Resource Optimization Method and System for Integrated Sensing and Communication System Based on STAR-RIS and RSMA

Abstract

The present invention discloses a method and system for optimizing the resources of a communication-sensing integrated system based on STAR-RIS and RSMA, belonging to the field of wireless communication; the optimization method includes: S1, constructing an initial system model; S2, setting an iteration number variable, and giving an initial radar detection signal covariance matrix, an initial power allocation, an initial decoding order, and an initial transmission matrix and reflection matrix of the STAR-RIS; S3, under the premise of the given power allocation, radar detection signal covariance matrix, decoding order, and transmission matrix and reflection matrix of the STAR-RIS, optimizing the initial system model to obtain the optimal power allocation, the optimal covariance matrix, the optimal decoding order, and the optimal transmission matrix and reflection matrix; S4, substituting the optimal parameters obtained above into S3 based on the BCD method, and updating the iteration number variable until the iteration number is reached, the iteration terminates, and the optimal solution of the total user rate is obtained.

Description

Technical Field

[0001] The present invention belongs to the field of wireless communication, and particularly relates to a method and system for optimizing the resources of a communication and sensing integrated system based on STAR-RIS and RSMA. Background Art

[0002] With the rapid development of wireless communication technology, people's requirements for the performance of communication systems are getting higher and higher. While meeting the requirements of high-rate and large-capacity communication, realizing the integration of communication and sensing has become a current research hotspot. An integrated sensing and communication (ISAC) system can effectively utilize spectrum resources and improve the overall performance of the system. In an ISAC system, how to further improve the spectrum efficiency and energy efficiency of the system is a key problem to be solved urgently.

[0003] RSMA (rate splitting multiple access) is a non-orthogonal multiple access technology. By splitting the messages of users into multiple sub-messages for transmission, it can effectively improve the spectrum efficiency of the system. As an extension of NOMA (non-orthogonal multiple access) technology, RSMA has greater flexibility and overcomes the limitations of NOMA in decoding complexity and interference management. At the same time, the emergence of reconfigurable intelligent surface (RIS) technology provides a new way to optimize the wireless communication environment. Among them, STAR-RIS, as a new type of RIS, has a higher degree of freedom and can simultaneously realize the reflection and transmission of signals, providing the possibility to further improve the system performance. Most of the existing studies are based on NOMA and RIS technologies for resource optimization in communication and sensing integrated systems. However, the above methods have certain limitations in interference management and degrees of freedom.

[0004] Therefore, adopting RSMA and STAR-RIS technologies in communication and sensing systems can further improve the system performance. For this reason, a method for optimizing the resources of a communication and sensing integrated system based on STAR-RIS and RSMA is proposed. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the purpose of the present invention is to provide a method and system for optimizing the resources of a communication and sensing integrated system based on STAR-RIS and RSMA, which solves the problems in the prior art.

[0006] The purpose of the present invention can be achieved by the following technical solutions:

[0007] A method for optimizing the resources of a communication and sensing integrated system based on STAR-RIS and RSMA, comprising the following steps:

[0008] S1. Construct an initial model of the integrated communication and sensing system based on STAR-RIS and RSMA;

[0009] S2. Set the iteration number variable and assign the initial covariance matrix of the radar detection signal , the initial power allocation , the initial decoding order and the transmission matrix of the initial STAR-RIS and the reflection matrix ;

[0010] S3. On the premise of the given power allocation , the covariance matrix of the radar detection signal , the decoding order and the transmission matrix of STAR-RIS and the reflection matrix , optimize the initial model of the integrated communication and sensing system based on STAR-RIS and RSMA to obtain the optimal power allocation , the optimal covariance matrix , the optimal decoding order and the optimal transmission matrix and the reflection matrix ;

[0011] S4. Based on the BCD method, substitute the optimal power allocation , the optimal covariance matrix , the optimal decoding order and the optimal transmission matrix and the reflection matrix into S3, update the iteration number variable, and perform iterations until the iteration number is reached, then the iteration terminates to obtain the optimal solution of the total user rate.

[0012] Further, the integrated communication and sensing system based on STAR-RIS and RSMA includes: The integrated communication and sensing system includes single-antenna users, a full-duplex base station and STAR-RIS. STAR-RIS includes transmission and reflection elements; The users transmit messages to the base station through STAR-RIS and direct connection in the uplink RSMA mode, and at the same time, the base station detects a single radar target through STAR-RIS with a certain transmission power.

[0013] Further, the initial model of the integrated communication and sensing system based on STAR-RIS and RSMA is:

[0014]

[0015]

[0016]

[0017]

[0018]

[0019]

[0020] Among them, is the covariance matrix of the base station radar detection signal, and the total power of the signal can be obtained through the trace of its covariance matrix to calculate, is the decoding order of the base station when receiving user messages, is the power allocation of the user, , are the transmission matrix and reflection matrix of the STAR-RIS respectively, is the th user's transmission rate when sending messages through uplink RSMA, is the SINR of the radar target, is the threshold of the radar target SINR; is the maximum available power of the base station; is the th user's allocated power, is the maximum power of each user; , are the transmission amplitude and reflection amplitude of the th transmission and reflection element of the STAR-RIS; The set is defined as the set of all possible decoding orders of all messages from users; represents the th user's th sub-message allocated power.

[0021] Furthermore, 's calculation formula is:

[0022]

[0023]

[0024]

[0025] Among them, is the th user's th sub-message's transmission rate, For the th user's th sub-message, the SINR, For the th user's channel with the base station, For the th user's channel with the STAR-RIS, And are respectively the channels between the STAR-RIS and the base station's receiving and transmitting antennas, represents the radar target amplitude variance, represents the received noise variance, is the self-interference of the base station, is the channel between the STAR-RIS and the radar target, And respectively represent the numbers of the base station's transmitting and receiving antennas, are respectively the channels of the base station's transmitting and receiving antennas with the radar target; represents the decoding order of the th user's th sub-message, represents that the decoding order of the th user's th sub-message is after the th user's th sub-message;

[0026] Among them, the equivalent channel of the radar detection target , the base station equivalent interference are respectively defined as:

[0027]

[0028] .

[0029] Furthermore, when obtaining the optimal power allocation , the optimal covariance matrix , the optimal transmission matrix and the reflection matrix , the initial model of the integrated communication and sensing system based on STAR-RIS and RSMA is optimized by using the successive convex approximation method and the gradient descent method.

[0030] Furthermore, when obtaining the optimal decoding order , the greedy algorithm is used to optimize the initial model of the integrated communication and sensing system based on STAR-RIS and RSMA.

[0031] STAR-RIS and RSMA-based integrated communication and sensing system resource optimization system, comprising:

[0032] Model construction module: constructing an initial model of an integrated communication and sensing system based on STAR-RIS and RSMA;

[0033] Parameter initialization module: setting an iteration number variable and giving an initial covariance matrix of radar detection signals , initial power allocation , initial decoding order and an initial transmission matrix of STAR-RIS and reflection matrix ;

[0034] Optimization module: optimizing the initial model of the integrated communication and sensing system based on STAR-RIS and RSMA under the premise of given power allocation , covariance matrix of radar detection signals , decoding order and transmission matrix of STAR-RIS and reflection matrix to obtain the optimal power allocation , optimal covariance matrix , optimal decoding order and optimal transmission matrix and reflection matrix ;

[0035] And an iteration module: substituting the optimal power allocation , optimal covariance matrix , optimal decoding order and optimal transmission matrix and reflection matrix into the optimization module, updating the iteration number variable, and performing iteration until the iteration number is reached, the iteration terminates, and the optimal solution of the user total rate is obtained.

[0036] A computer storage medium stores a readable program, which can execute the above-mentioned STAR-RIS and RSMA-based integrated communication and sensing system resource optimization method when the program runs.

[0037] An electronic device, comprising: a processor, a memory, a communication interface and a communication bus, and the processor, the memory and the communication interface complete mutual communication through the communication bus;

[0038] The memory is used to store at least one executable instruction, and the executable instruction causes the processor to perform the operations corresponding to the above-mentioned integrated communication and sensing system resource optimization method based on STAR-RIS and RSMA.

[0039] A computer program product includes computer instructions, and the computer instructions direct a computing device to perform the operations corresponding to the above-mentioned integrated communication and sensing system resource optimization method based on STAR-RIS and RSMA.

[0040] Advantages of the present invention:

[0041] 1. Under the constraints of the maximum transmit power of the base station, the minimum transmit power of the user, and the minimum SINR of the radar target, the present invention studies the joint optimization of the covariance matrix, power allocation, decoding order of the radar detection signal transmitted by the base station, and the transmission matrix and reflection matrix of STAR-RIS to achieve the problem of maximizing the user rate, and proposes an iterative algorithm based on the gradient descent method, successive convex approximation, and block coordinate descent method.

[0042] 2. The previously proposed STAR-RIS-assisted RSMA integrated sensing and communication system framework does not consider the situation of uplink users transmitting messages through RSMA. The framework proposed by the present invention can more flexibly and conveniently adapt to special application scenarios and improve the user rate; the present invention considers multiple uplink users transmitting messages to the base station through STAR-RIS and direct connection with a certain transmit power, and aims to maximize the user rate under the constraints of the maximum transmit power of the base station, the minimum transmit power of the user, and the minimum SINR of the radar target, and jointly optimizes the covariance matrix, power allocation, decoding order of the radar detection signal transmitted by the base station, and the transmission matrix and reflection matrix of STAR-RIS.

[0043] 3. Since the proposed optimization problem is non-convex and difficult to solve, the present invention solves four sub-problems through the gradient descent method, successive convex approximation technique (SCA), and block coordinate descent method (BCD); the five variables of the four sub-problems respectively correspond to four blocks, namely the covariance matrix of the radar detection signal transmitted by the base station, power allocation, decoding order, and the transmission matrix and reflection matrix of STAR-RIS; by keeping the other block variables unchanged, the four sub-problems are solved through the gradient descent method and SCA method, and the optimization variables of the four sub-problems are alternately optimized; and the solution obtained in each iteration will be used as the input for the next iteration.

[0044] 4. The convergence of the algorithms of the gradient descent method, successive convex approximation technique (SCA), and block coordinate descent method (BCD) proposed in the present invention can be guaranteed, and the required complexity is relatively low. The simulation results show the user rates in different scenarios. Compared with existing schemes, such as non-orthogonal multiple access (NOMA) schemes, time division multiple access (TDMA) schemes, and rate splitting multiple access (RSMA) schemes without power allocation, the performance is significantly improved by using the proposed algorithms. Brief Description of the Drawings

[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0046] Figure 1 It is a structural diagram of a communication and sensing integrated system based on STAR-RIS and RSMA of the present invention;

[0047] Figure 2 It is a flowchart of a resource optimization method for a communication and sensing integrated system based on STAR-RIS and RSMA of the present invention;

[0048] Figure 3 It is a curve graph showing the relationship between the proposed scheme in the present invention, non-orthogonal multiple access scheme, time division multiple access scheme, and rate splitting multiple access without power allocation (WP-RSMA) scheme. Detailed Embodiments

[0049] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0050] Embodiment 1

[0051] As Figure 1 shown, a communication and sensing integrated system based on STAR-RIS and RSMA includes: a base station (BS), STAR-RIS, and single-antenna users. The base station is a full-duplex base station, and STAR-RIS includes transmission and reflection elements; and the uplink users transmit messages to the base station through STAR-RIS and direct connection with a certain transmission power, and at the same time, the base station detects a single radar target through STAR-RIS with a certain transmission power.

[0052] The STAR-RIS is divided into a transmission area and a reflection area. Among them, the uplink users are distributed in the reflection area of the STAR-RIS. The message of each user is divided into 2 sub-messages. The signal transmitted by the th user is , where is the th sub-message of the th user, and is the power allocated to the th user. represents the th sub-message of the th user allocated power; The radar target is located in the transmission area of the STAR-RIS; is the transmission matrix of the STAR-RIS, is the reflection matrix of the STAR-RIS, where: is the transmission phase shift matrix, is the reflection phase shift matrix, and are the transmission and reflection phases of the th element of the STAR-RIS respectively, is the transmission amplitude matrix, is the reflection amplitude matrix, and

[0053] Embodiment 2

[0054] As Figure 2 shown, the resource optimization method for the integrated communication and sensing system based on STAR-RIS and RSMA includes the following steps:

[0055] S1, constructing an initial model of the integrated communication and sensing system based on STAR-RIS and RSMA;

[0056] The initial model of the integrated communication and sensing system based on STAR-RIS and RSMA is:

[0057]

[0058]

[0059]

[0060]

[0061]

[0062]

[0063] Among them, is the covariance matrix of the base station radar detection signal, and the total power of the signal can be obtained through the trace of its covariance matrix to calculate. is the decoding order of the base station when receiving user messages. is the power allocation of the user. 、 are the transmission matrix and reflection matrix of the STAR-RIS respectively. is the th user's transmission rate when sending messages through uplink RSMA. is the SINR of the radar target. is the threshold of the radar target SINR; is the maximum available power of the base station; is the th user's allocated power. is the maximum power of each user; 、 are the transmission amplitude and reflection amplitude of the th transmission and reflection element of the STAR-RIS; The set is defined as the set of all possible decoding orders of all messages from users. represents the th user's th sub-message allocated power.

[0064] The calculation formula of

[0065]

[0066]

[0067]

[0068] Among them, is the th user's th sub-message's transmission rate. is the th user's th sub-message's SINR. is the th user's channel with the base station. is the th user's channel with the STAR-RIS. and are the channels between the STAR-RIS, the base station receiving antenna, and the transmitting antenna, respectively, represents the amplitude variance of the radar target, represents the variance of the received noise, is the self-interference of the base station, is the channel between the STAR-RIS and the radar target, and represent the number of transmitting and receiving antennas of the base station, respectively, are the channels between the transmitting antenna, the receiving antenna of the base station, and the radar target, respectively. Among them, the equivalent channel of the radar detection target and the base station equivalent interference are defined as follows:

[0069]

[0070] .

[0071] In addition, the following expressions are defined:

[0072]

[0073]

[0074] limits the minimum SINR of the radar target, is the SINR of the radar target;

[0075] limits the transmission power of the base station, is the maximum transmission power of the base station;

[0076] limits the power of each user, is the maximum power of each user, is the number of users;

[0077] ensures that the transmission and reflection amplitudes of each STAR-RIS element are non-negative, where is the number of elements of the STAR-RIS;

[0078] represents the decoding order with a permutation denoted by, and the permutation belongs to the set , and the set is defined as the set of all possible decoding orders of all messages from users, Indicating the th user's th sub-message allocated power.

[0079] S2, set the iteration number variable and given the initial radar detection signal covariance matrix , initial power allocation , initial decoding order and the transmission matrix of the initial STAR-RIS and reflection matrix ;

[0080] S3, under the premise of given power allocation , radar detection signal covariance matrix , decoding order and the transmission matrix of the STAR-RIS and reflection matrix , optimize the initial model of the integrated communication and sensing system based on STAR-RIS and RSMA using the successive convex approximation (SCA) method and the gradient descent method to obtain the optimal power allocation , and serve as the input for the next iteration;

[0081] The steps to obtain the optimal power allocation are as follows:

[0082] S31, under the premise of given covariance matrix of the variable radar detection signal , decoding order , transmission matrix of the STAR-RIS and reflection matrix , study the optimization of power allocation. At this time, the problem is transformed into:

[0083]

[0084]

[0085]

[0086] Since the SINR expression in the objective function is non-convex, it is difficult to directly obtain the optimal solution of the problem. Here, the successive convex approximation (SCA) method and the gradient descent method are considered to solve the problem. By linearly approximating the non-convex objective function and constraint conditions, the original non-convex problem is gradually approximated, and a convergent solution is obtained. By calculating the gradient of the objective function with respect to the variable and updating the variable in the steepest descent direction, the algorithm can converge to a local optimal solution.

[0087] S32. First, initialize , assume the power allocation at the n-th iteration Perform a first-order Taylor expansion of the objective function at this point:

[0088]

[0089] The first-order approximation with respect to is:

[0090]

[0091] where is the value at the current iteration point, and the gradient of with respect to is:

[0092]

[0093] Through this linear approximation, the original non-convex optimization problem is transformed into a new convex optimization problem:

[0094]

[0095] S33. After transforming it into a convex optimization problem, use the gradient descent method to solve this problem. First, select an initial power allocation scheme such that it satisfies all the constraint conditions. Set the step size and the convergence threshold . At each iteration, calculate the gradient of the objective function:

[0096]

[0097] Substitute the previously calculated partial derivatives into the above formula to obtain the value of the gradient of the objective function. Update the power allocation:

[0098]

[0099] S34. After each update, ensure that the updated solution satisfies the non-negativity constraint and the power constraint, that is . If the updated power is negative, set it to zero. If the total power allocation of a certain user exceeds , then scale its power proportionally so that the total power is exactly equal to . If the norm of the gradient is less than the preset threshold, the algorithm converges.

[0100] S4. Given the power allocation , the covariance matrix of the radar detection signal , the decoding order and the transmission matrix of the STAR-RIS and the reflection matrix Based on the successive convex approximation (SCA) method and the gradient descent method, optimize the initial model of the integrated communication and sensing system based on STAR-RIS and RSMA to obtain the optimal covariance matrix and use it as the input for the next iteration;

[0101] The steps to obtain the optimal covariance matrix are as follows:

[0102] S41. Next, optimize the covariance matrix of the radar target . Fix other variables. At this time, the optimization problem is transformed into:

[0103]

[0104]

[0105]

[0106] The SINR expression in the objective function is still non-convex, and the SINR constraint of the radar target is also non-convex. Therefore, it is difficult to directly find the optimal solution to the problem. Consider using the successive convex approximation (SCA) method and the gradient descent method to solve the problem. By linearly approximating the non-convex objective function and constraint conditions through the first-order Taylor expansion, the problem gradually approaches the original non-convex problem, thereby obtaining a convergent solution. By calculating the gradient of the objective function with respect to the variables and updating the variables in the steepest descent direction, the algorithm can converge to a local optimal solution.

[0107] S42. However, the first-order Taylor expansion cannot ensure the effectiveness of the linear approximation. Therefore, the update amplitude of the optimization variables is restricted by the trust region method, so that the new optimization variables will not deviate too far from the current iteration point. First, initialize the covariance matrix :

[0108]

[0109] Set the trust region parameter: (such as ). Set the trust region adjustment factor: (such as ) (such as ), set the convergence threshold , the step size and the maximum number of iterations.

[0110] S43. Since the expression of is non-convex, both the numerator and denominator parts of the constraint conditions involve , assume that the first-order Taylor expansion of the objective function is performed at the covariance matrix at the n-th iteration:

[0111]

[0112] Regarding , the first-order approximation is:

[0113]

[0114] where is the value of calculated in the -th iteration. Since only the denominator part of the expression of contains , therefore, Regarding , the gradient is:

[0115]

[0116] Through this linear approximation, the original non-convex optimization problem is transformed into a new convex optimization problem:

[0117]

[0118] S44. Next, perform the first-order Taylor expansion on the constraint conditions:

[0119]

[0120] The expression of the gradient is:

[0121]

[0122] S45. The trust region constraint limits 's update amplitude:

[0123]

[0124] where is the trust region radius of the current iteration. This constraint ensures that 's update amplitude will not be too large. Therefore, the complete formulation of this sub-problem is:

[0125]

[0126]

[0127]

[0128]

[0129]

[0130] denote is positive semi - definite. Set the step size . At each iteration, calculate the gradient of the objective function:

[0131]

[0132] S46, Substitute the previously calculated partial derivatives into the above formula to obtain the value of the gradient of the objective function. Update :

[0133]

[0134] S47, Solve the above optimization sub - problem to obtain , Update the objective function and radar SINR and check if the constraints are satisfied:

[0135]

[0136]

[0137] If the objective function is improved and the constraints: and , then update the convergence domain to: and update the iteration point ; Otherwise, if the objective function is not improved or the constraints are not satisfied: or , then keep unchanged, update the convergence domain to: , and re - perform this iteration until the objective function is improved and the constraints are satisfied: ; Iterate a specified number of times or meet the convergence domain condition: , terminate the iteration.

[0138] S5, Given the power allocation radar detection signal covariance matrix decoding order and the transmission matrix of STAR - RIS and reflection matrix on the premise of, based on the greedy algorithm, optimize the initial model of the integrated communication and sensing system based on STAR - RIS and RSMA to obtain the optimal decoding order , and as the input for the next iteration;

[0139] The steps to obtain the optimal decoding order are as follows:

[0140] S51. For all possibilities of the decoding order, they form a discrete set. Therefore, this problem belongs to integer programming and the optimal exhaustive search can be used. Consider using a greedy algorithm to select the message that maximizes the target rate gain for decoding.

[0141] Initialize the set , which contains all possible decoding orders. In the initial state, no message is decoded, and the SINR of all messages is calculated according to the formula in the system model. For each sub-message of each user, calculate its current achievable rate .

[0142] S52. Select the message that maximizes the total rate increment for priority decoding. After selecting this message, add this message to the current decoding order and remove the decoded message from the set of undecoded messages. Repeat the above process until all messages are decoded. When all messages are decoded, a decoding order is obtained .

[0143] The computational complexity of this algorithm is relatively low. At each step, only the message that maximizes the total rate gain among the undecoded messages needs to be selected.

[0144] S6. Given the power allocation radar detection signal covariance matrix , decoding order , and the transmission matrix and reflection matrix of the STAR-RIS, based on the successive convex approximation (SCA) method and the gradient descent method, optimize the initial model of the integrated communication and sensing system based on STAR-RIS and RSMA, and obtain the optimal transmission matrix and reflection matrix in turn, and take and as the input for the next iteration;

[0145] The steps to obtain the optimal transmission matrix and reflection matrix are as follows:

[0146] S61. This paper considers the STAR-RIS operating in the UED mode. The UED mode means that each element of the STAR-RIS can simultaneously reflect and transmit the incident signal, but the reflected and transmitted signals have different amplitudes and phases. Since the reflection and transmission coefficients of each element can be adjusted independently, this mode provides the highest degree of freedom (DoFs).

[0147] The SINR expression in the objective function is still non-convex, and the SINR constraint of the radar target contains variables , so this constraint is also non-convex, and it is difficult to directly obtain the optimal solution of the problem. Consider using the successive convex approximation (SCA) method and the gradient descent method to solve the problem. The non-convex objective function and constraint conditions are linearly approximated by the first-order Taylor expansion to gradually approximate the original non-convex problem, so as to obtain a convergent solution. By calculating the gradient of the objective function with respect to the variables and updating the variables in the steepest descent direction, the algorithm can converge to a local optimal solution. However, the first-order Taylor expansion cannot ensure the effectiveness of the linear approximation. Therefore, the trust region method is used to limit the update amplitude of the optimization variables, so that the new optimization variables will not deviate too far from the current iteration point. Set the trust region parameter: (such as ). Set the trust region adjustment factor: , , set the convergence threshold , step size and the maximum number of iterations.

[0148] First, fix and other variables, and optimize . At this time, the optimization problem is:

[0149]

[0150]

[0151]

[0152] S62. Perform the first-order Taylor expansion on the objective function:

[0153]

[0154] The first-order approximation with respect to is:

[0155]

[0156] S63, where is the value of calculated in the th iteration. The gradient of with respect to

[0157]

[0158] Among them:

[0159]

[0160] Through this linear approximation, the original non-convex optimization problem is transformed into a new convex optimization problem. The first-order Taylor expansion of the constraint conditions is as follows:

[0161]

[0162] Gradient The expression of is:

[0163]

[0164] S64. Through the above linearized objective function and constraint conditions, a convex optimization sub-problem is constructed:

[0165]

[0166]

[0167]

[0168]

[0169] S65. Calculate the gradient of the objective function:

[0170]

[0171] Substitute the previously calculated partial derivatives into the above formula to obtain the value of the gradient of the objective function. Update :

[0172]

[0173] S66. Solve the above optimization sub-problem to obtain , update the objective function and radar SINR, and check whether the constraints are satisfied:

[0174]

[0175]

[0176] S67. If the objective function is improved and the constraints: and , then update the convergence domain to: and update the iteration point ; otherwise, if the objective function is not improved or the constraints are not satisfied: or , then keep unchanged, update the convergence domain to: , and re-perform this iteration until the objective function is improved and the constraints are satisfied: ; iterate a specified number of times or meet the convergence domain conditions: , terminate the iteration to obtain the optimized . Extract the amplitude from :

[0177]

[0178] S68. For the sub-problem of variable , the optimization method fixes and other variables, and optimizes . The optimization problem at this time is:

[0179]

[0180]

[0181]

[0182] Set the trust region parameter: (such as ). Set the trust region adjustment factor: , , set the convergence threshold , step size and the maximum number of iterations.

[0183] S69. Perform the first-order Taylor expansion on the objective function:

[0184]

[0185] The first-order approximation with respect to is:

[0186]

[0187] S610. Among them, is the value of calculated in the th iteration. For the convenience of calculation, approximate the gradient of with respect to as:

[0188]

[0189] Through this linear approximation, transform the original non-convex optimization problem into a new convex optimization problem. Next, perform the first-order Taylor expansion on the constraint conditions:

[0190]

[0191] The expression of the gradient is:

[0192]

[0193] S611. Construct a convex optimization sub - problem through the above linearized objective function and constraint conditions:

[0194]

[0195]

[0196]

[0197]

[0198] S612. Calculate the gradient of the objective function:

[0199]

[0200] Substitute the previously calculated partial derivatives into the above formula to obtain the value of the objective function gradient. Update :

[0201]

[0202] S613. Solve the above optimization sub - problem to obtain , update the objective function and radar SINR, and check whether the constraints are satisfied:

[0203]

[0204]

[0205] S614. If the objective function is improved and the constraints: and , then update the convergence domain to: and update the iteration point ; otherwise, if the objective function is not improved or the constraints are not satisfied: or , then keep unchanged, update the convergence domain to: , and re - perform this iteration until the objective function is improved and the constraints are satisfied: ; Iterate a specified number of times or meet the convergence domain condition: , terminate the iteration to obtain the optimized . Extract the amplitude from :

[0206]

[0207] After each iteration is completed, give the value of the optimized objective function.

[0208] S7, allocate the optimal power based on the BCD method , the optimal covariance matrix , the optimal decoding order and the optimal transmission matrix and the reflection matrix are substituted into S3, and the iteration count variable is updated . Repeat S3 to S6 until the iteration count is reached, the iteration terminates, and the optimal solution for the total user rate is obtained.

[0209] Embodiment 3

[0210] In this embodiment, the solution of the present invention is described in detail with a specific example. The specific example is simulated using MATLAB software. The specific parameters are set as shown in Table 1.

[0211] Table 1 Simulation Parameter Table

[0212]

[0213] As Figure 3 shown, by comparing different numbers of users, it can be observed that the larger the number of users, the greater the total system rate. This is because the system receives more user messages, increasing the total user rate of the system. In addition, the solution proposed in the present invention has a greater total user rate than the NOMA solution and the WP-RSMA solution under different numbers of users. This is because the RSMA solution can better allocate user power. It has better performance compared to the proposed NOMA solution because the RSMA solution provides flexible interference management.

[0214] Embodiment 4

[0215] In this embodiment, a resource optimization system for a communication and sensing integrated system based on STAR-RIS and RSMA is proposed, including:

[0216] Model construction module: construct an initial model of a communication and sensing integrated system based on STAR-RIS and RSMA;

[0217] Parameter initialization module: set the iteration count variable and give the initial covariance matrix of the radar detection signal , the initial power allocation , the initial decoding order and the initial transmission matrix of the STAR-RIS and the reflection matrix ;

[0218] Optimization module: given the power allocation , the covariance matrix of the radar detection signal , the decoding order and the transmission matrix of the STAR-RIS and reflection matrix Optimize the initial model of the integrated communication and sensing system based on STAR-RIS and RSMA to obtain the optimal power allocation 、optimal covariance matrix 、optimal decoding order and optimal transmission matrix and reflection matrix ;

[0219] And an iteration module: Based on the BCD method, substitute the optimal power allocation 、optimal covariance matrix 、optimal decoding order and optimal transmission matrix and reflection matrix into the optimization module, update the iteration count variable, and perform iterations until the iteration count is reached, at which point the iteration terminates and the optimal solution for the total user rate is obtained.

[0220] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present application. It should be understood that each process and / or block in the flowcharts and / or block diagrams, and the combination of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks

[0221] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks

[0222] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks

[0223] The foregoing has shown and described the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and what is described in the above embodiments and the specification is only to illustrate the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements fall within the scope of the present invention claimed.

Claims

1. A resource optimization method for the synaesthesia integrated system based on STAR-RIS and RSMA, characterized in that: The following steps are involved: S1, construct the initial model of synaesthesia integration system based on STAR-RIS and RSMA; S2, set the iteration number variable and give the initial radar detection signal covariance matrix , Initial power allocation , initial decoding order and the initial STAR-RIS transmission matrix and the reflection matrix ; S3, at a given power distribution , radar detection signal covariance matrix , decoding order and the STAR-RIS transmission matrix and the reflection matrix Under the premise of optimizing the initial model of the synaesthesia integration system based on STAR-RIS and RSMA, the optimal power distribution is obtained. , the optimal covariance matrix , optimal decoding order And the optimal transmission matrix and the reflection matrix ; S4, based on the BCD method to allocate the optimal power , the optimal covariance matrix , optimal decoding order And the optimal transmission matrix and the reflection matrix Substitute into S3 and update the iteration number variable, and iterate until the iteration number is reached. The iteration terminates and the optimal solution for the total user rate is obtained.

2. The method for optimizing synaesthesia integrated system resources based on STAR-RIS and RSMA according to claim 1, characterized in that: The synaesthesia integration system based on STAR-RIS and RSMA includes: single antenna users, a full-duplex base station and STAR-RIS, STAR-RIS includes The user transmits messages to the base station through STAR-RIS and direct connection in the uplink RSMA mode, and the base station detects a single radar target through STAR-RIS with a certain transmission power.

3. The method for optimizing synaesthesia integrated system resources based on STAR-RIS and RSMA according to claim 2, characterized in that: The initial model of the synaesthesia integration system based on STAR-RIS and RSMA is: in, is the covariance matrix of the base station radar detection signal, and the total power of the signal is calculated by the trace of its covariance matrix. To calculate, is the decoding order of the base station when receiving user messages, Power allocation for users, , are the transmission matrix and reflection matrix of STAR-RIS, For the The transmission rate when a user sends a message via the uplink RSMA, is the SINR of the radar target, is the threshold of radar target SINR; is the maximum available power of the base station; For the The power allocated to each user, The maximum power for each user; , For STAR-RIS The transmission amplitude and reflection amplitude of each transmission and reflection element; Defined as coming from All users The set of all possible decoding orders of a message; Indicates User's Sub-message Allocated power.

4. The method for optimizing synaesthesia integrated system resources based on STAR-RIS and RSMA according to claim 3, characterized in that: The calculation formula is: in, For the User's The transmission rate of sub-messages, For the User's The SINR of each sub-message is For the The channel between a user and the base station, For the channels between users and STAR-RIS, and are the channels between STAR-RIS and the base station receiving antenna and transmitting antenna, respectively. represents the radar target amplitude variance, represents the received noise variance, is the self-interference of the base station, is the channel between STAR-RIS and the radar target, and Respectively represent the number of base station transmission antennas and receiving antennas, They are the channels of the base station transmission antenna, receiving antenna and radar target respectively; Indicates User's The decoding order of sub-messages is Indicates User's The decoding order of sub-messages is User's After the sub-message; Among them, the equivalent channel of radar detection target , base station equivalent interference They are defined as: 。 5. The method for optimizing synaesthesia integrated system resources based on STAR-RIS and RSMA according to claim 1, characterized in that: To achieve the best power distribution , the optimal covariance matrix , the optimal transmission matrix and the reflection matrix When optimizing the initial model of the synaesthesia integration system based on STAR-RIS and RSMA, the method based on successive convex approximation and gradient descent is used.

6. The method for optimizing synaesthesia integrated system resources based on STAR-RIS and RSMA according to claim 1, characterized in that: To obtain the optimal decoding order When the greedy algorithm is used, the initial model of the synaesthesia integration system based on STAR-RIS and RSMA is optimized.

7. The synaesthesia integrated system resource optimization system based on STAR-RIS and RSMA is characterized by: include: Model building module: construct the initial model of synaesthesia integration system based on STAR-RIS and RSMA; Parameter initialization module: Set the number of iterations and give the initial radar detection signal covariance matrix , Initial power allocation , initial decoding order and the initial STAR-RIS transmission matrix and the reflection matrix ; Optimization module: Given power allocation , radar detection signal covariance matrix , decoding order and the STAR-RIS transmission matrix and the reflection matrix Under the premise of optimizing the initial model of the synaesthesia integration system based on STAR-RIS and RSMA, the optimal power distribution is obtained. , the optimal covariance matrix , optimal decoding order And the optimal transmission matrix and the reflection matrix ; And, iterative module: optimal power allocation based on BCD method , the optimal covariance matrix , optimal decoding order And the optimal transmission matrix and the reflection matrix Substitute into the optimization module and update the iteration number variable, iterate until the iteration number is reached, the iteration terminates, and the optimal solution for the total user rate is obtained.

8. A computer storage medium storing a readable program, characterized in that: When the program is running, the synaesthesia integration system resource optimization method based on STAR-RIS and RSMA described in any one of claims 1 to 6 can be executed.

9. An electronic device, characterized in that: include: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the synaesthesia integration system resource optimization method based on STAR-RIS and RSMA as described in any one of claims 1 to 6.

10. A computer program product comprising computer instructions, characterized in that: The computer instructions instruct the computing device to execute operations corresponding to the synaesthesia integration system resource optimization method based on STAR-RIS and RSMA as described in any one of claims 1 to 6.

Images