Full-duplex cooperative NOMA integrated sensing communication system based on RIS and optimization method

Abstract

The invention discloses a full-duplex cooperative NOMA integrated sensing communication system based on RIS and an optimization method, and relates to the technical field of wireless communication. The problem that an existing RIS-combined sensing communication system is not combined with a full duplex cooperation mechanism is solved, and meanwhile, a base station beam forming matrix, a power distribution factor and an RIS phase shift matrix are jointly optimized so as to solve the problems that in the RIS-combined sensing communication system, interference among users is large, the remote user communication quality is poor and the resource distribution efficiency is low. The communication performance of the user is improved through the help of the RIS and the optimization of the algorithm; through algorithm optimization, the beam forming gain is enhanced, and the target detection precision is improved; through integration of communication and perception, the cost of hardware is reduced; the communication quality of the remote users is guaranteed through the cooperation of decoding retransmission of the near users and the RIS; and through reasonable distribution of resources, the energy efficiency is greatly improved.

Description

Technical Field

[0001] The present invention belongs to the field of wireless communication technologies, and particularly relates to a full-duplex cooperative NOMA integrated sensing and communication system based on RIS and an optimization method. Background Art

[0002] With the increasing demand for effectively utilizing spectrum and energy in interconnected devices, integrated sensing and communication (ISAC) has become a cornerstone technology for next-generation wireless networks. ISAC realizes the dual functions of environmental sensing and data transmission within a shared frequency band, eliminates hardware costs, and significantly improves spectrum efficiency.

[0003] With the advancement of 6G communication technologies, non-orthogonal multiple access (NOMA), cooperative NOMA, and reconfigurable intelligent surface (RIS) technologies have gradually become key technologies for enhancing the performance of sensing and communication systems. NOMA improves system capacity by allowing multiple users to share channels on the same spectrum resources, but traditional NOMA has significant inter-user interference and far-user path loss problems. Cooperative NOMA enhances the capacity of far users through user cooperation but still faces self-interference and cross-cluster interference problems. As a new type of signal reflection device, RIS can enhance signal quality and transmission efficiency by dynamically adjusting the phase of the reflection surface. However, existing sensing and communication systems incorporating RIS do not combine a full-duplex cooperation mechanism, resulting in insufficient resource allocation and beamforming efficiency. Although hybrid NOMA schemes alleviate interference, they do not solve the coupling problem of power allocation and phase optimization in multi-user scenarios. Summary of the Invention

[0004] Aiming at the deficiencies of the prior art, the present invention provides a full-duplex cooperative NOMA integrated sensing and communication system based on RIS and an optimization method, which solves the problem that existing sensing and communication systems incorporating RIS do not combine a full-duplex cooperation mechanism. At the same time, the base station beamforming matrix, power allocation factor, and RIS phase shift matrix are jointly optimized to solve the problems of large inter-user interference, poor communication quality of far users, and low resource allocation efficiency in sensing and communication systems incorporating RIS.

[0005] The first aspect of the present invention provides a full-duplex cooperative NOMA integrated sensing and communication system based on RIS, including:

[0006] Base station: Generates signals and uses M antennas equipped to transmit signals to serve users in G' clusters through a reconfigurable intelligent surface; receives the echo signals reflected by radar targets to sense I radar targets; the signals generated by the base station can be used for communication and sensing simultaneously; the users include near users and far users;

[0007] Reconfigurable intelligent surface: Utilizes N reflection elements to adjust the phase of the incident signals and reflects the adjusted signals;

[0008] G clusters: Each cluster includes a near user U(g,n) and a far user U(g,f), where g is the cluster number. The near user in the same cluster represents the user closer to the base station, and the far user represents the user farther from the base station. At the same time, each cluster is assigned a sub-resource block, and the far users and near users within each cluster adopt non-orthogonal multiple access NOMA;

[0009] I radar targets: I targets to be sensed within the user-defined range.

[0010] The RIS-based full-duplex cooperative NOMA integrated sensing and communication system includes two communication modes: direct transmission and cooperative transmission, and the two communication modes are carried out simultaneously during the communication process;

[0011] In direct transmission, the base station sends signals to the users in each cluster. Then, the near user U(g,n) decodes the signals of the far user U(g,f) through successive interference cancellation, and then subtracts the decoded signals of the far user from the received signals to obtain the signals of the near user itself; on the other hand, the far user directly decodes its own signals;

[0012] In cooperative transmission, the near user U(g,n) receives the signals and decodes and retransmits them to the far user U(g,f), and the near user uses the full-duplex mode to retransmit the signals to the far user; the far user U(g,f) combines the signals retransmitted by the near user and the signals sent by the base station and decodes them.

[0013] The second aspect of the present invention provides an optimization method for an RIS-based full-duplex cooperative NOMA integrated sensing and communication system, which is used to optimize an RIS-based full-duplex cooperative NOMA integrated sensing and communication system, and includes the following steps:

[0014] Establish an objective function for maximizing the total data rate and total sensing beam gain of the users and its constraints;

[0015] Transform the established objective function for maximizing the total data rate and total sensing beam gain of the communication users into three sub-objective functions; the three sub-objective functions are the power allocation objective function within the cluster, the beamforming objective function, and the RIS phase shift objective function;

[0016] Use the AO algorithm to optimize and solve the three established sub-objective functions to obtain the optimized beamforming matrix, RIS phase shift matrix, and power allocation parameters; the power allocation parameters include the power allocation factor of the base station and the power allocation factor of the near user.

[0017] Further, the process of establishing the objective function for maximizing the total data rate and total sensing beam gain of the users and its constraints is specifically as follows:

[0018] The signal received by the near user is expressed as:

[0019]

[0020] where y g,n represents the signal received by the near user, represents the channel from the reconfigurable intelligent surface to the near user, H represents the conjugate transpose of the matrix, represents the set of complex numbers; Φ = diag(v) is the RIS phase shift matrix, and is the reflection vector, j represents the imaginary unit, φ l represents the phase shift angle of the l-th reflection element and follows φ l ∈[0, 2π), l represents the number of the reflection element, represents the channel from the base station to the reconfigurable intelligent surface, is the beamforming vector; x g,n and represent the signals required by the near user U(g, n) and the far user U(g, f) respectively, α g,n and α g,f represent the power allocation factors corresponding to the near user U(g, n) and the far user U(g, f) respectively, n g,n is the Gaussian white noise, obeying zero mean and standard deviation of P g,n represents the energy retransmitted by the near user, represents the self-interference coefficient, satisfies where i represents the i-th time slot, τ is the decoding time;

[0021] The signal received by the far user is expressed as:

[0022]

[0023] where y g,f represents the signal received by the far user, is the channel from the reconfigurable intelligent surface to the far user, and are the channels from the near user U(g, n) to the reconfigurable intelligent surface, from the reconfigurable intelligent surface to the far user U(g, f), and from the near user U(g, n) to the far user U(g, f) respectively in the cooperative transmission; n g,f is the Gaussian white noise, obeying zero mean and standard deviation of

[0024] The signal-to-noise ratio for the near user U(g, n) to decode the signal of the far user U(g, f) is:

[0025]

[0026] Among them, S g,n→f represents the signal-to-noise ratio for the near user U(g,n) to decode the signal of the far user U(g,f);

[0027] The signal-to-noise ratio for the near user to decode its own signal is:

[0028]

[0029] Among them, S g,n represents the signal-to-noise ratio for the near user to decode its own signal;

[0030] The data rate of the near user is:

[0031] R g,n = log 2 (1 + S g,n ) (5)

[0032] Among them, R g,n represents the data rate of the near user;

[0033] The signal-to-noise ratio of the far user is:

[0034]

[0035] Among them, represents the signal-to-noise ratio of the far user;

[0036] The actual data rate of the far user is:

[0037]

[0038] Among them, R g,f represents the actual data rate of the far user;

[0039] The total data rate of the user is:

[0040]

[0041] Among them, R sum represents the total data rate of the user;

[0042] The sensing beam gain for the target detection angle of θ i is:

[0043]

[0044] Among them, θ i represents the target detection angle, i represents the number of the radar target, and P(θ i ) is the sensing beam gain for the target detection angle of θ i ; is the steering vector of the reconfigurable intelligent surface, and λ is the carrier wavelength. represents the distance between adjacent reflecting elements;

[0045] The total sensing beam gain of I radar targets is:

[0046]

[0047] where P(θ) represents the total sensing beam gain;

[0048] The objective function for maximizing the total data rate and total sensing beam gain of communication users is:

[0049]

[0050] The constraints of the objective function for maximizing the total data rate and total sensing beam gain of communication users are:

[0051]

[0052] where W = [w 1 ,..., w g ,..., w G' is the beamforming matrix, α = [α 1,f ,..., α g,f ,..., α G',f is the power allocation factor of the base station, P N = [P 1,n ,..., P g,n ,..., P G',n is the energy retransmitted by all near users; β c and β s are given normalization parameters, P b is the maximum energy threshold of the base station, is the minimum communication rate threshold of the far user, is the minimum communication rate threshold of the near user, P min (θ i ) is the minimum sensing beam gain threshold, is the maximum retransmission power threshold.

[0053] Furthermore, the power allocation objective function within the cluster is:

[0054]

[0055] The constraints of the power allocation objective function within the cluster are:

[0056]

[0057]

[0058] Among them, S n→f =[S 1,n→f ,....,S G',n→f , S n =[S 1,n ,....,S G',n ; X = [X 1 ,...,X g ,...,X G' , X g is a slack variable; among them, S n→f =[S 1,n→f ,....,S G',n→f , S n =[S 1,n ,....,S G',n ; X = [X 1 ,...,X g ,...,X G' , X g is a slack variable; t represents the number of iterations, represents the real result of the power allocation factor of the nearby user U(g,n) obtained in the previous iteration, represents the real result of the energy retransmitted by the nearby user obtained in the previous iteration.

[0059] Furthermore, the beamforming objective function is:

[0060]

[0061] The constraints of the beamforming objective function are:

[0062]

[0063]

[0064] Among them, p is a slack variable, I n→f =[I 1,n→f ,...,I g,n→f ,...,I G',n→f , I g,n→f is a slack variable, is a slack variable, diag represents a diagonal matrix; represents taking the real part, v (t) represents the real result of the reflection vector obtained in the previous iteration, and represent the real results of the slack variables obtained in the previous iteration.

[0065] Furthermore, the RIS phase shift objective function is as follows:

[0066]

[0067] The constraints of the RIS phase shift objective function are:

[0068]

[0069] where Tr represents taking the trace, and rank represents taking the rank. V = vv H , K i = diag{a H (θ i )G}; |||| * is the nuclear norm, |||| 2 is the spectral norm, ρ > 0 is a penalty factor. is the real result of W g obtained from the previous iteration. is the eigenvector corresponding to the largest eigenvalue.

[0070] Furthermore, using the AO algorithm to optimize the beamforming matrix, RIS phase shift matrix, and power allocation parameters based on the established three sub-objective functions, the optimized beamforming matrix, RIS phase shift matrix, and power allocation parameters are obtained; the power allocation parameters include the power allocation factor of the base station and the power allocation factor of the near user, specifically:

[0071] S1: Initialize the BS beamforming matrix, RIS phase shift matrix, and power allocation parameters;

[0072] S2: Solve the power allocation objective function through multiple iterations of CVX to update the power allocation factors of the base station and the near user;

[0073] S3: Solve the RIS phase shift objective function through multiple iterations of CVX to update the RIS phase shift matrix;

[0074] S4: Solve the beamforming objective function through multiple iterations of CVX to update the beamforming matrix, and increase the penalty factor in each iteration until the objective function converges;

[0075] S5: Determine whether the objective function of maximizing the total data rate and total sensing beam gain of the communication users converges. If so, the optimized beamforming matrix, RIS phase shift matrix, and power allocation parameters are obtained; otherwise, return to S2.

[0076] The third aspect of the present invention provides an optimization system for a RIS-based full-duplex cooperative NOMA integrated sensing and communication system, which is used to implement the optimization method of the RIS-based full-duplex cooperative NOMA integrated sensing and communication system, including:

[0077] An objective function construction module, which is used to establish an objective function for maximizing the total data rate and total sensing beam gain of users and its constraints;

[0078] An objective function decomposition module, which is used to transform the established objective function for maximizing the total data rate and total sensing beam gain of communication users into three sub-objective functions; the three sub-objective functions are respectively the power allocation objective function within the cluster, the beamforming objective function, and the RIS phase shift objective function;

[0079] A solution module, which is used to optimize the BS beamforming matrix, the RIS phase shift matrix, and the power allocation parameters based on the established three sub-objective functions by using the AO algorithm, and obtain the optimized beamforming matrix, the RIS phase shift matrix, and the power allocation parameters; the power allocation parameters include the power allocation factor of the base station and the power allocation factor of the near user.

[0080] The fourth aspect of the present invention provides an electronic device, including: a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device runs, the processor communicates with the memory through the bus. When the machine-readable instructions are executed by the processor, the steps of the optimization method of the RIS-based full-duplex cooperative NOMA integrated sensing and communication system are executed.

[0081] The fifth aspect of the present invention provides a computer-readable storage medium, in which a computer program is stored. When the computer program is run by a processor, the steps of the optimization method of the RIS-based full-duplex cooperative NOMA integrated sensing and communication system are executed.

[0082] Compared with the prior art, the beneficial effects of the present invention are:

[0083] 1. Communication performance: With the help of RIS and the optimization of the algorithm, the communication performance of users is improved;

[0084] 2. Sensing performance: Through algorithm optimization, the beamforming gain is enhanced and the target detection accuracy is improved;

[0085] 3. Spectrum utilization rate: Because the communication signal and the sensing signal share the same spectrum and NOMA is used among users, the spectrum utilization rate is greatly improved;

[0086] 4. Hardware cost: Through the integration of communication and sensing, the hardware cost is reduced;

[0087] 5. Communication guarantee for distant users: Through the decoding and retransmission of nearby users and the cooperation of RIS, the communication quality of distant users is guaranteed;

[0088] 6. Energy efficiency: Through the reasonable allocation of resources, the energy efficiency is greatly improved. Description of the drawings

[0089] Figure 1 It is a schematic diagram of the RIS-based full-duplex cooperative NOMA integrated sensing and communication system in the embodiment of the present invention;

[0090] Figure 2 It is a convergence performance diagram of the RIS-based full-duplex cooperative NOMA integrated sensing and communication system optimization method in the embodiment of the present invention under different numbers of base station antennas and RIS reflection elements;

[0091] Figure 3 It is a schematic diagram of the total data rate of communication users under different numbers of RIS reflection elements in the embodiment of the present invention;

[0092] Figure 4 It is a schematic diagram of the total data rate of communication users under different numbers of BS antennas in the embodiment of the present invention;

[0093] Figure 5 It is a schematic diagram of the total data rate of communication users under different maximum energy thresholds of the base station in the embodiment of the present invention;

[0094] Figure 6 It is a schematic diagram of the total data rate of communication users under different communication rate thresholds in the embodiment of the present invention;

[0095] Figure 7 It is a schematic diagram of the sensing beam gain of three schemes under different target detection angles in the embodiment of the present invention. Detailed implementation manners

[0096] The present invention will be described in detail below with reference to the drawings and embodiments.

[0097] As Figure 1 shown, this embodiment provides a RIS-based full-duplex cooperative NOMA integrated sensing and communication system, including:

[0098] Base station (BS): Generates signals and uses the equipped M antennas to transmit signals to the reconfigurable intelligent surface (RIS) to serve users in G′ clusters; Receives the echo signals reflected by the radar targets to realize the sensing of I radar targets; The signals can be used for communication and sensing simultaneously; The users include nearby users and distant users;

[0099] Reconfigurable Intelligent Surface (RIS): Since the direct links from the base station to the users and radar targets are blocked, only the link from the base station to the RIS and then to the users and radar targets exists. Therefore, by using the equipped N reflection elements, the phase of the incident signal is adjusted and the adjusted signal is reflected, i.e., the echo signal, to optimize signal transmission;

[0100] G' clusters: Each cluster includes a near user U(g,n) and a far user U(g,f), where g is the cluster number. The near user in the same cluster represents the user closer to the base station, and the far user represents the user farther from the base station. At the same time, each cluster is assigned a sub - resource block (RB), and the near and far users within each cluster use non - orthogonal multiple access NOMA;

[0101] I radar targets: I targets to be sensed near the users;

[0102] In this embodiment, the base station (BS) equipped with an antenna is located at the origin, and the reconfigurable intelligent surface RIS equipped with reflection elements is located at (20, 20). Three clusters are set, and the users in each cluster are randomly distributed on circles with a radius of 3m centered at (10, 5), (26, 5), and (15, 11). The angles of the two radar targets are set to 30° and 20° respectively. The power budgets of the base station BS and the near users are 35dBm and 25dBm respectively. The noise power is set to - 90dBm. In addition, all channels follow the Rice fading distribution with a Rice factor of 3. The path loss exponent is set to 2.2, and the path loss at a reference distance of 1 meter is - 30dB.

[0103] The communication modes of the RIS - based full - duplex cooperative NOMA integrated sensing and communication system include two communication modes: direct transmission (DT) and cooperative transmission (CT), and these two communication modes are carried out simultaneously during the communication process;

[0104] In direct transmission (DT), the base station (BS) sends signals to the users in each cluster. Then, the near user U(g,n) decodes the signal of the far user U(g,f) through successive interference cancellation (SIC), and then subtracts the decoded signal of the far user from the received signal to obtain the signal of the near user itself; on the other hand, the far user directly decodes its own signal;

[0105] In cooperative transmission (CT), the near user U(g,n) receives the signal, decodes it and re - transmits it to the far user U(g,f), and the near user re - transmits the signal to the far user in full - duplex mode; the far user U(g,f) combines the signal re - transmitted by the near user and the signal sent by the base station and decodes it;

[0106] This embodiment provides an optimization method for a full-duplex cooperative NOMA integrated sensing and communication system based on RIS, including the following steps:

[0107] Step 1: Establish an objective function that maximizes the total data rate and total sensing beam gain of users and its constraints;

[0108] In this embodiment, the full-duplex mode is considered, which results in self-interference (SI) of the near user. Let SI represent the channel coefficient, which follows a complex symmetric Gaussian random variable with a mean of zero and a standard deviation of σ SI Therefore, the signal received by the near user is expressed as:

[0109]

[0110] where y g,n represents the signal received by the near user, represents the channel from the reconfigurable intelligent surface (RIS) to the near user, H represents the conjugate transpose of the matrix, represents the set of complex numbers; Φ = diag(v) is the RIS phase shift matrix, and is the reflection vector, j represents the imaginary unit, φ l represents the phase shift angle of the l-th reflection element and follows φ l ∈[0, 2π), l represents the number of the reflection element, represents the channel from the base station (BS) to the reconfigurable intelligent surface (RIS), is the beamforming vector; x g,n and x g,f respectively represent the signals required by the near user U(g,n) and the far user U(g,f), α g,n and α g,f respectively represent the power allocation factors corresponding to the near user U(g,n) and the far user U(g,f), n g,n is the Gaussian white noise, obeying a zero mean and a standard deviation of P g,n represents the energy retransmitted by the near user, represents the self-interference coefficient, and satisfies where i represents the i-th time slot, and τ is the decoding time, so it will cause interference when the near user decodes the far user;

[0111] For the far user U(g,f), it simultaneously receives signals from the base station (BS) and the near user U(g,n). Therefore, the signal received by the far user is expressed as:

[0112]

[0113] where, yg,f Denote the signal received by the far user, is the channel from the reconfigurable intelligent surface (RIS) to the far user, and are the channels from the near user U(g,n) to the reconfigurable intelligent surface (RIS), from the reconfigurable intelligent surface (RIS) to the far user U(g,f), and from the near user U(g,n) to the far user U(g,f) in cooperative transmission (CT), respectively; n g,f is Gaussian white noise, following a zero mean and a standard deviation of

[0114] Therefore, the signal-to-noise ratio (SINR) for the near user U(g,n) to decode the signal of the far user U(g,f) is:

[0115]

[0116] where, S g,n→f represents the signal-to-noise ratio for the near user U(g,n) to decode the signal of the far user U(g,f);

[0117] After successfully using successive interference cancellation (SIC), the signal of the far user can be removed. Therefore, the signal-to-noise ratio (SINR) for the near user to decode its own signal is:

[0118]

[0119] where, S g,n represents the signal-to-noise ratio for the near user to decode its own signal;

[0120] Therefore, the data rate of the near user is:

[0121] R g,n = log 2 (1 + S g,n ) (5)

[0122] where, R g,n represents the data rate of the near user;

[0123] The far user can receive its own signal from DT and CT and use the maximum ratio combining (MRC) technique. Since the signals from the base station and the near user are completely decomposable at the far user, the signal-to-noise ratio (SINR) of the far user is:

[0124]

[0125] where, represents the signal-to-noise ratio of the far user;

[0126] Due to the application of NOMA, the actual data rate of the far user is:

[0127]

[0128] Among them, R g,f represents the actual data rate of the far user;

[0129] Therefore, the total data rate of the user is:

[0130]

[0131] Among them, R sum represents the total data rate of the user;

[0132] Since the signal is dual-functional (communication and sensing), the target detection angle is θ i The sensing beam gain of is:

[0133]

[0134] Among them, θ i represents the target detection angle, i represents the number of the radar target, P(θ i ) is the sensing beam gain with the target detection angle of θ i ; is the steering vector of the reconfigurable intelligent surface (RIS), λ is the carrier wavelength, represents the distance between adjacent reflection elements;

[0135] Therefore, the total sensing beam gain of I radar targets is:

[0136]

[0137] Among them, P(θ) represents the total sensing beam gain;

[0138] To effectively utilize the proposed system, the objective of this application is to simultaneously optimize the communication and sensing performance, that is, under the given normalized parameters β c and β s , by jointly optimizing the beamforming matrix under the base station (BS), the RIS phase shift matrix and the transmission power of the near user, maximize the total rate and the total sensing beam gain of the communication user:

[0139] Therefore, the objective function for maximizing the total data rate and the total sensing beam gain of the communication user is:

[0140]

[0141] The constraints of the objective function for maximizing the total data rate and the total sensing beam gain of the communication user are:

[0142]

[0143]

[0144] Among them, \(W = [w 1 ,\cdots,w g ,\cdots,w G' \) is the beamforming matrix, \(\alpha = [\alpha 1,f ,\cdots,\alpha g,f ,\cdots,\alpha G',f \) is the power allocation factor of the base station, \(P N = [P 1,n ,\cdots,P g,n ,\cdots,P G',n \) is the energy for retransmission of all nearby users; \(\beta c \) and \(\beta s \) are given normalization parameters, \(P b \) is the maximum energy threshold of the base station, is the minimum communication rate threshold of the far user, is the minimum communication rate threshold of the nearby user, \(P min (\theta i )\) is the minimum perceived beam gain threshold, is the maximum retransmission power threshold; (12) is the power threshold of the base station BS, (13) and (14) ensure the quality of service (QoS), (15) ensures the minimum perceived beam gain of each target, (16) is the power limit for nearby users, (17) and (18) represent the constraint conditions of the power allocation factor, (19) is the modulus constraint of each RIS reflection element;

[0145] Step 2: Transform the established objective function that maximizes the total data rate and total perceived beam gain of communication users into three sub-objective functions; the three sub-objective functions are the in-cluster power allocation objective function, the beamforming objective function, and the RIS phase shift objective function respectively;

[0146] The proposed problem (11) is highly coupled and non-convex. To address this problem, it is divided into three solvable sub-problems;

[0147] To handle the non-convexity of the objective function, a relaxation variable is introduced. However, expressions (3), (4), and (6) in problem (11) still exhibit non-convexity. Therefore, relax expressions (3), (4), and (6), where problem (11) can be reformulated to obtain the in-cluster power allocation objective function as:

[0148]

[0149] The constraints of the in-cluster power allocation objective function are:

[0150]

[0151]

[0152] Among them, S n→f =[S 1,n→f ,....,S G',n→f , S n =[S 1,n ,....,S G',n , X = [X 1 ,...,X g ,...,X G' , X g is a slack variable;

[0153] However, constraints (21), (22), and (23) are not convex; to solve this problem, the SCA method is adopted, and the first-order Taylor expansion (FTE) approximation is used to transform the non-convex functions (21), (22), and (23) into affine forms. The reformulated form is:

[0154]

[0155] Among them, t represents the number of iterations, represents the real result of the power allocation factor of the near user U(g,n) obtained in the previous iteration, represents the real result of the energy retransmitted by the near user obtained in the previous iteration. Those without the superscript t are the parameters to be solved in this iteration;

[0156] Therefore, this problem (20) is further reformulated as:

[0157]

[0158] s.t. (14), (16), (17), (18), (24)-(27), (29)-(31) (33)

[0159] This problem (32) is now convex and can be easily solved by standard optimization tools such as CVX;

[0160] Similar to the design of power allocation, auxiliary variables are added, and the expressions of (3), (4), and (6) are relaxed, and the slack variables I g,n→f and are used to convexify the objective problem. Therefore, problem (11) can be reformulated as the RIS phase shift objective function:

[0161]

[0162] The constraints of the RIS phase shift objective function are:

[0163]

[0164] (13),(14),(15),(19) (43)

[0165] where p is the slack variable, I n→f = [I 1,n→f ,...,I g,n→f ,...,I G',n→f , I g,n→f is the slack variable, is the slack variable, diag represents a diagonal matrix;

[0166] Constraints (35), (36) and (37) are non-convex. Here, SCA is applied to convert them into affine forms:

[0167]

[0168]

[0169] where denotes taking the real part, v (t) represents the real result of the reflection vector obtained in the previous iteration, and represent the real results of the slack variables obtained in the previous iteration;

[0170] Similarly, the non-convex constraint (42) can also be rewritten as:

[0171]

[0172] Therefore, problem (34) can be reformulated as:

[0173]

[0174] s.t.(13),(14),(15),(19),(38)-(41),(44)-(47) (49)

[0175] This problem (48) is presented as convex and can be solved using CVX;

[0176] Problem (11) can be reshaped into a beamforming objective function:

[0177]

[0178] The constraints of the beamforming objective function are:

[0179]

[0180]

[0181] Among them, Tr represents taking the trace, and rank represents taking the rank. V = vv H , K i = diag{a H (θ i )G};

[0182] Constraints (52) and (53) are not convex and can be reformulated as:

[0183]

[0184] The non-convexity of these two constraints (59) and (60) is caused by the first term on the right side, which is underlined. Therefore, SCA is invoked to transform these terms through FTE, and these terms are respectively:

[0185]

[0186] Then, define:

[0187]

[0188] For the rank-1 constraint (58), it is transformed into an equivalent equality constraint:

[0189]

[0190] where || || * is the nuclear norm, || || 2 is the spectral norm. Considering that W g is positive semi-definite, so ||W g || * - ||W g || 2 > 0 always holds. Then, we introduce (65) as a penalty term to obtain a rank-1 matrix. The original problem (50) can be reformulated as:

[0191]

[0192] s.t. (58), (54)-(57), (63), (64) (67)

[0193] where ρ > 0 is a penalty factor. However, the penalty term makes the objective non-convex. Therefore, FTE of the penalty term is adopted, and this problem (66) can be transformed through the following problem:

[0194]

[0195] s.t. (58), (54)-(57), (63), (64) (69)

[0196] wherein, is the W obtained from the previous iteration g the real number result of is the eigenvector corresponding to the maximum eigenvalue. This problem (68) is currently convex and can be effectively solved by CVX;

[0197] Step 3: Use the AO algorithm to optimize and solve the three established sub-objective functions to obtain the optimized beamforming matrix, RIS phase shift matrix, and power allocation parameters; the power allocation parameters include the power allocation factor of the base station and the power allocation factor of the near user;

[0198] Step 3.1: Initialize the BS beamforming matrix, RIS phase shift matrix, and power allocation parameters;

[0199] Step 3.2: Solve the power allocation objective function shown in formula (32) through multiple iterations of CVX to update the power allocation factor of the base station (BS) and the power allocation factor of the near user;

[0200] Step 3.3: Solve the RIS phase shift objective function shown in formula (48) through multiple iterations of CVX to update the RIS phase shift matrix;

[0201] Step 3.4: Solve the beamforming objective function shown in formula (69) through multiple iterations of CVX to update the beamforming matrix. It should be noted that here the penalty factor needs to be increased in each iteration until the objective function converges;

[0202] Step 3.5: Determine whether the objective function of maximizing the total data rate and total sensing beam gain of the communication users shown in formula (11) converges. If so, obtain the optimized beamforming matrix, RIS phase shift matrix, and power allocation parameters, otherwise return to Step 3.2.

[0203] As Figure 2 shown, it shows the convergence behavior of the algorithm under different numbers of BS antennas and RIS reflection elements. It can be seen that the objective value converges to a stable value in several outer iterations. In addition, as the number of BS antennas and RIS reflection elements increases, the objective value also increases. In addition, it is worth noting that the increase in the number of antennas and RIS reflection elements has no significant impact on the number of iterations before the algorithm converges;

[0204] As Figure 3As shown, it shows the impact of the number of RIS reflection elements on the total data rate of communication users. It can be clearly observed that the total data rate of communication users increases with the increase in the number of RIS reflection elements. The reason is that the increase in the number of RIS reflection elements leads to an enhancement of the beamforming gain, thus bringing a greater SINR for communication users. In addition, it can be noted that this figure compares the optimal phase and random phase schemes of the RIS. It can be seen that the optimal phase scenario is superior to the random phase scenario, and the increase in the total data rate of communication users caused by the increase in the number of RIS reflection elements is not significant. In addition, this scheme has better performance compared with the NOMA scheme and the OMA scheme.

[0205] As Figure 4 shown, it is obvious that more BS antennas will cause an enhancement of the total data rate of communication users, which is due to the higher beamforming gain caused by more active beams. In addition, this figure also compares the schemes with fixed power allocation factors and optimal power allocation factors. The results show that the scheme with the optimal power allocation factor has better performance than the scheme with the fixed power allocation factor. In addition, the optimal power allocation scheme of this scheme is significantly superior to other baseline schemes;

[0206] As Figure 5 shown, it can be seen that the total data rate of communication users increases with the increase in the maximum energy threshold of the base station. This is because the signal strength increases with the increase in the transmission power. In addition, it is also found that by increasing the maximum energy threshold of the base station in the CNOMA scheme, there is a considerable increase in the total data rate of communication users. As expected, under the same maximum energy threshold of the base station, the scheme proposed in the present invention is superior to the NOMA and OMA methods;

[0207] As Figure 6 shown, it can be clearly observed that increasing the value of the communication rate threshold will reduce the total data rate of communication users, and it can also be observed that the increase in the sensing power threshold value at each radar target will lead to a decrease in the target value. This can be attributed to more sensing capabilities and less communication capabilities. Although at the same communication rate and sensing power, with the increase in the communication rate threshold, the total data rate value of communication users decreases, the effect of this scheme is superior to the NOMA and OMA schemes;

[0208] As Figure 7 shown, it shows the sensing beam gain of the proposed scheme and two baseline schemes at the target detection angle, where the target detection angle is in the directions of 20 degrees and -30 degrees. It can be clearly noted that the proposed scheme can reach the dominant peak in the direction of interest, which is sharper compared with other schemes. It can provide more sensing capabilities for the BS while taking into account the cooperation between communication users.

[0209] This embodiment provides an optimization system for a RIS-based full-duplex cooperative NOMA integrated sensing and communication system, which is used to implement the optimization method of the RIS-based full-duplex cooperative NOMA integrated sensing and communication system, including:

[0210] A target function construction module, which is used to establish a target function for maximizing the total data rate and total sensing beam gain of users and its constraints;

[0211] A target function decomposition module, which is used to transform the established target function for maximizing the total data rate and total sensing beam gain of communication users into three sub-target functions; the three sub-target functions are respectively the power allocation target function within the cluster, the beamforming target function, and the RIS phase shift target function;

[0212] A solution module, which is used to optimize the BS beamforming matrix, RIS phase shift matrix, and power allocation parameters based on the established three sub-target functions by using the AO algorithm, and obtain the optimized beamforming matrix, RIS phase shift matrix, and power allocation parameters; the power allocation parameters include the power allocation factor of the base station (BS) and the power allocation factor of the near user.

[0213] This embodiment provides an electronic device, including: a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device runs, the processor communicates with the memory through the bus. When the machine-readable instructions are executed by the processor, the steps of the optimization method of the RIS-based full-duplex cooperative NOMA integrated sensing and communication system are executed.

[0214] This embodiment provides a computer-readable storage medium, in which a computer program is stored. When the computer program is run by a processor, the steps of the optimization method of the RIS-based full-duplex cooperative NOMA integrated sensing and communication system as described above are executed.

Claims

1. A full-duplex cooperative NOMA integrated perception communication system based on RIS, characterized in that: include: Base station: generates signals and uses the M antennas to transmit signals to the reconstructed intelligent surface to provide services for users in G' clusters; Receiving echo signals reflected by radar targets to achieve the perception of I radar targets; the signals generated by the base station can be used for communication and perception at the same time; the users include near users and far users; Reconfigurable smart surface: uses N reflective elements to adjust the phase of the incident signal and reflect the adjusted signal; G' clusters: Each cluster includes a near user U(g,n) and a far user U(g,f), where g is the cluster number. The near users in the same cluster are users that are close to the base station, and the far users are users that are far from the base station. At the same time, each cluster is assigned a sub-resource block. Non-orthogonal multiple access (NOMA) is used for the far and near users in each cluster. I radar target: I target to be sensed within the range set by the user; The full-duplex cooperative NOMA integrated sensing communication system based on RIS includes two communication modes: direct transmission and cooperative transmission. The two communication modes are carried out simultaneously during the communication process. In direct transmission, the base station sends a signal to each cluster user. Then, the near user U(g,n) decodes the signal of the far user U(g,f) through serial interference cancellation, and then subtracts the decoded signal of the far user from the received signal to obtain the near user's own signal. On the other hand, the far user directly decodes its own signal. In cooperative transmission, the near user U(g,n) receives the signal, decodes it and retransmits it to the far user U(g,f), and the near user retransmits the signal to the far user in full-duplex mode. The far user U(g,f) combines the signal retransmitted by the near user with the signal sent by the base station and decodes them.

2. A RIS-based full-duplex cooperative NOMA integrated perception communication system optimization method, used to optimize the RIS-based full-duplex cooperative NOMA integrated perception communication system described in claim 1, characterized in that: The following steps are involved: Establish the objective function and its constraints to maximize the user's total data rate and total perceived beam gain; The established objective function of maximizing the total data rate and total perceived beam gain of the communication users is converted into three sub-objective functions; the three sub-objective functions are respectively a power allocation objective function within the cluster, a beamforming objective function, and a RIS phase shift objective function; The AO algorithm is used to optimize and solve the three established sub-objective functions to obtain the optimized beamforming matrix, RIS phase shift matrix and power allocation parameters; the power allocation parameters include the power allocation factor of the base station and the power allocation factor of the near user.

3. According to claim 2, the optimization method of the full-duplex cooperative NOMA integrated perception communication system based on RIS is characterized in that: The process of establishing the objective function and its constraints for maximizing the total data rate and total perceived beam gain of the user is specifically as follows: The signal received by the near user is expressed as: Among them, y g,n Indicates the signal received by the near user. represents the channel from the reconfigurable smart surface to the near user, H represents the conjugate transpose of the matrix, represents a set of complex numbers; Φ = diag(v) is the RIS phase shift matrix, and is the reflection vector, j represents the imaginary unit, φ l represents the phase shift angle of the lth reflective element and follows φ l ∈[0,2π), l represents the number of the reflective element, represents the channel from the base station to the reconfigurable smart surface, is the beamforming vector; x g,n and They represent the signals required by the near user U(g,n) and the far user U(g,f), respectively, g,n and α g,f They represent the power allocation factors corresponding to the near user U(g,n) and the far user U(g,f), n g,n is Gaussian white noise with zero mean and standard deviation , P g,n represents the energy of retransmission by the nearest user, represents the self-interference coefficient, satisfy Where i represents the i-th time slot and τ is the decoding time; The signal received by the far user is expressed as: Among them, y g,f represents the signal received by the distant user, It is a channel from a reconfigurable smart surface to a remote user. and They are the channel from the near user U(g,n) to the reconfigurable smart surface, the channel from the reconfigurable smart surface to the far user U(g,f), and the channel from the near user U(g,n) to the far user U(g,f) in cooperative transmission; n g,f is Gaussian white noise with zero mean and standard deviation The signal-to-noise ratio of the signal of the far user U(g,f) decoded by the near user U(g,n) is: Among them, S g,n→f represents the signal-to-noise ratio of the signal of the far user U(g,f) decoded by the near user U(g,n); The signal-to-noise ratio of the near user decoding its own signal is: Among them, S g,n It represents the signal-to-noise ratio of the near user decoding its own signal; The data rate of the near user is: R g,n =log2(1+S g,n ) (5) Among them, R g,n Indicates the data rate near the user; The signal-to-noise ratio of the far user is: in, represents the signal-to-noise ratio of the far user; The actual data rate for the far user is: Among them, R g,f Indicates the actual data rate of the far user; The total data rate for the user is: Among them, R sum Indicates the total data rate of the user; The target detection angle is θ i The perceptual beam gain of is: Among them, θ i represents the target detection angle, i represents the number of the radar target, P(θ i ) is the target detection angle θ i The perceived beam gain of is the steering vector of the reconfigurable smart surface, λ is the carrier wavelength, represents the distance between adjacent reflective elements; The total perception beam gain of I radar targets is: Where P(θ) represents the total perceptual beam gain; The objective function for maximizing the total data rate and total perceived beam gain of the communication user is: The constraints of the objective function to maximize the total data rate and total perceived beam gain of the communication users are: Where W = [w1, ..., w g ,...,w G' ] is the beamforming matrix, α=[α 1,f ,...,α g,f ,...,α G',f ] is the power allocation factor of the base station, P N =[P 1,n ,...,P g,n ,...,P G',n ] is the energy of all nearby users’ retransmissions; β c and β s For a given normalization parameter, P b is the maximum energy threshold of the base station, is the minimum communication rate threshold of the remote user, is the minimum communication rate threshold of the nearest user, P min (θ i ) minimum perceived beam gain threshold, is the maximum retransmission power threshold.

4. According to claim 3, the optimization method of the full-duplex cooperative NOMA integrated perception communication system based on RIS is characterized in that: The power allocation objective function within the cluster is: The constraints of the power allocation objective function within the cluster are: Among them, S n→f =[S 1,n→f ,....,S G',n→f ],S n =[S 1,n ,....,S G',n ], X=[X1,...,X g ,...,X G' ], X g is the slack variable; where S n→f =[S 1,n→f ,....,S G',n→f ],S n =[S 1,n ,....,S G',n ], X=[X1,...,X g ,...,X G' ], X g is the slack variable; t represents the number of iterations, represents the real number result of the power allocation factor of the near user U(g,n) obtained in the previous iteration, The real number result representing the energy of the near user retransmission obtained in the previous iteration.

5. According to claim 4, the optimization method of the full-duplex cooperative NOMA integrated perception communication system based on RIS is characterized in that: The beamforming objective function is: The constraints of the beamforming objective function are: Among them, p is the slack variable, I n→f =[I 1,n→f ,...,I g,n→f ,...,I G',n→f ], I g,n→f is the slack variable, is the slack variable, diag represents a diagonal matrix; represents the real part, v (t) represents the real result of the reflection vector obtained in the previous iteration, and Represents the real result for the slack variable from the previous iteration.

6. According to claim 5, the optimization method of the full-duplex cooperative NOMA integrated perception communication system based on RIS is characterized in that: The RIS phase shift objective function is: The constraints of the RIS phase shift objective function are: Among them, Tr means finding the trace, and rank means taking the rank. V=vv H , K i =diag{a H (θ i )G}; ‖‖ * is the nuclear norm, ‖‖2 is the spectral norm, ρ>0 is a penalty factor, is the W obtained in the previous iteration g The real result of yes The eigenvector corresponding to the largest eigenvalue.

7. The RIS-based full-duplex cooperative NOMA integrated perception communication system optimization method according to claim 6 is characterized in that: The AO algorithm is used to optimize the beamforming matrix, RIS phase shift matrix and power allocation parameters based on the three established sub-objective functions to obtain the optimized beamforming matrix, RIS phase shift matrix and power allocation parameters; the power allocation parameters include the power allocation factor of the base station and the power allocation factor of the near user, specifically: S1: Initialize BS beamforming matrix, RIS phase shift matrix and power allocation parameters; S2: Solve the power allocation objective function through multiple iterations of CVX, and update the power allocation factor of the base station and the power allocation factor of the near user; S3: Solve the RIS phase shift objective function through multiple iterations of CVX and update the RIS phase shift matrix; S4: Solve the beamforming objective function through multiple iterations of CVX and update the beamforming matrix. The penalty factor is increased in each iteration until the objective function converges. S5: Determine whether the objective function of maximizing the total data rate and total perceived beam gain of the communication users converges. If so, obtain the optimized beamforming matrix, RIS phase shift matrix and power allocation parameters. Otherwise, return to S2.

8. A full-duplex cooperative NOMA integrated perception communication system optimization system based on RIS, used to implement the full-duplex cooperative NOMA integrated perception communication system optimization method based on RIS according to any one of claims 2-7, characterized in that: include: An objective function building module, used to establish an objective function and its constraints for maximizing the total data rate and total perceived beam gain of the user; An objective function decomposition module is used to convert the established objective function of maximizing the total data rate and total perceived beam gain of communication users into three sub-objective functions; the three sub-objective functions are respectively a power allocation objective function within the cluster, a beamforming objective function and a RIS phase shift objective function; The solution module is used to optimize the BS beamforming matrix, RIS phase shift matrix and power allocation parameters based on the established three sub-objective functions using the AO algorithm to obtain the optimized beamforming matrix, RIS phase shift matrix and power allocation parameters; the power allocation parameters include the power allocation factor of the base station and the power allocation factor of the near user.

9. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate through the bus, and when the machine-readable instructions are executed by the processor, the steps of the RIS-based full-duplex collaborative NOMA integrated perception communication system optimization method described in any one of claims 2-7 are performed.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, executes the steps of the RIS-based full-duplex collaborative NOMA integrated perception communication system optimization method as described in any one of claims 2 to 7.

Images