STAR-RIS assisted NOMA symbiotic radio system resource allocation method in hybrid communication mode

Abstract

The invention relates to a simultaneous transmission and reflection reconfigurable intelligent surface (STAR-RIS) assisted non-orthogonal multiple access (NOMA) symbiotic radio system resource allocation method in a hybrid communication mode, and belongs to the field of Internet of Things. The method comprises the following steps that an STAR-RIS-assisted NOMA symbiotic radio system in a hybrid communication mode is established, a backscattering device (BD) in an NOMA cluster adopts the hybrid communication mode to carry out information transmission, and the STAR-RIS adopts a time switching protocol to assist a primary system and a secondary system; on the premise that the target throughput of the main system is met, a resource allocation model is constructed by taking maximization of the BD reachable throughput of the secondary system as a target; a block coordinate descent method is adopted, and the resource allocation model with the maximum throughput capacity is converted into a convex optimization problem; and solving the optimization problem by using a convex optimization toolbox to obtain transmission time blocks of three stages, active beam forming of a base station, a reflection coefficient and BD transmitting power of STAR-RIS, a transmission coefficient of STAR-RIS and a reflection coefficient of BD, namely, obtaining a resource allocation scheme. According to the method, better throughput performance can be obtained, and it is proved that the performance of the symbiotic radio system can be remarkably improved through fusion of the hybrid communication mode, the STAR-RIS technology and the NOMA technology.

Description

Technical Field

[0001] The invention belongs to the field of Internet of Things, and relates to a resource allocation method for a non-orthogonal multiple access (NOMA) symbiotic radio system assisted by a simultaneously transmitting and reflecting reconfigurable smart surface (STAR-RIS) in a hybrid communication mode. Background Art

[0002] With the rapid penetration of the Internet of Things in various fields of social production, the number of sensor nodes in the Internet of Things has exploded. However, the scarcity of spectrum resources and the lack of equipment endurance have become key factors restricting the sustainable development of the Internet of Things. In recent years, symbiotic radio, as an emerging communication paradigm, has been considered as one of the core technologies of the future Internet of Things due to its reciprocal sharing characteristics in the spectrum and energy domains and the ability to achieve highly reliable backscattering communication through joint decoding. It is expected to play a more important role in the future communication field.

[0003] However, the transmission performance of the symbiotic radio system is limited by the influence of dual-channel fading. Existing research mainly focuses on improving system performance by introducing RIS technology to assist communication links or optimizing beamforming. Although these methods have improved communication quality to a certain extent, they still fail to fully solve the problems of resource allocation and transmission efficiency, especially in complex communication environments. Summary of the invention

[0004] In view of this, the object of the present invention is to provide a NOMA symbiotic radio system resource allocation method assisted by STAR-RIS in a hybrid communication mode.

[0005] In order to achieve the above object, the present invention provides the following technical solutions:

[0006] A method for allocating resources of a NOMA symbiotic radio system assisted by STAR-RIS in a hybrid communication mode, the method comprising the following steps:

[0007] Step 1: Establish a symbiotic radio system of NOMA assisted by STAR-RIS in hybrid communication mode, where the backscatter device (BD) in the NOMA cluster adopts hybrid communication mode for information transmission, and STAR-RIS uses time switching protocol to assist the primary and secondary systems;

[0008] Step 2: Under the premise of meeting the target throughput of the primary system, a resource allocation model is constructed with the goal of maximizing the achievable throughput of the secondary system BD;

[0009] Step 3: Use the block coordinate descent method to transform the throughput maximization resource allocation model into a convex optimization problem;

[0010] Step 4: Use the convex optimization toolbox to solve the optimization problem and obtain the transmission time blocks of the three stages, the active beamforming of the base station, the reflection coefficient of STAR-RIS and the BD transmission power, the transmission coefficient of STAR-RIS, and the BD reflection coefficient, that is, the resource allocation plan.

[0011] Further, in step 1, the NOMA symbiotic radio system assisted by STAR-RIS in the hybrid communication mode includes a base station configured with N antennas, a single-antenna cooperative receiver (CR), K single-antenna BDs and a STAR-RIS with M units; the coverage area is divided into a reflection area and a transmission area according to the STAR-RIS working mode, the base station and the K BDs are located in the reflection area of ​​the STAR-RIS, and the CR is located in the transmission area. Definition represents the reflection coefficient vector of STAR-RIS, where the subscript r represents the STAR-RIS reflection area, represents the M×1 dimensional complex matrix space, represents a unit complex vector, represents the phase shift of the mth STAR-RIS element, 1≤m≤M, [·] H represents the conjugate transpose, and the corresponding reflection coefficient matrix is ​​Θ r =diag{u r}, diag{·} represents a diagonal matrix; similarly, the transmission coefficient vector and its transmission coefficient matrix of STAR-RIS are and

[0012] Further, in step 2, under the premise of satisfying the target throughput of the primary system, with the goal of maximizing the achievable throughput of the secondary system BD, the resource allocation model is configured as follows:

[0013]

[0014] C6:t 1 +t 2 +t 3 ≤T,t 1 ,t 2 ,t 3 ≥0

[0015] C7:t R =t 1 ,t T =t 2 +t 3

[0016]

[0017] Among them, max means that the objective function takes the maximum value, They represent the total BD throughput and the total base station throughput in the three transmission stages, The superscript of represents the throughput value sum operation, and the subscript corresponds to BD. i={1,2,3}, the superscript i corresponds to the i-th stage of transmission, The subscripts of correspond to the main system base station; t = [t 1 ,t 2 ,t 3 ] is the transmission time block of the three phases, the subscript i corresponds to the three phases of transmission, w is the base station transmit beamforming vector, α is the BD reflection coefficient, P b is the BD transmission power, the subscript corresponds to BD; C1 represents the minimum limit of the throughput that can be achieved by the main system, is the target throughput of the primary system, and the superscript min indicates the minimum value of the base station throughput; C2 indicates the QoS constraint during BD decoding of the secondary system, is the signal-to-interference-plus-noise ratio (SINR) of BD k at stage i, the superscript corresponds to the corresponding transmission stage, and the subscript corresponds to BDk, Indicates BD k decoding c k The SINR threshold, the superscript min indicates the minimum value of the decoded SINR; C3 indicates that the maximum limit of the total transmission power of the base station is The superscript max indicates the maximum value, and the subscript corresponds to the main system base station; C4 requires that the energy collected by BD in the first two stages must be greater than the energy required for BD to achieve active communication in the third stage, P b,k Expressed as the transmission power of BD k, and are the energies collected by BDs in phase 1 and phase 2, respectively. The subscripts correspond to the corresponding BDs and the superscripts correspond to the corresponding transmission phases. C5 represents the non-negative constraint on the transmission power of BDs when performing active communication in phase 3. C6 represents the time constraints of the three phases, where T represents the working time block of the entire system. C7 represents the relationship between the reflection and transmission time allocation factor of STAR-RIS and the time proportion of each of the three phases, where the reflection and transmission time of STAR-RIS are t R and t T ; C8 is the reflection coefficient constraint of BD, α i,k represents the reflection coefficient of BD k at stage i; C9 is the reflection and transmission phase shift constraint for STAR-RIS, and They represent the phase shift of STAR-RIS in the mth reflection unit and transmission unit respectively.

[0018] Furthermore, in step 3, the process of solving the resource allocation model for maximizing throughput includes the following steps:

[0019] a. For the optimization of the transmission time blocks in the three stages, P1 is transformed into the following optimization problem, which can be solved directly.

[0020]

[0021] stC1,C4,C6,C7

[0022] b. For base station active beamforming optimization, the optimization problem of P1 transformation is as follows:

[0023]

[0024] stC1,C2,C3,C4

[0025] P1-2 is still a non-convex optimization problem, in which the objective function is a non-concave function. At the same time, C 1 and C 2 is a non-convex constraint.

[0026] c. For the optimization of STAR-RIS reflection coefficient and BD transmission power, the optimization problem of P1 conversion is as follows:

[0027]

[0028] stC1,C2,C4,C5,C9

[0029] d. For the optimization of the STAR-RIS transmission coefficient matrix, the optimization problem of P1 transformation is as follows:

[0030]

[0031] stC1,C2,C9

[0032] e. For BD reflection coefficient optimization, the optimization problem of P1 transformation is as follows:

[0033]

[0034] stC1,C2,C4,C8

[0035] Furthermore, in step 4, the optimization problem has been transformed into a convex optimization problem. The convex optimization (CVX) tool can be used directly to solve the optimization problem and obtain the time slots between NOMA clusters, the reflection coefficients of the backscattering devices, and the full-duplex transmission power, that is, the resource allocation plan.

[0036] The beneficial effects of the present invention are:

[0037] In the face of the higher requirements of spectrum efficiency, energy efficiency and communication quality of the future Internet of Things, this paper integrates NOMA and STAR-RIS technologies in the symbiotic radio system, and designs a STAR-RIS-assisted NOMA symbiotic radio system resource allocation method in hybrid communication mode. Under the constraint of the target throughput of the main system, a non-convex optimization function constrained by multiple conditions is constructed with the goal of maximizing the total throughput of BDs. Successive convex approximation (SCA), variable substitution and semi-positive relaxation methods are used to transform it into a solvable convex optimization problem; finally, based on the idea of ​​alternating iteration, the three-stage transmission time block, active beamforming of the base station, reflection coefficient of STAR-RIS and BD transmission power, transmission coefficient of STAR-RIS, and joint optimization of BD reflection coefficient are realized. The results show that compared with the other four traditional methods, the method of the present invention can increase the throughput by 14.36%, 43.43%, 67.78% and 439.69% respectively, which proves that the integration of the three technologies can significantly improve the performance of the symbiotic radio system. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 is a flow chart of the resource allocation method of the present invention;

[0039] Figure 2 It is an overall architecture diagram of the NOMA symbiotic radio system assisted by STAR-RIS in the hybrid communication mode of the present invention;

[0040] Figure 3 is a convergence characteristic curve diagram of the BD total throughput of the present invention;

[0041] Figure 4 is a curve diagram of the total throughput of BDs of the present invention and the comparative method under the base station transmission power; DETAILED DESCRIPTION

[0042] See also Figure 1-Figure 4 , which is a resource allocation method for NOMA symbiotic radio system assisted by STAR-RIS in hybrid communication mode.

[0043] The present invention provides a NOMA symbiotic radio system resource allocation method assisted by STAR-RIS in a hybrid communication mode, such as Figure 1 As shown, the method comprises the following steps:

[0044] Step 1: Establish a symbiotic radio system of NOMA assisted by STAR-RIS in hybrid communication mode, where the BD in the NOMA cluster adopts hybrid communication mode for information transmission, and STAR-RIS uses time switching protocol to assist the primary and secondary systems;

[0045] Step 2: Under the premise of meeting the target throughput of the primary system, a resource allocation model is constructed with the goal of maximizing the achievable throughput of the secondary system BDs;

[0046] Step 3: Use the block coordinate descent method to transform the throughput maximization resource allocation model into a convex optimization problem;

[0047] Step 4: Use the convex optimization toolbox to solve the optimization problem and obtain the transmission time blocks of the three stages, the active beamforming of the base station, the reflection coefficient of STAR-RIS and the BD transmission power, the transmission coefficient of STAR-RIS, and the BD reflection coefficient, that is, the resource allocation plan.

[0048] Further, in step 1, if Figure 2 The system includes a base station with N antennas, a single-antenna cooperative receiver (CR), K single-antenna BDs, and a STAR-RIS with M units. The coverage area is divided into a reflection area and a transmission area according to the STAR-RIS working mode. The base station and the K BDs are located in the reflection area of ​​the STAR-RIS, and the CR is located in the transmission area. Definition represents the reflection coefficient vector of STAR-RIS, where represents the phase shift of the mth STAR-RIS element, 1≤m≤M, and the corresponding reflection coefficient matrix is ​​Θ r =diag{u r}; Similarly, the transmission coefficient vector and transmission coefficient matrix of STAR-RIS are and

[0049] The transmission process is divided into three stages. First, in the first two stages when the authorized spectrum of the primary system is busy, BD performs backscattering communication to form a symbiotic relationship with the primary system. STAR-RIS improves BD's energy collection through reflection mode in stage one and assists the primary and secondary systems through transmission mode in stage two. Then, in the third stage when the spectrum is idle, STAR-RIS assists BD's active communication through transmission mode.

[0050] When STAR-RIS works in T mode, the channel from the base station to the CR via STAR-RIS is recorded as The subscript corresponds to the main system, i.e. The backscatter channel from BD k to CR via STAR-RIS is denoted as The subscripts correspond to the subsystems, i.e. When STAR-RIS works in R mode, the channel from the base station to BD k via STAR-RIS is recorded as The subscript corresponds to the main system, i.e. Among them, the channel coefficients from the base station to CR, BD k, and STAR-RIS are Subscript k corresponds to BD k; the channel coefficients from STAR-RIS to BDk and CR are Subscript r corresponds to the reflection area, and t corresponds to the transmission area; the channel coefficients from BD k to CR and STAR-RIS are h c,k , where k∈{1,...,K}.

[0051] Furthermore, in step 2, the process of constructing a resource allocation model for maximizing BDs throughput includes:

[0052] Consider the mutualistic primary and secondary systems, and let the periods of the primary and secondary system symbols be T s and T c , and T c =LT S , satisfying L>>1,>> indicates much greater than. Let the main system symbol be s(l), representing the lth main symbol in a single BD symbol period, and when the base station transmits the beamforming vector as w, the base station equivalently sends the signal as ws(l), l∈{1,...,L}.

[0053] In the first stage, BD performs backscatter communication. In the lth main symbol period within a single symbol period of BD, the BD k backscatter signal can be expressed as Among them, α 1,k ∈[0,1] is the reflection coefficient of BD k in stage 1, c k represents the transmission symbol of BD k. Correspondingly, the energy collected by BD k is:

[0054]

[0055] Among them, η k ∈(0,1) represents the energy harvesting efficiency of BD k.

[0056] The signal received at CR is:

[0057]

[0058] in, It means that CR has a mean of 0 and a variance of Additive white Gaussian noise.

[0059] In CSR, the equivalent channel for decoding s(l) can be expressed as Therefore, for the primary system, the SINR of decoding s(l) is:

[0060]

[0061] Therefore, the achievable throughput of the primary system in the first phase is:

[0062]

[0063] where log 2 (·) represents the logarithmic function with base 2, E[·] represents the expected value, and the variable is c k .

[0064] For L>>1, the achievable throughput of s(l) at CR is approximately:

[0065]

[0066] For the secondary system, CR uses serial interference cancellation (SIC) technology to first decode the primary signal and then remove the successfully decoded primary signal; thus, the intermediate decoded signal corresponding to L primary symbols in the symbol period of the secondary system is It can be expressed as the following formula, where [·] T Indicates transpose.

[0067]

[0068] where s = [s(1), s(1), ..., s(L)] T is the symbol vector of the main system, n c =[n c (1),n c (2),...,n c (L)] T represents the intermediate noise vector corresponding to the SIC decoding process.

[0069] According to the SIC principle, assuming c i In c j was previously decoded (i<j). According to the maximum ratio combining technique, c k Decoding, we can get CR decoding c k The corresponding SINR is:

[0070]

[0071] Therefore, the total throughput of BD in the first phase is:

[0072]

[0073] In the second stage, BD performs backscatter communication. The signal backscattered by BD k can be expressed as Correspondingly, the energy collected by BD k is:

[0074]

[0075] The signal received at CR is:

[0076]

[0077] Which decodes s The equivalent channel of (l) is Therefore, for the primary system, the SINR of decoded s(l) is:

[0078]

[0079] Correspondingly, if L>>1, the achievable throughput of s(l) at CR can be approximately expressed as:

[0080]

[0081] For the secondary system, the SIC technology is also used at the CR to first decode the primary system symbols and then remove the successfully decoded symbols. for:

[0082]

[0083] Similar to the first stage, decoding c k The SINR is:

[0084]

[0085] Therefore, the total BD throughput in the second phase is:

[0086]

[0087] In the third stage, BD exclusively uses the licensed frequency band to transmit information and perform active communication. To this end, the CR receives the signal:

[0088]

[0089] Among them, P b,k is the transmission power of BD k, which must satisfy the following conditions:

[0090]

[0091] Decodec k The SINR is:

[0092]

[0093] Since the decoding order of uplink NOMA does not affect the total throughput of all BDs, the total throughput of BDs in the third-stage subsystem is:

[0094]

[0095] The goal of the present invention is to optimize the transmission time block t, base station active beamforming, STAR-RIS reflection coefficient and BD transmission power P in three stages by joint optimization. b ,STAR-RIS transmission coefficient, and BD reflection coefficient α, under the premise of meeting the target throughput of the primary system, with the goal of maximizing the achievable throughput of the secondary system BDs, the resource optimization problem can be described as:

[0096]

[0097] C6:t 1 +t 2 +t 3 ≤T,t 1 ,t 2 ,t 3 ≥0

[0098] C7:t R =t 1 ,t T =t 2 +t 3

[0099]

[0100] in, C1 represents the minimum limit of the throughput that can be achieved by the primary system; C2 represents the QoS constraint when the secondary system is decoding BD. Indicates BD k decoding c k SNR threshold; C3 indicates that the maximum limit of the total transmission power of the base station is C4 requires that the energy collected by BD in the first two stages must be greater than the energy required for BD to achieve active communication in the third stage; C5 represents the non-negative constraint on the transmission power of BD when performing active communication in stage three; C6 is the time constraint for the three stages; C7 is the relationship between the reflection and transmission time allocation factor of STAR-RIS and the respective time proportions of the three stages; C8 is the reflection coefficient constraint for BD; C9 is the reflection and transmission phase shift constraint for STAR-RIS.

[0101] For P1, we first use the block coordinate descent method to transform it into the following five sub-optimization problems:

[0102] a. Optimization of transmission time blocks in three stages,P1 is transformed into the following optimization problem, which can be solved directly.

[0103]

[0104] stC1,C4,C6,C7

[0105] b. Base station active beamforming optimization, the optimization problem of P1 transformation is as follows:

[0106]

[0107] stC1,C2,C3,C4

[0108] For w, the objective function is non-convex. First, the objective function is converted into a form that is easy to handle and the base station active beamforming matrix is ​​defined as Satisfies W>0 and Rank(W)=1, where Rank(·) represents the rank of the matrix. To facilitate subsequent analysis, define Further definition The subscript k corresponds to BD k. Since only the first and second stages in the objective function involve the active beamforming vector w, the objective function is rewritten as:

[0109]

[0110] Where Tr(·) represents the trace of the matrix. The objective function is still non-convex. For ease of processing, the SCA method is used. By introducing the slack variable l = [l i ]and i=1,2, i corresponds to transmission stage i. So we have

[0111]

[0112]

[0113] Where exp(·) represents the exponential function.

[0114] Using the first-order Taylor approximation, that is:

[0115] Further:

[0116]

[0117] Therefore, the objective function is reformulated as:

[0118]

[0119] Furthermore, for C1, the throughput of the main system can be re-expressed as:

[0120]

[0121] in, The superscript indicates the conjugate transpose, and the subscript is mainly used to distinguish.

[0122] For C4, let Then the sum of the energy collected by BD in stages one and two is:

[0123]

[0124] Through the above conversion, P1-2 can be transformed into:

[0125]

[0126] C10:Rank(W)=1

[0127] C11:C11-1,C11-2,C11-3,C11-4

[0128] Among them, C10 is still a non-convex constraint condition; after removing the rank 1 constraint in P1-2′, the problem can be relaxed into a semi-positive definite programming problem. If the optimal solution W after SDP relaxation is * The rank is 1, which can be solved by singular value decomposition W*=w*(w*) H Solve for w, otherwise the suboptimal solution of P1-2′ is obtained by Gaussian randomization method, and the superscript represents the optimal value.

[0129] c. The optimization of STAR-RIS reflection coefficient and BD transmission power, the optimization problem of P1 conversion is as follows:

[0130]

[0131] stC1,C2,C4,C5,C9

[0132] Among them, the objective function and C1 are non-convex. To facilitate subsequent processing, according to the reflection coefficient vector satisfy

[0133] For the objective function, Use the trace operation of the matrix and the inner product operation of the vector to convert it into a matrix form Tr(Q brk1 U r ),in And satisfy U r ≥0 and Rank(U r )=1, where r in the subscript represents the reflection area, and b corresponds to BD in the subsystem.

[0134] Since the STAR-RIS reflection coefficient and BD transmission power optimization are not involved in the second stage, the objective function can be rewritten as follows:

[0135]

[0136] The objective function is still non-convex, and the SCA method is also used. By introducing the slack variable q1 =[q 1,1 ,...,q 1,K-1 ] and φ 1 =[φ 1,1 ,...,φ 1,K-1 ], the subscript corresponds to BD k, so for k, k∈{1,...,K-1}:

[0137]

[0138] Similarly, using the first-order Taylor approximation, we have:

[0139]

[0140] In summary, the objective function can be rewritten as:

[0141]

[0142] Then, for C1, a similar treatment as in the objective function can be used. First, The first item Expressed as in Thus, the first term can be re-expressed as Tr(Q cr1 U r ),in The second item It can be re-expressed as Tr(Q cr2 U r ),in Therefore, the throughput of the primary system is:

[0143]

[0144] Similarly, C4 can be rewritten as:

[0145]

[0146] According to the above conversion, P1-3 becomes P1-3' as shown.

[0147]

[0148] C13:Rank(U r )=1

[0149] C14:C14-1,C14-2

[0150] Among them, C13 is non-convex and can be relaxed using SDR technology. After removing C13, P1-3′ can be solved using the convex optimization tool CVX. Note that the solution may not satisfy the rank-one constraint. In this case, The method can be similar to solving w in P1-2′ * .

[0151] d. STAR-RIS transmission coefficient matrix optimization, the optimization problem of P1 transformation is as follows:

[0152]

[0153] stC1,C2,C9

[0154] Among them, the objective function and C1 are non-convex. Similar to the optimization of the reflection coefficient vector, the transmission coefficient vector is defined as Then there is Therefore, the objective function It can be rewritten as Tr(Q btk1 U t ),in Meet U t ≥0 and Rank(U t )=1; in The t in the subscript represents the transmission area, and b corresponds to the BD in the subsystem. In addition, since the transmission coefficient vector of STAR-RIS is not involved in stage 1, the objective function can be rewritten as follows.

[0155]

[0156] However, the above equation is still non-convex, so the SCA method is used. Introducing the slack variable q 2 =[q 2,1 ,...,q 2,K-1 ] and φ 2 =[φ 2,1 ,...,φ 2,K-1 ], for any k∈{1,...,K-1}:

[0157]

[0158] Similarly, using the first-order Taylor approximation, we get:

[0159]

[0160] The objective function can be reformulated as:

[0161]

[0162] The third and fourth items Can be expressed as Tr(Q ct1 U t ), Tr(Q ct2 U t ),in

[0163] where subscript c corresponds to the primary system. Therefore, the throughput of the primary system can be re-expressed as:

[0164]

[0165] According to the above conversion, P1-4 becomes P1-4'.

[0166]

[0167] C16:Rank(U t )=1

[0168] C17:C17-1,C17-2

[0169] Among them, in P1-4′, C16 is non-convex and can also be relaxed using SDR technology to solve The method is similar to solving w in P1-2′ * ; After removing C16, P1-4′ can be solved by the convex optimization tool CVX.

[0170] e. BD reflection coefficient optimization, the optimization problem of P1 transformation is as follows:

[0171]

[0172] stC1,C2,C4,C8

[0173] Among them, the objective function and C1 are non-convex; the SCA method is also used to deal with this problem.

[0174] For the objective function, introduce the slack variable u=[u i ] and ψ=[ψ i ], where u i =[u i,1 ,...,u i,K-1 ],ψ i =[ψ i,1 ,...,ψ i,K-1 ], i = 1, 2. So, for any k∈{1,...,K-1}:

[0175]

[0176] Similarly, using the first-order Taylor approximation, we get:

[0177]

[0178] in, Indicates the mth iteration when the SCA method is used 1,k feasible point.

[0179] In the objective function, since only stages one and two involve variables, the objective function is rewritten as:

[0180]

[0181] For the non-convex constraint C1, use The lower bound of the main system throughput is approximated by for:

[0182]

[0183] in, The superscript represents the lower bound of the main system throughput.

[0184] According to the above conversion, P1-5 becomes P1-5'.

[0185]

[0186] C2,C4,C8

[0187] C18:C18-1,C18-2,C18-3,C18-4

[0188] This problem is a convex optimization problem and can be solved efficiently using CVX tools.

[0189] Example

[0190] This example describes the application effect of the present invention in detail in combination with simulation.

[0191] This embodiment verifies the convergence and effectiveness of the method through simulation results. Considering the three-dimensional space coordinate model, the base station, CR, STAR-RIS and BD1, 2 are located at (0,0,10), (0,50,0), (0,10,10), (0,1,0), (0,-1,0) (unit: meter). The path loss index from the base station to CR, the base station to BD and the BD to CR are set to 3.8, 3 and 2.8 respectively, and the other path loss index is 2. Other parameters are set as follows: K = 2, L = 128, T = 1, N=6, M=40, η=0.8, ε 0 =-30dB, κ=3dB.

[0192] In this embodiment, the proposed STAR-RIS-assisted NOMA symbiotic radio system resource allocation method in hybrid communication mode is transformed into Figure 3It can be seen that the total throughput of BD of this method converges. In this figure, the total throughput of BD tends to stabilize after about 2 iterations, which shows that after 2 iterations of this iterative algorithm, BD can obtain a stable throughput; at the same time, as the number of base station antennas or the number of STAR-RIS units increases, the total throughput of BD shows an upward trend.

[0193] Compared with traditional methods, Figure 4 The performance of the method of the present invention and four traditional methods in terms of BD throughput increase trend and difference is compared. The method of the present invention is always better than other traditional methods at all power levels. Compared with traditional method one, the method of the present invention reasonably allocates the time of stage one, so that the BD can collect more energy, thereby increasing the duration of stage three, and thus improving the throughput of the subsystem; compared with traditional method two, since multiple BDs use the NOMA scheme to achieve active communication in stage three, the throughput of the subsystem is improved; compared with traditional method three, the phase shift of STAR-RIS is optimized to improve the BD communication environment, thereby significantly improving the throughput of the subsystem; and since traditional method four does not have the active communication stage of stage three, the total throughput of the BD is relatively small, and its trend with power change is not obvious.

[0194] Finally, it should be noted that the above preferred embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail through the above preferred implementation cases, those skilled in the art should understand that various changes can be made in form and details without departing from the scope defined by the claims of the present invention.

Claims

1. A resource allocation method for non-orthogonal multiple access (NOMA) symbiotic radio system assisted by simultaneously transmissive and reflective reconfigurable smart surface (STAR-RIS) in hybrid communication mode, characterized in that: The following steps are involved: Step 1: Establish a symbiotic radio system of NOMA assisted by STAR-RIS in hybrid communication mode, where the backscatter device (BD) in the NOMA cluster adopts hybrid communication mode for information transmission, and STAR-RIS uses time switching protocol to assist the primary and secondary systems; Step 2: Under the premise of meeting the target throughput of the primary system, a resource allocation model is constructed with the goal of maximizing the achievable throughput of the secondary system BD; Step 3: Use the block coordinate descent method to transform the throughput maximization resource allocation model into a convex optimization problem; Step 4: Use the convex optimization toolbox to solve the optimization problem and obtain the transmission time blocks of the three stages, the active beamforming of the base station, the reflection coefficient of STAR-RIS and the BD transmission power, the transmission coefficient of STAR-RIS, and the BD reflection coefficient, that is, the resource allocation plan.

2. The method according to claim 1, characterized in that: In step 1, the NOMA symbiotic radio system assisted by STAR-RIS in hybrid communication mode includes a base station configured with N antennas, a single-antenna cooperative receiver (CR), K single-antenna BDs, and a STAR-RIS with M units; According to the STAR-RIS working mode, the coverage area is divided into a reflection area and a transmission area. The base station and K BDs are located in the reflection area of ​​STAR-RIS, and the CR is located in the transmission area. represents the reflection coefficient vector of STAR-RIS, where the subscript r represents the STAR-RIS reflection area, represents the M×1 dimensional complex matrix space, represents a unit complex vector, represents the phase shift of the mth STAR-RIS element, 1≤m≤M, [·] H represents the conjugate transpose, and the corresponding reflection coefficient matrix is ​​Θ r =diag{u r }, diag{·} represents a diagonal matrix; similarly, the transmission coefficient vector and its transmission coefficient matrix of STAR-RIS are and 3. The method according to claim 1, characterized in that: In step 2, under the premise of meeting the target throughput of the primary system, the resource allocation model is constructed with the goal of maximizing the achievable throughput of the secondary system BD as the goal, and is configured as follows: Among them, max means that the objective function takes the maximum value, They represent the total BD throughput and the total base station throughput in the three transmission stages, The superscript of represents the throughput value sum operation, and the subscript corresponds to BD. i={1,2,3}, the superscript i corresponds to the i-th stage of transmission, The subscripts of correspond to the main system base station; t = [t1, t2, t3] is the transmission time block of the three phases, the subscript i corresponds to the three phases of transmission, w is the base station transmit beamforming vector, α is the BD reflection coefficient, P b is the BD transmission power, the subscript corresponds to BD; C1 represents the minimum limit of the throughput that can be achieved by the main system, is the target throughput of the primary system, and the superscript min indicates the minimum value of the base station throughput; C2 indicates the QoS constraint during BD decoding of the secondary system, is the signal-to-interference-plus-noise ratio (SINR) of BD k in stage i, the superscript corresponds to the corresponding transmission stage, and the subscript corresponds to BD k, Indicates BD k decoding c k The SINR threshold, the superscript min indicates the minimum value of the decoded SINR; C3 indicates that the maximum limit of the total transmission power of the base station is The superscript max indicates the maximum value, and the subscript corresponds to the main system base station; C4 requires that the energy collected by BD in the first two stages must be greater than the energy required for BD to achieve active communication in the third stage, P b,k Expressed as the transmission power of BD k, and are the energies collected by BDs in phase 1 and phase 2, respectively. The subscripts correspond to the corresponding BDs and the superscripts correspond to the corresponding transmission phases. C5 represents the non-negative constraint on the transmission power of BDs when performing active communication in phase 3. C6 represents the time constraint of the three phases, where T represents the working time block of the entire system. C7 represents the relationship between the reflection and transmission time allocation factor of STAR-RIS and the time proportion of each of the three phases, where the reflection and transmission time of STAR-RIS are t R and t T ; C8 is the reflection coefficient constraint of BD, α i,k represents the reflection coefficient of BD k at stage i; C9 is the reflection and transmission phase shift constraint for STAR-RIS, and They represent the phase shift of STAR-RIS in the mth reflection unit and transmission unit respectively.

4. The method according to claim 1, characterized in that: In step 3, the process of solving the resource allocation model for maximizing throughput includes the following steps: a. Optimization of transmission time blocks in three stages,P1 is transformed into the following optimization problem, which can be solved directly. b. For the optimization of active beamforming of base stations, P1 is transformed into the following optimization problem: For w, the objective function is non-convex. First, the objective function is converted into a form that is easy to handle and the base station active beamforming matrix is ​​defined as Satisfy W>0 and Rank(W)=1, where Rank(·) represents the rank of the matrix. The successive convex approximation (SCA) method is used to introduce a slack variable l=[l i ]and i=1,2, the subscript represents the stage i, and the first-order Taylor approximation is used to complete the processing of the objective function and relax the problem into a semi-positive definite programming problem. c. For the optimization of STAR-RIS reflection coefficient and BD transmission power, P1 is transformed into the following optimization problem: Among them, the objective function and C1 are non-convex. To facilitate subsequent processing, according to the reflection coefficient vector satisfy For the objective function, Converted into matrix form Tr(Q brk1 U r ), where Tr(·) represents the trace of the matrix, |·| 2 represents taking the square of the complex modulus, And satisfy U r ≥0 and Rank(U r )=1. Then, the SCA method is used to complete the processing of the objective function. For C1, similar processing in the objective function can be used. The first term can be re-expressed as Tr(Q cr1 U r ),in The second term can be re-expressed as Tr(Q cr2 U r ),in The subscript r represents the reflection area, and b corresponds to the BD in the subsystem. The problem is relaxed into a semi-definite programming problem. d. For the optimization of the STAR-RIS transmission coefficient matrix, P1 is transformed into the following optimization problem: Among them, the objective function and C1 are non-convex. Similar to the optimization of the reflection coefficient vector, the transmission coefficient vector is defined as Then there is In the objective function It can be rewritten as Tr(Q btk1 U t ),in Meet U t ≥0 and Rank(U t )=1; in The t in the subscript represents the transmission area, and b corresponds to the BD in the subsystem. Then, the SCA method is used to relax the problem into a semi-positive definite programming problem. e. For BD reflection coefficient optimization, P1 is transformed into the following optimization problem: Among them, the objective function and C1 are non-convex; the SCA method is also used to deal with this problem. Introduce the slack variable u=[u i ] and ψ=[ψ i ], where u i =[u i,1 ,...,u i,K-1 ],ψ i =[ψ i,1 ,...,ψ i,K-1 ], i = 1, 2, the subscript represents the stage i, and the first-order Taylor approximation is used to relax the problem into a semi-positive definite programming problem.

5. The method according to claim 1, characterized in that: In step 4, the optimization problem has been transformed into a convex optimization problem. The convex optimization (CVX) tool can be used directly to solve the optimization problem and obtain the transmission time blocks of the three stages, the active beamforming of the base station, the reflection coefficient of STAR-RIS and the BD transmission power, the transmission coefficient of STAR-RIS, and the BD reflection coefficient, that is, the resource allocation plan.

Images