (13a) makes the problem difficult to solve directly; the penalty dual decomposition method is used to deal with the restriction (13a); specifically, the penalty dual decomposition transfers the equation restriction to the objective function by introducing the Lagrangian dual matrix and the penalty factor, transforming it into a two-layer iterative problem, in which the outer layer is used to update the introduced dual matrix and penalty factor, and the inner layer is used to solve the augmented Lagrangian problem; by introducing the Lagrangian dual matrix D and the penalty parameter ρ, the augmented Lagrangian problem can be modeled as

[0053]

[0054] Problem (14) is separable with respect to all optimization variables, so the following steps 2 to 4 will use a rotation optimization method to solve problem (14);

[0055] Step 2: Under the condition of fixing the auxiliary matrix P, optimize the receiving filter U. The specific process is as follows: Since the receiving filter u of the mth perception target is optimized m It will not affect the performance of other perception targets, so problem (14) can be divided into M independent sub-problems, as follows

[0056]

[0057] in Problem (15) is a generalized Rayleigh quotient problem, and its optimal solution is

[0058]

[0059] where v max (A) represents the eigenvector corresponding to the largest eigenvalue of matrix A;

[0060] Step 3, under the condition of fixed hybrid beamforming FW and receiving filter U, optimize the auxiliary matrix P and the common rate allocation variable c. The specific process is as follows: Problem (14) can be simplified to

[0061]

[0062] Problem (17) contains fractional signal-to-noise ratio and coupled auxiliary variables, which are difficult to solve directly. Therefore, the weighted minimum mean square error and quadratic transformation methods are used to reconstruct the communication rate and perception rate respectively. Specifically, the kth user uses the equalizer ω c，k Estimate the public information flow s0, that is After removing the public information flow, use ωp，K Estimate the private information flow, i.e. It can be obtained that the mean square error between public information flow and private information flow can be expressed as

[0063]

[0064] By solving and The minimum mean square error equalizer can be obtained as

[0065]

[0066] Substituting equation (19) into equation (18), we can get the minimum mean square error:

[0067]

[0068] Using equation (20), the signal-to-interference-noise ratio of the public information flow and the private information flow can be reconstructed as and That is, the transmission rate is and Introducing the weighted minimum mean square error function β c，k =η c，k -log(η c，k ) and β p,k =η p，k δ p，k -log(η p，k ), where η c，k and η p，k is the weight parameter; by taking the equalizer and weight as optimization variables, the relationship between the communication rate and the weighted minimum mean square error can be expressed as

[0069]

[0070] Where τ=1 / ln2+log(ln2); by solving and The optimal equalizer can be obtained and Similarly, the optimal weights can be expressed as

[0071]

[0072] Although the weighted minimum mean square error can be used to reconstruct the communication rate, it cannot be used to reconstruct the perception rate. To address this challenge, the quadratic transformation method is used to reconstruct the perception rate by introducing auxiliary variables. Specifically, by introducing the constructed proxy function f(x, P) = 2Re(x H s(P)-x H I(P)x), where and I(P)>0, we can get And the optimal solution on the right side of the equation is

[0073]

[0074] By replacing the variables in the proxy function with the useful signal and interference signal in the perception rate, we can get

[0075]

[0076] At this point, problem (17) can be reconstructed as

[0077]

[0078] in and They include inherent variables and introduced auxiliary variables respectively; although and The coupling between them makes problem (25) difficult to solve directly, but it can be iteratively optimized by using a rotation optimization method. and Fixed Solving equations (19), (22) and (23) Fixed variables Solved by Convex Optimization Toolbox

[0079] Step 4: Optimize the hybrid beamforming matrices F and W. The specific process is as follows: Variables F and W only appear in the last two terms of the objective function, so problem (14) can be simplified to

[0080]

[0081] Problem (26) involves the coupling between F and W, which is difficult to solve directly. Therefore, a rotation optimization method is used to optimize one variable while fixing the other variable. When the variable F is fixed, the objective function of problem (26) is transformed into This problem can be solved by the first-order optimal condition to find the optimal W, that is,

[0082] W * =(F H F) -1 F H (P+ρD) (57)

[0083] When the variable W is fixed, problem (26) can be transformed into

[0084]

[0085] Where Y = WW H and Z = (P + ρD)W H; Since the module constraints in constraint (28) are separable, the optimization problem (28) can be equivalent to

[0086]

[0087] The optimal solution to problem (29) is

[0088]

[0089] in and is the optimal solution obtained in the previous iteration;

[0090] Step 5: Based on the above steps 1 to 4, a two-layer iterative optimization algorithm can be designed to solve the optimization problem (14). The pseudo code of the specific solution process is as follows:

[0091] Pseudocode of the algorithm for solving problem (14):

[0092] 1. Initialize F (0) , W (0) , D (0) , (0) and ψ (0) , set the outer iteration index n = 1;

[0093] 2. While non-convergent execution

[0094] 3. Set F (1) =F (n-1) and W (1) =W (n-1 ) and inner iteration index i=1;

[0095] 4. While non-convergent execution

[0096] 5. Update the receive filter U according to equation (16) (i) ;

[0097] 6. Update according to equations (19), (22) and (23)

[0098] 7. Solve problem (25) through CVX and obtain P (i) and c (i) ;

[0099] 8. Update the digital beamforming matrix W according to equation (27): (i+1) ;

[0100] 9. Update the analog beamforming matrix F according to equation (30) (i+1) ;

[0101] 10. Update iteration index i=i+1;

[0102] 11. end while

[0103] 12. Get the best P (n) 、F (n) and W (n) ;

[0104] 13. if||P (n) -F (n) W (n) || ∞ ≤ψ (n-1) implement

[0105] 14. Update ρ (n) =ρ (n-1) ;

[0106] 15. else

[0107] 16. Update ρ (n) =0.9ρ (n-1) , D (n) =D (n-1) ;

[0108] 17. end if

[0109] 18. Update ψ (n) =0.9||P (n) -F (n) W (n) || ∞ , n = n + 1;

[0110] 19. end while

[0111] 20. Output the optimal communication rate.

[0112] Compared with the prior art, the present invention adopts the above technical solution and has the following technical effects:

[0113] 1. The present invention discloses a hybrid beamforming method for near-field synaesthesia integration assisted by rate splitting, which comprises four steps: S1, constructing a near-field synaesthesia integration network with a hybrid beamforming architecture, and establishing a channel model for near-field communication and near-field perception; S2, a base station uses rate splitting multiple access technology to send required data to communication users; S3, a base station collects echo signals reflected by perceived targets for target detection; S4, an optimization problem is established with the goal of meeting target perception requirements and maximizing communication rate, and a double-layer rotation optimization algorithm is designed by using penalty dual decomposition, weighted minimum mean square error and quadratic transformation methods to jointly process variables such as receiving filters, digital beamforming, analog beamforming and public rate allocation; the present invention can effectively improve the communication and perception rates of the synaesthesia integration network. BRIEF DESCRIPTION OF THE DRAWINGS

[0114] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without paying any creative work.

[0115] Figure 1 is a network diagram of the rate splitting assisted near-field ISAC proposed in the present invention;

[0116] Figure 2 The relationship between the communication rate and the base station transmission power threshold was simulated;

[0117] Figure 3 The relationship between the communication rate and the number of RF links is shown;

[0118] Figure 4 The relationship between the communication rate and the number of users was simulated;

[0119] Figure 5 The effect of the number of sensing targets on the communication rate was simulated. DETAILED DESCRIPTION

[0120] The technical solution of the present invention is further described in detail below in conjunction with the accompanying drawings:

[0121] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. The present invention is described in detail below based on the drawings and preferred embodiments, and the purpose and effect of the present invention will become clearer. It should be understood that the specific embodiments described here are only used to explain the present invention and are not used to limit the present invention.

[0122] Figure 1 The network diagram of the proposed rate splitting assisted near-field ISAC is shown. The designed near-field ISAC consists of a base station, K>1 single-antenna communication users and M>1 sensing targets. The sets of communication users and sensing targets are denoted as and The base station is equipped with N t Transmitting antennas and N r A uniform linear array of N receiving antennas f RF links (N f <N t), where the distance between the connected antennas is d, then the boundary between the near-field and far-field regions is given by the Rayleigh distance Determine, among which D in Rayleigh distance i =(N i -1)d and λ are the antenna aperture and signal wavelength respectively.

[0123] The coordinates of the midpoint of the transmission antenna array and the nth transmission antenna are recorded as (0, 0) and s n =(0, nd), where For a user or target with a distance r and an angle θ relative to the center of the transmission antenna array, its coordinates are r = (r cosθ, r sinθ), then the distance between the user or target and the nth transmission antenna is

[0124]

[0125] The corresponding near-field channel It can be modeled as in is the path loss. Specifically, where f and c are the carrier frequency and the speed of light, respectively. Use the second-order Taylor expansion to approximate d n (r, θ) can obtain the spherical wave characteristics of the near-field channel, that is, d n (r, λ, θ)≈r-δ n (r, θ), where δ n (r, θ) = ndsinθ - (nd) 2 cos 2 θ / 2r. Substituting this into The near-field channel between the user or target and the nth transmission antenna can be equivalent to: in Combining all the transmitting antennas, the total near-field channel of the user or target is

[0126]

[0127] where a(r,θ) is the near-field array response vector of the transmitting antenna.

[0128] The present invention considers a multipath channel model, which includes a line-of-sight path and Q non-line-of-sight paths introduced by Q scatterers. The distance and angle of the kth user are denoted as r k and θ k The distance and angle of the qth scatterer are denoted as r k，q and θ k，q , then the channel Can be characterized as

[0129]

[0130] in and are the complex channel gains of the line-of-sight path and the qth non-line-of-sight path, respectively, is the distance between the kth user and the qth scatterer.

[0131] Target perception relies on the echo signals collected by the base station, so the base station needs to send a detection signal and collect the echo signals. Using the array response vector of the near-field channel established by equation (2), the perception channel matrix of the mth perception target is It can be modeled as

[0132]

[0133] in and are the channel gain and the coordinates of the mth perceived target; and are the near-field array response vectors of the receiving and transmitting antennas, respectively.

[0134] In order to accurately manage co-channel interference, the present invention proposes a data transmission scheme based on rate splitting. According to the coding and decoding principle of rate splitting multiple access technology, the information W of the kth user k Split into a common part W c，k and a private section W p，k ; Then all the common parts {W c，1 , …, W c，K} is encoded as a public information stream x0, the private part <W p，1 , …, W p，K} is encoded as independent private information streams {s1, ..., s K Therefore, the signal flow vector transmitted at the lth time index can be expressed as s(l)=[s0(l), s1(l), ..., s K (l)] T ,in is a discrete time index, and L is the total number of time indices in a coherent processing interval. Subsequently, the signal stream s(l) passes through the hybrid beamforming matrix Linear precoding is performed, where is the analog beamforming matrix, is the digital beamforming matrix. In W, and are the digital beamforming vectors of the public information stream and the kth private information stream, respectively. Therefore, the signal transmitted by the base station at the lth time index can be expressed as

[0135]

[0136] The signal received by the kth user is

[0137]

[0138] in is additive Gaussian white noise. Combining equation (5) and equation (6), the power received by the kth user can be calculated as

[0139]

[0140] To reconstruct the required information, the kth user treats the private information stream as interference to decode the public information stream, so the signal-to-interference-to-noise ratio is At the same time, to ensure that all users can successfully decode the public information flow, the public rate should meet Where R c，k =log(1+γ c，k ). In addition, since all users share the common information rate, Among them C c，k is the public information stream assigned to the kth user. After removing the interference of the public information stream using serial interference cancellation, the kth user starts decoding the private information stream, and its signal-to-interference-noise ratio and transmission rate are and Since the transmission rate of the kth user includes the public rate and the private rate, R k =C c，k +R p，k .

[0141] The sensing target reflects the signal x(l) to the base station, so the echo signal received by the base station at the lth time index is

[0142]

[0143] where α m is the energy reflection coefficient of the mth sensing target, is additive white Gaussian noise. After receiving the echo signal, the base station uses the receiving filter Obtain the expected signal of the mth perceived target, and we can get

[0144]

[0145] From equation (9), the signal-to-interference-noise ratio of the mth perceived target can be obtained as

[0146]

[0147] in is the transmission covariance matrix. Therefore, the rate of the mth sensing target is

[0148] The present invention aims to maximize the communication rate of the worst user by optimizing variables such as receiving filter, digital beamforming, analog beamforming and public rate allocation under the condition of satisfying the perceived rate. Therefore, the optimization problem can be modeled as:

[0149]

[0150] Where U=[u1,...,u M ] and c=[C c，1 , ..., C c，K ]; P th and R th are the maximum transmission power threshold and target perceived rate requirement threshold of the base station, respectively. (11b) and (11c) are the transmission power and perceived rate requirement limits, respectively; (11d) and (11e) are the common rate allocation limits of the rate splitting technique; (11f) and (11g) are the modulus constraints of the analog beamforming and receiving filters.

[0151] Hybrid beamforming algorithm design:

[0152] The optimization problem (11) is non-convex and non-smooth, and is difficult to solve directly. To solve problem (11), the present invention designs a hybrid beamforming optimization algorithm using penalized dual decomposition, weighted minimum mean square error, and quadratic transformation methods. The specific solution steps are as follows:

[0153] Step 1: Use penalty dual decomposition to reconstruct the problem (11). The specific process is as follows: Introduce the auxiliary matrix where p k =Fw k , refactoring equation (7) into

[0154]

[0155] Similarly, the covariance matrix R should be updated to R = PP H To eliminate non-smoothness, a non-negative auxiliary variable R is introduced s Refactoring problem (11) into

[0156]

[0157] in Contains all optimization variables. In the optimization problem (13), the equation restriction (13a) makes the problem difficult to solve directly. To this end, the penalty dual decomposition method is used to deal with the restriction (13a). Specifically, the penalty dual decomposition transfers the equation restriction to the objective function by introducing the Lagrangian dual matrix and the penalty factor, transforming it into a two-layer iterative problem, where the outer layer is used to update the introduced dual matrix and penalty factor, and the inner layer is used to solve the augmented Lagrangian problem. By introducing the Lagrangian dual matrix D and the penalty parameter ρ, the augmented Lagrangian problem can be modeled as

[0158]

[0159] Problem (14) is separable with respect to all optimization variables, so the following steps 2 to 4 will use a rotation optimization method to solve problem (14).

[0160] Step 2: Under the condition of fixing the auxiliary matrix P, optimize the receiving filter U. The specific process is as follows: Since the receiving filter u of the mth perception target is optimized m It will not affect the performance of other perception targets, so problem (14) can be divided into M independent sub-problems, as follows

[0161]

[0162] in Problem (15) is a generalized Rayleigh quotient problem, and its optimal solution is

[0163]

[0164] where v max (A) represents the eigenvector corresponding to the largest eigenvalue of matrix A.

[0165] Step 3, under the condition of fixed hybrid beamforming FW and receiving filter U, optimize the auxiliary matrix P and the common rate allocation variable c. The specific process is as follows: Problem (14) can be simplified to

[0166]

[0167] Problem (17) contains fractional signal-to-noise ratio and coupled auxiliary variables, which are difficult to solve directly. Therefore, the weighted minimum mean square error and quadratic transformation methods are used to reconstruct the communication rate and perception rate, respectively. Specifically, the kth user uses the equalizer ω c，k Estimate the public information flow s0, that is After removing the public information flow, use ω p，k Estimate the private information flow, i.e. It can be obtained that the mean square error between public information flow and private information flow can be expressed as

[0168]

[0169] By solving and The minimum mean square error equalizer can be obtained as

[0170]

[0171] Substituting equation (19) into equation (18), we can get the minimum mean square error:

[0172]

[0173] Using equation (20), the signal-to-interference-noise ratio of the public information flow and the private information flow can be reconstructed as and That is, the transmission rate is and Then, the weighted minimum mean square error function β is introduced c，k =η c，k δ c，k -log(η c，k ) and β p，k =η p，k δ p，k -log(η p，k ), where η c，k and η p，k is the weight parameter. By taking the equalizer and weight as optimization variables, the relationship between the communication rate and the weighted minimum mean square error can be expressed as

[0174]

[0175] Where τ=1 / ln2+log(ln2). By solving and The optimal equalizer can be obtained and Similarly, the optimal weights can be expressed as

[0176]

[0177] Although the weighted minimum mean square error can be used to reconstruct the communication rate, it cannot be used to reconstruct the perceived rate. To address this challenge, the quadratic transformation method is used to reconstruct the perceived rate by introducing auxiliary variables. Specifically, by introducing the constructed proxy function f(x, P) = 2Re(x H s(P)-x H I(P)x), where and I(P)>0, we can get And the optimal solution on the right side of the equation is

[0178]

[0179] By replacing the variables in the proxy function with the useful signal and interference signal in the perception rate, we can get

[0180]

[0181] At this point, problem (17) can be reconstructed as

[0182]

[0183] in and They include inherent variables and introduced auxiliary variables. and The coupling between them makes problem (25) difficult to solve directly, but it can be iteratively optimized by using a rotation optimization method. and Fixed Solving equations (19), (22) and (23) Fixed variables Solved by Convex Optimization Toolbox

[0184] Step 4: Optimize the hybrid beamforming matrices F and W. The specific process is as follows: Variables F and W only appear in the last two terms of the objective function, so problem (14) can be simplified to

[0185]

[0186] Problem (26) involves the coupling between F and W, which is difficult to solve directly. Therefore, a rotation optimization method is used to optimize one variable while fixing the other variable. When the variable F is fixed, the objective function of problem (26) will be transformed into This problem can be solved by the first-order optimal condition to find the optimal W, that is,

[0187] W * =(F H F) -1 F H (P+ρD) (87)

[0188] When the variable W is fixed, problem (26) can be transformed into

[0189]

[0190] Where Y = WW H and Z = (P + ρD)W H Since the module constraints in constraint (28) are separable, the optimization problem (28) can be equivalent to

[0191]

[0192] The optimal solution to problem (29) is

[0193]

[0194] in and is the optimal solution obtained in the previous iteration.

[0195] Step 5: Based on the above steps 1 to 4, a two-layer iterative optimization algorithm can be designed to solve the optimization problem (14). The pseudo code of the specific solution process is as follows:

[0196]

[0197] Simulation results: Simulation parameter settings: The base station of near-field ISAC is equipped with N t = 64 transmission antennas and N r = 64 receiving antennas, and the operating frequency is f c =30GHz, antenna aperture size is D t =D r = 0.5 m, so the near field area formed by the antenna aperture and the operating frequency is 50 m; the number of RF links and scatterers are N f =8 and Q = 2, and the scatterers are distributed within a range of 20 to 30 meters from the base station; K = 6 users and M = 4 sensing targets are randomly distributed in the near field area; the maximum transmission power threshold and background noise of the base station are P th =30dBm and σ 2 =-80dBm; the communication rate requirement of the sensing target is R th =6bps / Hz.

[0198] Figure 2 The relationship between the communication rate and the base station transmission power threshold is simulated. The simulation results show that the proposed scheme has three advantages over the near-field space division multiple access technology, the far-field rate split multiple access technology and the pure communication rate split technology: 1) The proposed transmission scheme is significantly better than the space division multiple access technology, and when P th When the transmission rate is ≥26dBm, the communication rate of spatial division multiple access technology tends to saturation, but the proposed scheme can avoid rate saturation; 2) The achievable communication rate of the proposed transmission scheme is significantly better than that of ISAC under the far-field channel model; 3) The proposed transmission scheme achieves a perceived rate of 6bps / Hz at the cost of reducing the communication rate by 0.05bps / Hz, which shows the advantage of the proposed transmission scheme.

[0199] Figure 3The relationship between the communication rate and the number of RF links is shown. The results show that as the number of RF links increases, the communication rate of all transmission schemes will increase. In addition, under the condition of the same RF links, the scheme proposed in this invention can keep up with the achievable rate of the pure communication network and outperform the space division multiple access technology and far-field ISAC.

[0200] Figure 4 The relationship between the communication rate and the number of users was simulated. The results show that as the number of users increases, the communication rate of all transmission schemes will decrease. The performance difference between the proposed transmission scheme and the space division multiple access technology gradually narrows. This is because each user only uses one layer of serial interference elimination to decode the common information stream, and the transmission rate of the common information stream depends on the user with the worst channel quality and is shared by all users. In addition, the transmission scheme proposed in the present invention has more obvious advantages over far-field ISAC. This is because the spherical wave introduces distance information, which can focus the beam energy at a specific spatial position, thereby reducing interference between users.

[0201] Figure 5 The effect of the number of sensing targets on the communication rate was simulated. The simulation results show that the performance of the far-field ISAC will seriously degrade when serving more than five sensing targets, resulting in a significant advantage in the scheme proposed in the present invention. For example, when M=7, the near-field ISAC achieves a performance gain of approximately 0.25bps / Hz compared to the far-field ISAC. This is because as the number of sensing targets increases, the probability of sensing targets with the same angle and distance increases. Spherical waves contain distance and angle information and can better cope with this situation. In addition, compared to the space division multiple access technology, the transmission scheme proposed in the present invention provides a performance gain of approximately 0.35bps / Hz.

[0202] Those skilled in the art can understand that the above are only preferred examples of the invention and are not intended to limit the invention. Although the invention is described in detail with reference to the above examples, those skilled in the art can still modify the technical solutions recorded in the above examples or replace some of the technical features with equivalents. Any modification, equivalent replacement, etc. made within the spirit and principle of the invention should be included in the protection scope of the invention. All technical features in this embodiment can be freely combined according to actual needs.

[0203] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A hybrid beamforming method for near-field synaesthesia integration assisted by rate splitting, characterized by: The specific steps include: Step S1, constructing a near-field synaesthesia integrated network with a hybrid beamforming architecture, and establishing a channel model for near-field communication and near-field perception; Step S2: The base station uses rate splitting multiple access technology to send required data to the communication user; Step S3: The base station collects and senses the echo signal reflected by the target to perform target detection; Step S4, establish an optimization problem with the goal of meeting the target perception requirements and maximizing the communication rate, and use the penalty dual decomposition, weighted minimum mean square error and quadratic transformation method to design a two-layer rotation optimization algorithm to jointly process the receiving filter, digital beamforming, analog beamforming and common rate allocation variables.

2. The hybrid beamforming method of near-field synaesthesia integrated with rate splitting assistance according to claim 1, characterized in that: The near-field synaesthesia integrated network includes a base station, K>1 communication users with single antennas and M>1 sensing targets; the sets of communication users and sensing targets are respectively denoted as and The base station is equipped with N t Transmitting antennas and N r A uniform linear array of N receiving antennas f RF links (N f <N t ), where the distance between the connected antennas is denoted as d, then the boundary between the near-field and far-field regions is given by the Rayleigh distance Determine, among which, D in Rayleigh distance i =(N i -1)d and λ are the antenna aperture and signal wavelength respectively; The coordinates of the midpoint of the transmission antenna array and the nth transmission antenna are recorded as (0, 0) and s n =(0, nd), where For a user or target with a distance r and an angle θ relative to the center of the transmission antenna array, its coordinates are r = (rcosθ, rsinθ). Then the distance between the user or target and the nth transmission antenna is The corresponding near-field channel It can be modeled as in is the path loss; specifically, where f and c are the carrier frequency and the speed of light respectively; d is approximated using a second-order Taylor expansion n (r, θ) can obtain the spherical wave characteristics of the near-field channel, that is, d n (r, θ)≈r-δ n (r, θ), where δ n (r, θ) = nd sin θ - (nd) 2 cos 2 θ / 2r; Substitute it into The near-field channel between the user or target and the nth transmission antenna can be equivalent to: in Combining all the transmitting antennas, the total near-field channel of the user or target is Where a(r, θ) is the near-field array response vector of the transmitting antenna; A multipath channel model is considered, which includes a line-of-sight path and Q non-line-of-sight paths introduced by Q scatterers; the distance and angle of the kth user are denoted by r and k and θ k The distance and angle of the qth scatterer are denoted as r k，q and θ k，q , then the channel Can be characterized as in and are the complex channel gains of the line-of-sight path and the qth non-line-of-sight path, respectively, is the distance between the kth user and the qth scatterer; Target perception relies on the echo signal collected by the base station, that is, the base station needs to send a detection signal and collect the echo signal; the array response vector of the near-field channel established using equation (2), the perception channel matrix of the mth perception target It can be modeled as in, and are the channel gain and the coordinates of the mth perceived target; and are the near-field array response vectors of the receiving and transmitting antennas, respectively.

3. The hybrid beamforming method of near-field synaesthesia integrated with rate splitting assistance according to claim 2, characterized in that: According to the coding and decoding principle of rate split multiple access technology, the information W of the kth user k Split into a common part W c，k and a private section W p，k ; Then all the common parts {W c，1 , ..., W c，K } is encoded as a public information stream x0, and the private part {W p，1 , ..., W p，K } is encoded as independent private information streams {s1, ..., s K }; The signal flow vector transmitted at the lth time index can be expressed as s(l)=[s0(l), s1(l), ..., s K (l)] T ,,in is a discrete time index, L is the total number of time indices under a coherent processing interval; the signal stream s(l) passes through the hybrid beamforming matrix Linear precoding is performed, where is the analog beamforming matrix, is the digital beamforming matrix; in W, and are the digital beamforming vectors of the public information flow and the kth private information flow respectively; the signal transmitted by the base station at the lth time index can be expressed as The signal received by the kth user is in is additive Gaussian white noise; Combining equation (5) and equation (6), the power received by the kth user can be calculated as To reconstruct the required information, the kth user treats the private information stream as interference to decode the public information stream, so the signal-to-interference-to-noise ratio is At the same time, to ensure that all users can successfully decode the public information flow, the public rate should meet Where R c，k =log(1+γ c，k ); Since all users share the common information rate, Among them C c，k is the public information stream allocated to the kth user; after removing the interference of the public information stream using serial interference cancellation, the kth user starts decoding the private information stream, and its signal-to-interference-noise ratio and transmission rate are and R p，k =log(1+γ p，k ); Since the transmission rate of the kth user includes the public rate and the private rate, R k =C c，k +R p，k ; The sensing target reflects the signal x(l) to the base station, so the echo signal received by the base station at the lth time index is where α m is the energy reflection coefficient of the mth sensing target, is additive white Gaussian noise; after receiving the echo signal, the base station uses the receiving filter Obtain the expected signal of the mth perceived target, and we can get From equation (9), the signal-to-interference-noise ratio of the mth perceived target can be obtained as in is the transmission covariance matrix; therefore, the rate of the mth sensor target is Under the condition of satisfying the perceived rate, the communication rate of the worst user is maximized by optimizing the variables such as receiving filter, digital beamforming, analog beamforming and public rate allocation. Therefore, the optimization problem can be modeled as: Where U=[u1,...,u M ] and c=[C c，1 , ..., C c，K ]; P th and R th are the maximum transmission power threshold and target perceived rate requirement threshold of the base station respectively; (11b) and (11c) are the transmission power and perceived rate requirement limits respectively; (11d) and (11e) are the common rate allocation limits of the rate splitting technology; (11f) and (11g) are the module value constraints of the analog beamforming and receiving filters.

4. The hybrid beamforming method of near-field synaesthesia integrated with rate splitting assistance according to claim 3 is characterized by: A hybrid beamforming optimization algorithm is designed to solve problem (11) by using penalized dual decomposition, weighted minimum mean square error and quadratic transformation method. The specific solution steps are as follows: Step 1: Use penalty dual decomposition to reconstruct the problem (11). The specific process is as follows: Introduce the auxiliary matrix where p k =Fw k , refactoring equation (7) into Similarly, the covariance matrix R should be updated to R = PP H ; To eliminate non-smoothness, a non-negative auxiliary variable R is introduced s Refactoring problem (11) into in includes all optimization variables; in the optimization problem (13), the equation restriction (13a) makes the problem difficult to solve directly; the penalty dual decomposition method is used to deal with the restriction (13a); specifically, the penalty dual decomposition transfers the equation restriction to the objective function by introducing the Lagrangian dual matrix and the penalty factor, transforming it into a two-layer iterative problem, in which the outer layer is used to update the introduced dual matrix and penalty factor, and the inner layer is used to solve the augmented Lagrangian problem; by introducing the Lagrangian dual matrix D and the penalty parameter ρ, the augmented Lagrangian problem can be modeled as Problem (14) is separable with respect to all optimization variables, so the following steps 2 to 4 will use a rotation optimization method to solve problem (14); Step 2: Under the condition of fixing the auxiliary matrix P, optimize the receiving filter U. The specific process is as follows: Since the receiving filter u of the mth perception target is optimized m It will not affect the performance of other perception targets, so problem (14) can be divided into M independent sub-problems, as follows in Problem (15) is a generalized Rayleigh quotient problem, and its optimal solution is where v max (A) represents the eigenvector corresponding to the largest eigenvalue of matrix A; Step 3, under the condition of fixed hybrid beamforming FW and receiving filter U, optimize the auxiliary matrix P and the common rate allocation variable c. The specific process is as follows: Problem (14) can be simplified to Problem (17) contains fractional signal-to-noise ratio and coupled auxiliary variables, which are difficult to solve directly. Therefore, the weighted minimum mean square error and quadratic transformation methods are used to reconstruct the communication rate and perception rate respectively. Specifically, the kth user uses the equalizer ω c，k Estimate the public information flow s0, that is After removing the public information flow, use ω p，k Estimate the private information flow, i.e. It can be obtained that the mean square error between public information flow and private information flow can be expressed as By solving and The minimum mean square error equalizer can be obtained as Substituting equation (19) into equation (18), we can get the minimum mean square error: Using equation (20), the signal-to-interference-noise ratio of the public information flow and the private information flow can be reconstructed as and That is, the transmission rate is and Introducing the weighted minimum mean square error function β c，k =η c，k δ c，k -log(η c，k ) and β p，k =η p，k δ p，k -log(η p，k ), where η c，k and η p，k is the weight parameter; by taking the equalizer and weight as optimization variables, the relationship between the communication rate and the weighted minimum mean square error can be expressed as Where τ=1 / ln2+log(ln2); by solving and The optimal equalizer can be obtained and Similarly, the optimal weights can be expressed as Although the weighted minimum mean square error can be used to reconstruct the communication rate, it cannot be used to reconstruct the perception rate. To address this challenge, the quadratic transformation method is used to reconstruct the perception rate by introducing auxiliary variables. Specifically, by introducing the constructed proxy function f(x, P) = 2Re(x H s(P)-x H I(P)x), where and I(P)>0, we can get And the optimal solution on the right side of the equation is By replacing the variables in the proxy function with the useful signal and interference signal in the perception rate, we can get At this point, problem (17) can be reconstructed as in and They include inherent variables and introduced auxiliary variables respectively; although and The coupling between them makes problem (25) difficult to solve directly, but it can be iteratively optimized by using a rotation optimization method. and Fixed Solving equations (19), (22) and (23) Fixed variables Solved by Convex Optimization Toolbox Step 4: Optimize the hybrid beamforming matrices F and W. The specific process is as follows: Variables F and W only appear in the last two terms of the objective function, so problem (14) can be simplified to Problem (26) involves the coupling between F and W, which is difficult to solve directly. Therefore, a rotation optimization method is used to optimize one variable while fixing the other variable. When the variable F is fixed, the objective function of problem (26) is transformed into This problem can be solved by the first-order optimal condition to find the optimal W, that is, W * =(F H F) -1 FH(P+ρD) (27) When the variable W is fixed, problem (26) can be transformed into Where Y = WW H and Z = (P + ρD)W H ; Since the module constraints in constraint (28) are separable, the optimization problem (28) can be equivalent to The optimal solution to problem (29) is in and is the optimal solution obtained in the previous iteration; Step 5: Based on the above steps 1 to 4, a two-layer iterative optimization algorithm can be designed to solve the optimization problem (14). The pseudo code of the specific solution process is as follows: Pseudocode of the algorithm for solving problem (14):

Images