STAR-RIS assisted SWIPT system energy efficiency optimization method

Abstract

The invention discloses an STAR-RIS assisted SWIPT system energy efficiency optimization method, and the method is an energy efficiency optimization method which considers a rate division multiple access (RSMA) scheme under a wireless co-energy transmission (SWIPT) system assisted by an intelligent metasurface (STAR-RIS) capable of simultaneous reflection and transmission. According to the method provided by the invention, an optimization problem taking maximization of system energy efficiency as an optimization target is constructed, and the optimization problem is simplified by adopting a precoding scheme suitable for an RSMA scene; the simplified optimization problem is decomposed into three sub-problems: a sub-problem 1 for solving an STAR-RIS side amplitude coefficient, a sub-problem 2 for solving an STAR-RIS side phase coefficient and a sub-problem 3 for solving a base station side rate segmentation proportion and a user side energy recovery proportion; and solving the three sub-problems through alternate iteration until the optimization problem is converged to solve the optimization problem.

Description

Technical Field

[0001] The invention belongs to the fields of wireless communication, STAR-RIS, SWIPT and RSMA, and in particular relates to a STAR-RIS-assisted SWIPT system energy efficiency optimization method. Background Art

[0002] In recent years, the Reconfigurable Intelligent Surface (RIS) technology has been widely considered as one of the candidate technologies for 6G due to its low power consumption and ability to flexibly adjust the wireless communication environment [1]. Its principle is to adjust the propagation direction, phase and amplitude of electromagnetic waves by controlling the reflective elements on the surface, thereby achieving dynamic adjustment of the signal. Smart reflective surfaces can be used in various communication and radar systems, providing a new performance optimization method for wireless communications, the Internet of Things, radar and other fields.

[0003] However, in current research results, smart metasurfaces can only reflect incident signals, which means that RIS cannot be used to assist communication when the receiver is located on the opposite side of the transmitter [2]. To address this problem, [3] proposed a hybrid transmission metasurface (Simultaneously Transmitting and Reflecting RIS, STAR-RIS), which can simultaneously reflect and transmit incident signals and achieve 360° communication coverage.

[0004] STAR-RIS has three transmission protocols: Time Switching (TS), Energy Splitting (ES), and Mode Switching (MS) [4]. The TS transmission protocol means that the working time of STAR-RIS is divided into T time and R time. During T time, all working elements can only transmit the incident signal, and during R time, all working elements can only reflect the incident signal. Under the ES transmission protocol, each working unit can reflect and transmit the incident signal at the same time. By adjusting the transmission-reflection ratio coefficient of the working unit, the energy of the transmission signal and the reflected signal can be flexibly allocated, which provides additional degrees of freedom for the signal propagation process and increases the flexibility of communication scheme design. Under the MS transmission protocol, each working unit of STAR-RIS can independently choose to reflect or transmit. In actual applications, these three transmission protocols can be flexibly used according to the scenario to obtain better comprehensive performance.

[0005] As metasurface technology continues to develop, multiple access technology is also continuously innovating. Orthogonal multiple access (OMA) includes time division multiple access (TDMA), frequency division multiple access (FDMA), and code division multiple access (CDMA). They all eliminate interference between multiple users by allocating orthogonal resources to different users [5]. However, as the capacity of communication networks increases and more and more devices participate in communication, the spectrum efficiency of OMA cannot be guaranteed. In order to further improve the spectrum efficiency, space division multiple access (SDMA) and non-orthogonal multiple access (NOMA) [6] have been widely studied. By properly utilizing the spatial resources and multi-antenna structure at the transmitter, SDMA can provide services to multiple users in the same time-frequency resources, and each user can directly decode the expected data stream by treating the interference between multiple users as noise. NOMA, on the other hand, adjusts the transmission power levels of different users. By allocating different powers to different users, the system can utilize the channel differences between users, making it easier to distinguish the user signals at the receiving end and directly decode the interference between multiple users [7].

[0006] However, SDMA and NOMA also have their limitations. SDMA is suitable for situations where the interference between multiple users is weak. When the interference between multiple users is too severe, the performance of SDMA will be greatly reduced. Since NOMA directly decodes the interference between multiple users, its receiver is relatively complex and is suitable for situations where the interference between multiple users is severe. Recently, a new multiple access technology: rate division multiple access (RSMA) has received widespread attention [8]. Its principle is to divide user messages into public and private parts, decode the public part at the receiving end, and treat the private parts of other users as noise. By flexibly adjusting the ratio of the public and private parts at the transmitting end, flexible management of any level of multi-user interference can be achieved, which is in sharp contrast to the extreme interference management strategies used in SDMA and NOMA.

[0007] Currently, research on STAR-RIS mainly focuses on its combination with NOMA technology, while there are relatively few studies on the combination of RSMA and STAR-RIS. The performance analysis of the combination of these two technologies in various communication scenarios is still an open question.

[0008] [1]Q.Wu and R.Zhang,“Intelligent reflecting surface enhanced wirelessnetwork via joint active and passive beamforming,”IEEE Trans.WirelessCommun.,vol.18,no.11,pp.5394–5409,Nov.2019.

[0009] [2]L.Yang,Y.Yang,M.O.Hasna and M.-S.Alouini,"Coverage,Probability ofSNR Gain,and DOR Analysis of RIS-Aided Communication Systems,"in IEEEWireless Communications Letters,vol.9,no.8,pp.1268-1272,Aug.2020,doi:10.1109 / LWC.2020.2987798.

[0010] [3]Y.Liu,X.Mu,J.Xu,R.Schober,Y.Hao,H.V.Poor,and L.Hanzo,“STAR:Simultaneous transmission and reflection for 360°coverage by intelligentsurfaces,”IEEE Wireless Communications,vol.28,no.6,pp.102–109,Dec,2021.

[0011] [4]X.Mu,Y.Liu,L.Guo,J.Lin and R.Schober,"Simultaneously Transmittingand Reflecting(STAR)RIS Aided Wireless Communications,"in IEEE Transactionson Wireless Communications,vol.21,no.5,pp.3083-3098,May 2022

[0012] [5] Y.Mao and B.Clerckx, Multiple Access Techniques. Cham, Switzerland: Springer Int., 2021, pp.63–100.

[0013] [6] L.Dai, B.Wang, Y.Yuan, S.Han, I.Chih-lin, and Z.Wang, "Nonorthogonalmultiple access for 5G: Solutions, challenges, opportunities, and future researchtrends," IEEE Commun.Mag., vol.53, no.9, pp.74–81, Sep.2015.

[0014] [7] Y. Saito, Y. Kishiyama, A. Benjebbour, T. Nakamura, A. Li, and K. Higuchi, "Non-orthogonal multiple access (NOMA) for cellular future radio access," in Proc. IEEE 77th Veh. Technol. Conf. (VTC Spring), 2013, pp. 1–5.

[0015] [8] Y.Mao, B.Clerckx, and VOKLi, "Rate-splitting multiple access for downlink communication systems: Bridging, generalizing, and outperforming SDMAand NOMA," EURASIP J.Wireless Commun.Netw., no.1, p.133, May 2018. Summary of the invention

[0016] The purpose of the present invention is to solve the problems existing in the prior art and provide a STAR-RIS-assisted SWIPT system energy efficiency optimization method.

[0017] In order to achieve the above-mentioned invention object, the present invention specifically adopts the following technical solutions:

[0018] A STAR-RIS-assisted SWIPT system energy efficiency optimization method comprises the following steps:

[0019] S1: Based on the constructed system model of the STAR-RIS-assisted SWIPT system, the rate-splitting multiple access scheme is considered on the base station side, and an optimization problem with the optimization goal of maximizing the system energy efficiency is constructed. Under the consideration of the communication rate requirement constraints of the receiving side users, the energy collection requirement constraints, and the energy limit constraints on the base station side, the active beams on the base station side, the rate splitting ratio, and the passive beams on the STAR-RIS side are jointly optimized to maximize the system energy efficiency.

[0020] S2: The precoding design suitable for the base station side of the RSMA-SWIPT system is realized through the maximum ratio transmission-zero forcing precoding scheme;

[0021] S3: The optimization problem in S1 is decomposed into three sub-problems, namely, sub-problem 1 for solving the amplitude coefficient on the STAR-RIS side, sub-problem 2 for solving the phase coefficient on the STAR-RIS side, and sub-problem 3 for solving the rate split ratio on the base station side and the energy recovery ratio on the user side. The three sub-problems are solved by alternate iteration until the optimization problem converges, and the solution of the optimization problem is obtained. The values ​​of each optimization variable when the optimization problem converges are used as the corresponding design for achieving the optimal energy efficiency of the SWIPT system.

[0022] Based on the above scheme, each step can be implemented in the following preferred specific manner.

[0023] Preferably, in step S1, the system model constructed is as follows: Consider a two-user STAR-RIS-assisted SWIPT system, the base station has M antennas, the STAR-RIS has N reflection units and N transmission units, wherein a single-antenna user user 1 is in the reflection area of ​​the STAR-RIS, and another single-antenna user user 2 is in the transmission area of ​​the STAR-RIS. The STAR-RIS adopts the ES transmission mode. The single-antenna user use r 1 receives the signal reflected by STAR-RIS, and the single-antenna user user 2 receives the signal transmitted by STAR-RIS. The information sent by the base station side to the single-antenna user user 1 is divided into a first public information part and a first private information part, and the information sent to the single-antenna user user 2 is divided into a second public information part and a second private information part. The first public information part and the second public information part are encoded into a public stream, the first private information part is encoded into a first private stream, and the second private information part is encoded into a second private stream. The superimposed signal obtained after precoding the public stream, the first private stream and the second private stream is used as the signal finally sent by the base station.

[0024] Preferably, in step S1, the function form of the optimization problem is as follows:

[0025]

[0026] c i +R p，i ≥λ i (1b)

[0027] P i EH ≥ζ i (1c)

[0028]

[0029] θ i，n ∈[0,2π] (1e)

[0030] β 1，n 2 +β 2，n 2 =1 (1f)

[0031] Among them, (1a) is the constraint to ensure that each user can successfully decode the public stream, (1b) is the communication rate requirement constraint of user i, (1c) is the energy collection requirement constraint of user i, (1d) is the energy limit constraint on the base station side, (1e) is the phase constraint of the STAR-RIS unit, and (1f) is the amplitude coefficient constraint when STAR-RIS works in ES mode;

[0032] Where: c i represents the public information rate of user i; w k Represents the precoding vector of the public or private stream; α i represents the energy split ratio at user i; θ i，n represents the phase coefficient of the nth unit of STAR-RIS located on the user i side; β i，n represents the amplitude coefficient of the reflection or transmission side of the nth unit of STAR-RIS located on the user i side; R p，i R represents the information rate of the private stream decoded by user i; c，i represents the information rate of the public stream decoded by user i; i represents the minimum communication rate requirement of user i; P i EH Represents the power used for energy recovery; i represents the minimum energy collection requirement of user i; P max It is the maximum transmission power that the base station can provide; β 1，n represents the amplitude coefficient of the reflection side of the nth unit of STAR-RIS located on the user 1 side; β 2，nRepresents the transmission amplitude coefficient of the nth unit of STAR-RIS located on the user 2 side.

[0033] Preferably, in step S2, the maximum ratio transmission-zero forcing precoding scheme is specifically:

[0034]

[0035] q i =h i H Φ i G

[0036] Where: H represents conjugate transpose; q i represents an intermediate variable; z k Represents the intermediate variable Z = [z 1 , z 2 ]; Z = Q H (QQ H ) represents an intermediate variable; Q = [q 1 ,q 2 ] represents an intermediate variable; q 1 ,q 2 Both represent intermediate variables; h i is the channel parameter from STAR-RIS to useri; Φ i is the coefficient matrix of STAR-RIS on the user i side; G is the channel parameter from the base station to STAR-RIS.

[0037] Preferably, in step S3, the specific process of obtaining the solution result of the optimization problem is: in the tth iteration round, determine whether the optimization problem converges: if it converges, use the solution results of the three sub-problems when the optimization problem converges as the solution result of the optimization problem; if it does not converge, solve the three sub-problems separately to obtain the solution results of the three sub-problems in the tth iteration round, and enter the t+1th iteration round until the optimization problem converges.

[0038] Preferably, in step S3, the function form of sub-problem 1 is:

[0039] P1: β T Pβ+q T β(3)

[0040]

[0041]

[0042] q=[-A 1 , -A 2 ] T

[0043] β=[β1,β2] T

[0044] Where: β represents the STAR-RIS amplitude coefficient value vector; T represents transposition; β i represents the amplitude coefficient of the STAR-RIS working unit close to user i; β 1 represents the amplitude coefficient of the reflection side of any unit of STAR-RIS located on the user 1 side; β 2 represents the amplitude coefficient of the transmission side of any unit of STAR-RIS located on the user 2 side; P,t ik t i0 t ii , B 1 , B 2 , A 1 , A 2 Both represent intermediate variables; σ 2 represents the variance of complex additive Gaussian white noise; γ i Represents auxiliary variables.

[0045] Preferably, in step S3, the function form of sub-problem 2 is:

[0046]

[0047] st|θ i，n |≤1 (4f)

[0048]

[0049] Where: θ i represents the phase coefficient vector of the STAR-RIS working unit located on the user i side; C represents a positive number; a i , b i Represent auxiliary variables The index of g ik , C i，k , g i0 , C ii , C i，0 , D i , g ii All represent intermediate variables; Respectively represent the θ in the t-1th iteration round i，n , Θ i , a i , b i The value of; N represents the number of STAR-RIS working units.

[0050] Preferably, in step S3, the function form of sub-problem 3 is:

[0051]

[0052] Where: u i，0 ,u i,k ,u i,i , d i , e i , f i All represent intermediate variables; Respectively represent the intermediate variable d in the t-1th iteration round i , e i , f i The value of .

[0053] Compared with the prior art, the present invention has the following beneficial effects:

[0054] (1) Better energy efficiency performance: The present invention adopts the RSMA multiple access scheme in the STAR-RIS-assisted SWIPT system, and its energy efficiency performance is better than the existing multiple access schemes using NOMA and SDMA.

[0055] (2) Green communication network: The communication scheme of the present invention takes wireless energy into consideration. The energy collection process at the receiving end can reduce the number of battery replacements of the receiving end IOT device, which has certain research value for the construction and evaluation of future 6G green communication networks. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 is a flow chart of the steps of the method of the present invention;

[0057] Figure 2 Schematic diagram of the system model of the method of the present invention;

[0058] Figure 3 A schematic diagram of a specific implementation process of the method of the present invention for solving optimization problems;

[0059] Figure 4 This is a graph showing the convergence performance results of algorithms for different transmission schemes according to an embodiment of the present invention;

[0060] Figure 5 This is a diagram showing how the energy efficiency of different transmission schemes varies with the number of antennas at the base station side according to an embodiment of the present invention;

[0061] Figure 6 This is a graph showing how the energy efficiency of different transmission schemes varies with the number of working units on the STAR-RIS side according to an embodiment of the present invention. DETAILED DESCRIPTION

[0062] In order to make the above-mentioned purpose, features and advantages of the present invention more obvious and easy to understand, the specific implementation mode of the present invention is described in detail below in conjunction with the accompanying drawings. In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below. The technical features in each embodiment of the present invention can be combined accordingly without conflicting with each other.

[0063] In the description of the present invention, it should be understood that the terms "first" and "second" are only used for the purpose of distinguishing descriptions, and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features.

[0064] like Figure 1 As shown, in a preferred implementation of the present invention, the above-mentioned STAR-RIS-assisted SWIPT system energy efficiency optimization method includes the following steps S1 to S4. The specific implementation process is described in detail below.

[0065] 1. Building a system model

[0066] S1: Construct a system model of the STAR-RIS-assisted SWIPT system.

[0067] It should be noted that in step S1, the system model constructed is as follows: Consider a two-user STAR-RIS-assisted SWIPT system, the base station has M antennas, STAR-RIS has N reflection units and N transmission units, wherein a single-antenna user user 1 is in the reflection area of ​​STAR-RIS, and another single-antenna user user2 is in the transmission area of ​​STAR-RIS. STAR-RIS adopts ES transmission mode, and the single-antenna user user 1 receives the signal reflected from STAR-RIS, and the single-antenna user user2 receives the signal transmitted from STAR-RIS, and the information W sent by the base station side to the single-antenna user user1 is 1 Divided into the first public information part W c，1 and the first private information part W p，1 , the information W sent to the single-antenna user user2 2 Divided into the second public information part W c,2 and the second private information part W p，2 , the first public information part W c,1 and the second public information part W c,2 Encode into public streams 0, the first private information part W p，1 Encoded into the first private stream s 1 , the second private information part W p，2 Encoded into the second private stream s 2 , a superimposed signal x obtained by precoding the public stream, the first private stream and the second private stream is used as the signal finally sent by the base station.

[0068] In this embodiment, if Figure 2 As shown, a two-user STAR-RIS-assisted SWIPT system is considered, the base station has M antennas, STAR-RIS has N reflection units and N transmission units, wherein a single-antenna user user 1 is in the reflection area of ​​STAR-RIS, and another single-antenna user user 2 is in the transmission area of ​​STAR-RIS. STAR-RIS adopts the Energy Splitting (ES) transmission mode. In the ES mode, STAR-RIS can simultaneously reflect and transmit the incident signal from the base station to the user on the corresponding side. At the same time, the present invention considers that the direct channel between the base station and the user is blocked by an obstacle, so the single-antenna user user 1 receives the signal reflected from STAR-RIS, and the single-antenna user user2 receives the signal transmitted from STAR-RIS. In addition, the present invention considers the rate division multiple access scheme, and the information W sent to user 1 on the base station side is 1 It will be divided into two parts: public information part W c,1 and private information part W p，1 ; Information W sent to user2 2 It will be divided into two parts: public information part W c，2 and private information part W p，2 The common part W of user 1 and user 2 c,1 , W c，2 Will be encoded into a public stream s 0 , private information part W p，1 , W p，2 Will be encoded into private stream s 1 ,s 2 .

[0069] The signal finally sent by the base station is s 0 ,s 1 ,s 2 The superimposed signal x obtained after precoding is specifically expressed as:

[0070]

[0071] Among them, w k Yes k The precoding vector is a complex matrix of M×1; sk The subscript k is 0, 1 or 2, which refers to the public flow, the first private flow and the second private flow respectively. The public flow or the private flow can be considered to obey the Gaussian distribution with a mean of 0 and a variance of 1, that is, E[s k s k H ]=1, H represents conjugate transpose.

[0072] Considering that the received signals of single-antenna user user 1 and single-antenna user user 2 are in the same form, in order to simplify the representation, useri refers to any single-antenna user in user 1 or user 2, i∈{1,2}. Therefore, the received signal y at single-antenna user user i is i It can be expressed as:

[0073] y i =h i H Φ i Gx+n i (2)

[0074] Where G is the channel parameter from the base station to STAR-RIS, which is an N×M complex matrix; Φ i is the coefficient matrix of STAR-RIS on the useri side, which is an N×N complex matrix; h i is the channel parameter from STAR-RIS to user i, which is an N×1 complex matrix; n i is zero mean and has a variance of σ 2 The additive white Gaussian noise (AWGN) is a kind of noise that is generated by the

[0075] The coefficient matrix Φ of STAR-RIS on the user i side i The expression is as follows:

[0076]

[0077] Among them, β i , 1， β i , 2， …, β i , n， …, β i , N They represent the amplitude coefficients of the reflection or transmission side of the 1st, 2nd, ...n, ..., Nth units of STAR-RIS located on the user i side; θ i，1 ,θ i，2 ,…,θ i，n ,…,θ i，Nis the phase coefficient of the 1st, 2nd, ...n, ..., Nth units of STAR-RIS located on the user i side. The phase coefficient of the nth unit of STAR-RIS located on the user i side satisfies θ i，n ∈[0, 2π], n∈{1, 2, 3,...,N}.

[0078] Furthermore, the amplitude coefficient of the nth unit of STAR-RIS located on the user 1 side is denoted as β 1，n , the amplitude coefficient of the nth unit of STAR-RIS located on the user 2 side is recorded as β 2，n , then β 1，n ∈[0, 1], β 2，n ∈[0, 1], β 1，n 2 +β 2，n 2 =1.

[0079] The present invention considers the PS (Power Splitting) transmission protocol in the SWIPT system, that is, an energy splitting device is used at the user to divide the energy of the received signal into two parts, one for information processing and the other for energy recovery. The energy splitting ratio at user i is set to α i ∈[0, 1], that is, the received energy α i times for information processing, (1-α i ) times for energy recovery. When a user performs information processing, according to the rate division multiple access strategy, the user needs to decode the public stream first. Channel state information is needed when decoding the public stream. The present invention considers that the user end has obtained perfect channel state information by means of channel estimation. When decoding the public stream, the private stream will be regarded as noise. After decoding the public stream, the user removes the public stream part of the signal used for information processing and continues to decode their respective private streams.

[0080] Therefore, the information rate R of user i decoding the public stream is c，i and the information rate R of the private flow p，i It can be expressed as:

[0081]

[0082] Among them, w 0 Indicates 0 The precoding vector of 1 Indicates 1 The precoding vector of 2 Indicates 2 The precoding vector of 2 represents the noise variance.

[0083] Power P for energy recovery i EH It can be expressed as:

[0084]

[0085] 2. Design Optimization Problem

[0086] S2: Based on the system model of the STAR-RIS-assisted SWIPT system, the rate split multiple access (RSMA) scheme is considered on the base station side, and an optimization problem with maximizing the system energy efficiency as the optimization goal is constructed. Considering the communication rate requirement constraints of the receiving side users, the energy collection requirement constraints, and the energy limit constraints on the base station side, the active beams on the base station side, the rate split ratio, and the passive beams on the STAR-RIS side are jointly optimized to maximize the system energy efficiency.

[0087] It should be noted that in step S2 of the present invention, the goal is to maximize the energy efficiency of the system under the conditions of considering the power overhead limit of the base station, the user information rate and the energy recovery requirement, and the corresponding optimization problem is as follows:

[0088]

[0089] c i +R p，i ≥λ i (7b)

[0090] P i EH ≥ζ i (7c)

[0091]

[0092] θ i，n ∈[0,2π] (7e)

[0093] β 1，n 2 +β 2，n 2 =1 (7f)

[0094] Among them, (7a) is the constraint to ensure that each user can successfully decode the public stream, (7b) is the communication rate requirement constraint of user i, (7c) is the energy collection requirement constraint of user i, (7d) is the energy limit constraint on the base station side, (7e) is the phase constraint of the STAR-RIS unit, and (7f) is the amplitude coefficient constraint when STAR-RIS works in ES mode.

[0095] Where: c i represents the public information rate of user i; irepresents the minimum communication rate requirement of user i; i represents the minimum energy collection requirement of user i; P max It represents the maximum transmit power that the base station can provide.

[0096] 3. Solving Optimization Problems

[0097] S3: The precoding design suitable for the base station side of the RSMA-SWIPT system is realized through the maximum ratio transmission-zero forcing precoding scheme.

[0098] In the present invention, due to the non-convexity of the objective function of the optimization problem p0 and the mutual coupling of the optimization variables within each constraint leading to the non-convexity of the constraint conditions, the optimization problem p0 is a non-convex problem. To solve the above problems, the present invention first adopts a maximum ratio transmission-forcing zero precoding scheme to simplify the original optimization problem p0, and then decomposes the simplified optimization problem into three sub-problems: sub-problem 1 for solving the amplitude coefficient on the STAR-RIS side, sub-problem 2 for solving the phase coefficient on the STAR-RIS side, and sub-problem 3 for solving the rate division ratio on the base station side and the energy recovery ratio on the user side. The original optimization problem p0 is solved by alternately iteratively solving these three sub-problems until the objective function converges.

[0099] It should be noted that in step S3, according to the prior art document B. Clerckx, Y. Mao, R. Schober, and H. V. Poor, "Rate-splitting unifying sdma, oma, noma, and multicasting in misobroadcast channel: A simple two-user rate analysis," IEEE Wireless Communications Letters, vol. 9, no. 3, pp. 349-353, 2020, in the RSMA scheme, the public stream will carry the main energy and all users need to decode it, so the maximum ratio transmission (MRT) precoding strategy is adopted to improve the signal-to-noise ratio of the receiving end; and the private stream is the part that different users decode and belongs to themselves. The private stream that does not belong to the user will be regarded as interference during the decoding process, so the zero-forcing (ZF) precoding scheme should be adopted for the private stream. Therefore, the present invention adopts a base station side precoding design suitable for the RSMA-SWIPT system for the base station side precoding design, which is implemented by the maximum ratio transmission-zero-forcing precoding scheme MRT-ZF. The specific precoding design scheme is as follows:

[0100]

[0101] q i =hi H Φ i G

[0102] Where H represents the conjugate transpose; q i represents an intermediate variable; z k Represents the intermediate variable Z = [z 1 , z 2 ]; Z = Q H (QQ H ) represents an intermediate variable; Q = [q 1 ,q 2 ] represents an intermediate variable; q 1 ,q 2 Both represent intermediate variables.

[0103] S4: Decompose the optimization problem in S2 into three sub-problems, namely, sub-problem 1 for solving the amplitude coefficient on the STAR-RIS side, sub-problem 2 for solving the phase coefficient on the STAR-RIS side, and sub-problem 3 for solving the rate split ratio on the base station side and the energy recovery ratio on the user side. Solve the three sub-problems by alternating iterations until the optimization problem converges, and obtain the solution to the optimization problem. The values ​​of each optimization variable when the optimization problem converges are used as the corresponding design for achieving the optimal energy efficiency of the SWIPT system.

[0104] In the present invention, for the energy efficiency maximization problem of the STAR-RIS-RMSA-assisted SWIPT system, since there are many variables to be optimized and different constraints are coupled with each other, an alternating optimization method can be used to optimize the variables separately. This embodiment decomposes the original optimization problem into three sub-problems according to different optimization variables, namely: Sub-problem 1. Solving the amplitude coefficient on the STAR-RIS side, Sub-problem 2. Solving the phase coefficient on the STAR-RIS side, Sub-problem 3. Solving the rate split ratio on the base station side and the energy recovery ratio on the user side. For different sub-problems, due to their non-convexity, the present invention adopts different optimization algorithms to convert them into convex problems, and then alternately iterates to solve these sub-problems until the above optimization problem converges.

[0105] In step S4, the solution of the optimization problem is obtained. The specific implementation process is as follows: Figure 3 As shown: In the t-th iteration round, determine whether the optimization problem converges: if converged, the solution results of the three sub-problems when the optimization problem converges are used as the solution results of the optimization problem; if not converged, the three sub-problems are solved separately to obtain the solution results of the three sub-problems in the t-th iteration round, and enter the t+1-th iteration round until the optimization problem converges.

[0106] In this embodiment, the specific process of converting each of the three sub-problems into a solvable convex problem is introduced below.

[0107] 1) Solve the sub-problem 1 of the STAR-RIS side amplitude coefficient

[0108] In subproblem 1, the goal is to optimize the amplitude coefficient β of the reflection or transmission side of the nth unit of STAR-RIS located on the user i side i，n In the present invention, a scheme for setting amplitude coefficients suitable for actual STAR-RIS deployment is first considered, and then other optimization variables are fixed, a sub-optimization problem about optimizing the amplitude coefficients is given, and sub-problem 1 is transformed into a convex problem by using fractional programming (FP) and semi-definite relaxation (SDR).

[0109] Since the number of STAR-RIS working units deployed in actual communication scenarios will be large, if the amplitude coefficient of each working unit is to be independently regulated, this will greatly increase the complexity of hardware control during actual deployment. Therefore, the present invention considers an amplitude coefficient setting scheme that is more suitable for actual deployment and has lower optimization complexity: the amplitude coefficient of the reflection or transmission side of each working unit is equal, that is:

[0110] β i , 1 =β i , 2 =…=β i , n=…=β i , N =β i (9)

[0111] Then, the optimization problem p0 can be reformulated as:

[0112]

[0113]

[0114] t ik =h i H diag(Gw k )Θ i

[0115] Where: β i represents the amplitude coefficient of the STAR-RIS working unit close to user i; γ i , γ 1 , γ 2 All represent auxiliary variables; t ik ,t i0 ,t ii Both represent intermediate variables, Θ i =[θ i，1 ,θ i，2 , ..., θi，N ] T represents the phase coefficient vector of the STAR-RIS working unit located on the user i side; β 1 represents the amplitude coefficient of the reflection side of any unit of STAR-RIS located on the user 1 side; β 2 Represents the amplitude coefficient of the transmission side of any unit of STAR-RIS located on the user 2 side.

[0116] According to the related theory of fractional programming in the prior art "K.Shen and W.Yu, "Fractional programming for communication systems-part ii: Uplink scheduling via matching," IEEE Transactions on Signal Processing, vol.66, no.10, pp.2631-2644, 2018," an auxiliary variable y is introduced i , through Lagrange dual transformation and quadratic transformation, the optimization problem (10) is transformed into formula (11):

[0117]

[0118] In Beta i With y i When the value of is fixed, the optimization problem (11) only has Y i A variable to be optimized, according to the literature K.Shenand W.Yu. "Fractional programming for communication systems-part ii: Uplinkscheduling via matching," IEEE Transactions on Signal Processing, vol.66.no.10, pp.2631-2644, 2018, introduces the value of the amplitude coefficient βi in the t-1th iteration round The optimization problem (11) is used to obtain the optimal value of the auxiliary variable γi It is expressed as formula (12):

[0119]

[0120] When γi is fixed, the optimal value of the auxiliary variable yi is It is expressed as formula (13):

[0121]

[0122] When updating γ in each iteration through equations (12) and (13) i and i When , only the terms related to these two variables in the optimization problem (11) can be regarded as constants. Therefore, the optimization problem (11) can be re-expressed as formula (14):

[0123]

[0124] For simplicity of expression, the present invention introduces intermediate variables P, q, B 1 , B 2 , A 1 , A 2 and two constants A i and B i , introducing the STAR-RIS amplitude coefficient value vector β, and then re-expressing equation (14) as equation (15):

[0125] P1: β T Pβ+q T β(15)

[0126]

[0127] q=[-A 1 , -A 2 ] T

[0128] β=[β1,β2] T

[0129]

[0130] The subproblem P1 in formula (15) is a real variable quadratic constrained quadratic programming problem. The prior art Z.-q.Luo, W.-k.Ma, AM-c.So, Y.Ye, and S.Zhang, “Semidefinite relaxation of quadratic optimization problems,” IEEE Signal Processing Magazine, vol.27, no.3, pp.20-34, 2010. shows that this problem can be solved by semidefinite programming.

[0131] 2) Solve the sub-problem 2 of the phase coefficient on the STAR-RIS side

[0132] In the present invention, the goal of sub-problem 2 is to optimize the STAR-RIS side phase coefficient θ i，n Therefore, the values ​​of other optimization variables are fixed and the phase coefficient θ of STAR-RIS is optimized. i，n, by introducing the phase coefficient vector Θ of the STAR-RIS working unit located on the user i side i and the intermediate variable g ik , g i0 , g i1 , g i2 , g ii , the optimization problem p0 is reformulated as formula (16):

[0133]

[0134] |θ i，n |=1 (16b)

[0135]

[0136] g ik =h i H diag(Gw k )β i

[0137] Due to the non-convexity of constraint (16b), a penalty function-based method is used to deal with this constraint. Equation (16) can be re-expressed as Equation (17):

[0138]

[0139] st|θ i，n |≤1

[0140] (17a), (16a), (16b), (16c), (16d)

[0141] Where: C is a large enough positive number to force This term is in θ i The result when the optimal value is obtained is 0.

[0142] In order to deal with other non-convex constraints, the sequential convex approximation method is used. Introducing auxiliary variables as well as Introducing intermediate variable C i，k , C ii , C i，0 , D i , and These items are first-order Taylor approximated according to the prior art literature “Z. Yang, J. Shi, Z. Li, M. Chen, W. Xu, and M. Shikh-Bahaei, “Energy efficient rate splitting multiple access (RSMA) with reconfigurable intelligent surface,” in 2020 IEEE International Conference on Communications Workshops (ICC Workshops), 2020, pp. 1-6”, and equation (17) is re-expressed as equation (18). P2 in equation (18) is a convex problem, which can be solved using a convex optimization toolbox such as the matlab cvx toolbox.

[0143]

[0144] C i，k =Re{(g ik Θ i t-1 ) H g ik (Θ i -Θ i t-1 )}

[0145] D i =α i |g i1 Θ i | 2 +α i |g i2 Θ i | 2 +σ 2

[0146] Where: ai and bi represent auxiliary variables Index of Respectively represent the θ in the t-1th iteration round i，n , Θ i , a i , b i The value of .

[0147] 3) Solve the sub-problem 3 of the rate split ratio on the base station side and the energy recovery ratio on the user side

[0148] In the present invention, the goal of sub-problem 3 is to optimize the public information rate c of user i i And the energy split ratio α iIn order to optimize these two optimization variables, the values ​​of other optimization variables are fixed and the intermediate variable u is introduced. i，0 ,u i，k ,u i，i , then the optimization problem p0 can be transformed into only c i and α i The sub-problem to be optimized is shown in formula (19):

[0149]

[0150]

[0151] u i,k =|h i H Φ i G k | 2

[0152] By introducing intermediate variables And as well as These terms are approximated by the first-order Taylor approximation, and equation (19) is re-expressed as the following convex problem equation (20). Considering that equation (20) P3 is a convex problem, it can be solved using a convex optimization toolbox such as the cvx toolbox.

[0153]

[0154] (19c), (19d)

[0155] Where: d i , e i , f i All represent intermediate variables; Respectively represent the intermediate variable d in the t-1th iteration round i , e i , f i The value of .

[0156] The present invention will now use a specific example to demonstrate the application effect of the STAR-RIS-assisted SWIPT system energy efficiency optimization method described in S1 to S4 in the above embodiments, so as to facilitate understanding of the essence of the present invention.

[0157] Example

[0158] The specific implementation process of the STAR-RIS-assisted SWIPT system energy efficiency optimization method used in this embodiment is as described above and will not be repeated here.

[0159] The simulation results are used below to demonstrate the performance of the proposed method. The simulation parameters are set as follows: the base station and STAR-RIS are located at (0,0) and (0,20m) respectively. The number of working units on STAR-RIS is N=20, the transmission channel gains in the system all obey the Rice distribution, the Rice factor is set to 10, and the path loss model is set to 35.6+22lgd(dB). The maximum transmission power P of the base station is set to 40dBm, and the noise power is set to -90dBm. The minimum communication transmission rate required by the user is 0.1Mbit / s, and the minimum energy recovery requirement is set to -80dBm. The system bandwidth is set to 1MHz.

[0160] Figure 4 The simulation results of the convergence performance of the method of the present invention are shown in Figure 1. This embodiment considers three different transmission schemes: Scheme 1, STAR-RIS adopts the proposed method to obtain the optimal phase design; Scheme 2, the phase of the working unit on STAR-RIS is set to a random phase; Scheme 3, the traditional RIS is used for transmission. From the perspective of the convergence performance of the method, the three schemes can converge quickly within 10 rounds using the proposed method; from the perspective of the effectiveness of the method, the convergence value of Scheme 1 is better than that of Scheme 2, because the use of the optimal phase setting can obtain richer STAR-RIS passive beamforming gain. In addition, the convergence value of Scheme 1 is better than that of Scheme 3, because STAR-RIS has a wider coverage and can serve users on both sides at the same time.

[0161] Figure 5 and Figure 6 The comparisons are made on various multiple access schemes. From the simulation results, in the communication system studied by the present invention, the performance of RSMA is better than that of SDMA and NOMA. This is because RSMA can flexibly adapt to different levels of multi-user interference by adjusting the rate division ratio on the base station side, providing a higher degree of freedom and thus achieving better energy efficiency. Among them, Figure 5 Describing the changing trend of the energy efficiency of different communication schemes with the number of antennas on the base station side, it can be seen that with the increase in the number of antennas on the base station side, the energy efficiency performance of all communication schemes has been improved. This is because increasing the number of antennas can obtain higher spatial division multiplexing gain. Figure 6 The energy efficiency of different communication schemes is described as the number of STAR-RIS working units changes. It can be seen that with the increase in the number of working units, the energy efficiency performance of all communication schemes is improved. This is because increasing the number of working units can obtain higher passive beam gains.

[0162] The above-described embodiment is only a preferred solution of the present invention, but it is not intended to limit the present invention. A person skilled in the relevant technical field may make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, any technical solution obtained by equivalent replacement or equivalent transformation falls within the protection scope of the present invention.

Claims

1. A STAR-RIS-assisted SWIPT system energy efficiency optimization method, characterized in that: The following steps are involved: S1: Based on the constructed system model of the STAR-RIS-assisted SWIPT system, the rate-splitting multiple access scheme is considered on the base station side, and an optimization problem with the optimization goal of maximizing the system energy efficiency is constructed. Under the consideration of the communication rate requirement constraints of the receiving side users, the energy collection requirement constraints, and the energy limit constraints on the base station side, the active beams on the base station side, the rate splitting ratio, and the passive beams on the STAR-RIS side are jointly optimized to maximize the system energy efficiency. S2: The precoding design suitable for the base station side of the RSMA-SWIPT system is realized through the maximum ratio transmission-zero forcing precoding scheme; S3: The optimization problem in S1 is decomposed into three sub-problems, namely, sub-problem 1 for solving the amplitude coefficient on the STAR-RIS side, sub-problem 2 for solving the phase coefficient on the STAR-RIS side, and sub-problem 3 for solving the rate split ratio on the base station side and the energy recovery ratio on the user side. The three sub-problems are solved by alternate iteration until the optimization problem converges, and the solution of the optimization problem is obtained. The values ​​of each optimization variable when the optimization problem converges are used as the corresponding design for achieving the optimal energy efficiency of the SWIPT system.

2. The STAR-RIS-assisted SWIPT system energy efficiency optimization method according to claim 1, characterized in that: In step S1, the constructed system model is as follows: consider a two-user STAR-RIS-assisted SWIPT system, the base station has M antennas, the STAR-RIS has N reflection units and N transmission units, wherein a single-antenna user user 1 is in the reflection area of ​​the STAR-RIS, and another single-antenna user user 2 is in the transmission area of ​​the STAR-RIS, the STAR-RIS adopts the ES transmission mode, the single-antenna user user 1 receives the signal reflected by the STAR-RIS, and the single-antenna user user 2 receives the signal transmitted by the STAR-RIS, the information sent by the base station side to the single-antenna user user 1 is divided into a first public information part and a first private information part, the information sent to the single-antenna user user 2 is divided into a second public information part and a second private information part, the first public information part and the second public information part are encoded into a public stream, the first private information part is encoded into a first private stream, the second private information part is encoded into a second private stream, and the superimposed signal obtained after precoding the public stream, the first private stream and the second private stream is used as the signal finally sent by the base station.

3. The STAR-RIS-assisted SWIPT system energy efficiency optimization method according to claim 1, characterized in that: In step S1, the function form of the optimization problem is as follows: c i +R p，i ≥λ i (1b) P i EH ≥ζ i (1c) i i，n ∈[0,2π] (1e) β 1，n 2 +β 2，n 2 ＝1 (1f) Among them, (1a) is the constraint to ensure that each user can successfully decode the public stream, (1b) is the communication rate requirement constraint of user i, (1c) is the energy collection requirement constraint of user i, (1d) is the energy limit constraint on the base station side, (1e) is the phase constraint of the STAR-RIS unit, and (1f) is the amplitude coefficient constraint when STAR-RIS works in ES mode; Where: c i represents the public information rate of user i; w k Represents the precoding vector of the public or private stream; α i represents the energy split ratio at user i; θ i，n represents the phase coefficient of the nth unit of STAR-RIS located on the user i side; β i，n represents the amplitude coefficient of the reflection or transmission side of the nth unit of STAR-RIS located on the user i side; R p，i R represents the information rate of the private stream decoded by user i; c，i represents the information rate of the public stream decoded by user i; i represents the minimum communication rate requirement of user i; P i EH Represents the power used for energy recovery; i represents the minimum energy collection requirement of user i; P max It is the maximum transmission power that the base station can provide; β 1，n represents the amplitude coefficient of the reflection side of the nth unit of STAR-RIS located on the user 1 side; β 2，n Represents the transmission amplitude coefficient of the nth unit of STAR-RIS located on the user 2 side.

4. The STAR-RIS-assisted SWIPT system energy efficiency optimization method according to claim 3, characterized in that: In step S2, the maximum ratio transmission-zero forcing precoding scheme is specifically: q i =h i H Φ i G Where: H represents conjugate transpose; q i represents an intermediate variable; z k represents the kth column of the intermediate variable Z = [z1, z2]; Z = Q H (QQ H ) represents an intermediate variable; Q = [q1, q2] represents an intermediate variable; q1, q2 both represent intermediate variables; h i is the channel parameter from STAR-RIS to useri; Φi is the coefficient matrix of STAR-RIS on the user i side; G is the channel parameter from the base station to STAR-RIS.

5. The STAR-RIS-assisted SWIPT system energy efficiency optimization method according to claim 1, characterized in that: In step S3, the specific process of obtaining the solution result of the optimization problem is as follows: in the tth iteration round, judging whether the optimization problem converges; if converged, taking the solution results of the three sub-problems when the optimization problem converges as the solution result of the optimization problem; If it does not converge, the three sub-problems are solved separately to obtain the solution results of the three sub-problems in the t-th iteration round, and then enter the t+1-th iteration round until the optimization problem converges.

6. The STAR-RIS-assisted SWIPT system energy efficiency optimization method according to claim 3, characterized in that: In step S3, the functional form of sub-problem 1 is: P1: b T Pβ+q T b (3) <h2 style=";text-align:left;direction:ltr">q=[-A1,-A2]<h2 style=";text-align:left;direction:ltr"> T β＝[β1，β2] T Where: β represents the STAR-RIS amplitude coefficient value vector; T represents transposition; β i represents the amplitude coefficient of the working unit of STAR-RIS close to user i; β1 represents the amplitude coefficient of the reflection side of any unit of STAR-RIS located on the user 1 side; β2 represents the amplitude coefficient of the transmission side of any unit of STAR-RIS located on the user 2 side; P, t ik t i0 t ii , B1, B2, A1, A2 all represent intermediate variables; σ 2 represents the variance of complex additive Gaussian white noise; γ i Represents auxiliary variables.

7. The STAR-RIS-assisted SWIPT system energy efficiency optimization method according to claim 6, characterized in that: In step S3, the functional form of sub-problem 2 is: s.t.|θ i，n |≤1 (4f) Where: Θ i represents the phase coefficient vector of the STAR-RIS working unit located on the user i side; C represents a positive number; a i , b i Represent auxiliary variables The index of g ik , C i，k , g i0 , C ii , C i，0 , D i , g ii All represent intermediate variables; Respectively represent the θ in the t-1th iteration round i，n , Θ i , a i , b i The value of; N represents the number of STAR-RIS working units.

8. The STAR-RIS-assisted SWIPT system energy efficiency optimization method according to claim 7, characterized in that: In step S3, the functional form of sub-problem 3 is: Where: u i，0 ,u i，k ,u i，i , d i , e i , f i All represent intermediate variables; Respectively represent the intermediate variable d in the t-1th iteration round i , e i , f i The value of .

Images