Transceiver optimization method for IRS-assisted multicarrier MIMO simultaneous wireless information and power transfer system

Abstract

The invention provides a transceiver optimization method of an IRS-assisted multicarrier MIMO information and power transfer system, and relates to the technical field of information and power transfer systems, and the method comprises the following steps: carrying out linear pre-coding or Thomlinson-Harat nonlinear pre-coding processing on symbol vectors to be carried by different subcarriers; therefore, a signal vector corresponding to the linear transceiver or the nonlinear transceiver is obtained; establishing a receiving signal of a transceiver of the multi-carrier MIMO simultaneous wireless information and power transfer system based on the signal vector; on the premise that given transmitting power constraint and energy collection power constraint are met, a transceiver optimization problem taking the sum of mean square errors of minimized received signals as a target function is established and solved, and a locally optimal transceiver optimization scheme is obtained. According to the invention, the linear transceiver and the nonlinear transceiver are effectively unified, and the nonlinear transceiver can be applied and expanded to the linear transceiver so as to adapt to different use scenes.

Description

Technical Field

[0001] The present invention relates to the technical field of wireless power-carrying communication systems, and in particular to a transceiver optimization method of an IRS-assisted multi-carrier MIMO wireless power-carrying communication system. Background Art

[0002] In the era of B5G and 6G, the literature [1] proposed that the Internet of Things (IoT) is a key technology with great development potential. This is because it can significantly improve people's quality of life and has many specific application scenarios, such as autonomous driving, implantable devices, and smart medical care. In the future, the use of a large number of IoT devices will consume more and more electricity. Therefore, how to extend the working cycle of the IoT network and improve energy efficiency is an urgent issue to be solved.

[0003] In this demand context, wireless power transfer technology has attracted widespread attention from academia and industry. This technology can achieve parallel transmission of information and energy, providing support for green communication in energy-constrained IoT networks. Reference [2] proposed that power splitting (PS) and time-switching (TS) are two feasible technical solutions for achieving wireless information and energy transmission. Power splitting is that the receiver uses a power divider to divide the received signal into two parts according to a certain ratio, one part is used for information decoding (ID) and the other part is used for energy harvesting (EH). Time-slot switching is to alternately perform information decoding and energy harvesting in different time periods. However, for traditional wireless information and power transfer (SWIPT) systems, due to severe channel attenuation, the received signal strength will be significantly reduced, which limits the EH and ID performance of SWIPT. In order to overcome this dilemma, reference [3] proposed that intelligent reflecting surface (IRS) technology has become a very promising solution in recent years. IRS improves the received signal strength by reconstructing a favorable wireless propagation environment. Reference [4] proposed that IRS is a metal surface composed of a certain number of passive reflective units, and the phase of each reflective unit can be adjusted by an intelligent controller. In view of the respective advantages of SWIPT and IRS in wireless communications, the IRS-assisted SWIPT system has also become a research hotspot in academia and industry in recent years.

[0004] So far, IRS-assisted PS-based SWIPT systems have been studied in depth in references [5-8]. Reference [5] proposed a joint optimization framework for IRS-assisted PS-based MIMO SWIPT IoT networks and maximized the weighted sum rate under transmit power and EH constraints. Reference [6] studied a multi-user IRS-assisted PS-based MISO SWIPT system and maximized the system's energy efficiency under the premise of satisfying transmit power constraints, EH constraints and data rate constraints. Reference [7] studied an IRS-assisted cooperative non-orthogonal multiple access PS-based MISO SWIPT system and maximized the system's fairness rate under quality of service (QoS) and transmit power constraints. Reference [8] designed a physical layer security beamforming scheme for IRS-assisted PS-based NOMA MISO SWIPT networks and achieved the maximum sum rate under EH, interference power and signal-to-interference-noise ratio constraints. The above references [5-8] mainly use data rate as the design criterion of the system, so this is not suitable for characterizing the accuracy of signal recovery at the receiving end. In addition, these works mainly study the linear transceiver design of IRS-assisted SWIPT systems. References [9-10] proposed that for traditional wireless information transfer (WIT) systems, combining multiple-input multiple-output (MIMO) with nonlinear Tomlinson-Harashimap precoding (THP) or decision feedback equalization technology can achieve potential communication performance improvement. References [11-12] have applied THP to MIMO SWIPT and IRS-assisted MIMO systems respectively. The IRS-assisted MIMO SWIPT system combines the advantages of IRS and SWIPT, but the research on IRS-assisted MIMO SWIPT system based on THP is still an open problem. Therefore, how to apply THP to IRS-assisted MIMO SWIPT systems to improve communication performance and unify the design schemes of linear transceivers and nonlinear transceivers to adapt to different usage scenarios is an extremely important technical problem to be solved.

[0005] The literature is:

[0006] [1]P.X.Nguyen et al.,"Backscatter-Assisted Data Offloading in OFDMA-Based Wireless-Powered Mobile Edge Computing for IoTNetworks,"in IEEEInternet Things J.,vol.8,no.11,pp.9233-9243,1June1,2021.

[0007] [2]R.Zhang and C.K.Ho,"MIMO Broadcasting for Simultaneous WirelessInformation and Power Transfer,"in IEEE Trans.Wirel.Commun.,vol.12,no.5,pp.1989-2001,May 2013.

[0008] [3]C.Pan et al.,"Intelligent Reflecting SurfaceAided MIMOBroadcasting for Simultaneous Wireless Information andPower Transfer,"in IEEEJ.Sel.Areas Commun.,vol.38,no.8,pp.1719-1734,Aug.2020.

[0009] [4]S.Zhang and R.Zhang,"Capacity Characterization for IntelligentReflecting SurfaceAided MIMO Communication,"in IEEE J.Sel.Areas Commun.,vol.38,no.8,pp.1823-1838,Aug.2020.

[0010] [5]X.Xie et al.,"A Joint Optimization Framework for IRS-AssistedEnergy Self-Sustainable IoT Networks,"in IEEE Internet Things J.,vol.9,no.15,pp.

[0011] 13767-13779,1Aug.1,2022.

[0012] [6]J.Tang,Z.Peng,D.K.C.So,X.Zhang,K.-K.Wong and J.A.Chambers,"EnergyEfficiency Optimization for a Multiuser IRS-Aided MISO System With SWIPT,"inIEEE Trans.Commun.,vol.71,no.10,pp.5950-5962,Oct.2023.

[0013] [7]Z.Yang et al.,"User Fairness Optimization ofIRS-AssistedCooperative MISO-NOMA for ITS With SWIPT,"in IEEE Trans.Intell.Transp.Syst.,vol.25,no.7,pp.6861-6872,July 2024.

[0014] [8]R.Sun,W.Wang,L.Xu,N.Zhao,N.Al-Dhahir,and X.Wang,“Securebeamforming for IRS-assistedNOMA SWIPT networks,”IEEE Trans.Commun.,pp.1–1,2024.

[0015] [9]M.B.Shenouda and T.N.Davidson,"A framework for designing MIMOsystems with decision feedback equalization or Tomlinson-harashimaprecoding,"in IEEE J.Sel.Areas Commun.,vol.26,no.2,pp.401-411,February 2008.

[0016]

[10] AAD'Amico, "Tomlinson-Harashima Precoding in MIMO Systems: AUnifiedApproach to Transceiver Optimization Based on Multiplicative Schur-Convexity," in IEEE Trans.Signal Process., vol.56, no.8, pp.3662-3677, Aug.2008.

[0017]

[11] Q.Li, Q.Zhang and J.Qin, "Robust Tomlinson–Harashima Precoding WithGaussian Uncertainties for SWIPT in MIMO Broadcast Channels," in IEEETrans.Signal Process., vol.65, no.6, pp.1399-1411, 15March15, 2017.

[0018]

[12] S.Gong, C.Xing, Summary of the invention

[0019] In order to solve the problem in the above-mentioned prior art that the THP-based nonlinear transceiver has not been applied to the IRS-assisted multi-carrier MIMO system to improve the performance of the communication system and fill this technical gap, the present invention proposes a transceiver optimization method for an IRS-assisted multi-carrier MIMO wireless energy-carrying communication system, which effectively unifies the design schemes of linear transceivers and nonlinear transceivers to meet different usage scenarios.

[0020] In order to achieve the above technical effects, the technical solution of the present invention is as follows:

[0021] A transceiver optimization method for an IRS-assisted multi-carrier MIMO wireless energy-carrying communication system comprises the following steps:

[0022] S1. Linear precoding or Tomlinson-Harashima nonlinear precoding of symbol vectors to be carried by different subcarriers to obtain a signal vector corresponding to a nonlinear transceiver or a linear transceiver;

[0023] S2 based on the signal vector, establish a multi-carrier MIMO wireless communication system transceiver receiving signal;

[0024] S3. Under the premise of satisfying the given transmit power constraint and collection power constraint, establish a transceiver optimization problem with minimizing the sum of the mean square error of the received signal as the objective function;

[0025] S4. Solve the transceiver optimization problem to obtain a locally optimal transceiver optimization solution.

[0026] Preferably, the multi-carrier MIMO wireless energy communication system includes a t A transmitter with N antennas r Receiver with N antennas and s The multi-carrier MIMO wireless energy communication system has a total bandwidth of W, which is evenly distributed to N subcarriers, and the calculation expression of the equivalent frequency domain channel matrix of the kth subcarrier is:

[0027]

[0028] Among them, H k represents the equivalent frequency domain channel matrix of the kth subcarrier, represents the equivalent frequency domain channel matrix of the k-th subcarrier transmitter-receiver link, represents the equivalent frequency domain channel matrix of the k-th subcarrier transmitter-intelligent reflecting surface IRS link, represents the equivalent frequency domain channel matrix of the k-th subcarrier intelligent reflecting surface IRS-receiver link; represents the reflection coefficient matrix of the intelligent reflective surface IRS, diag(.) represents the diagonal matrix, φ m represents the reflection coefficient of the mth passive reflection unit, φ m The adjustment range of the phase is [0,2π).

[0029] Preferably, the performing linear precoding or Tomlinson-Harashima nonlinear precoding processing on the symbol vectors to be carried by different subcarriers includes:

[0030] The symbol vector carried by the kth subcarrier is defined as L k ≤min{N t ,N r}, through the modulus operator MOD and the lower triangular feedback matrix The symbol vectors to be carried by different subcarriers are processed nonlinearly, and the calculation expression of the signal vector corresponding to the nonlinear transceiver is obtained as follows:

[0031]

[0032] Among them, v k (n) represents the signal vector v k The nth element in represents a complex vector, v k The real and imaginary parts of all elements are limited to The modulo operation expression is defined as:

[0033]

[0034] Among them, MOD M (.) represents the modulo operation, x represents the input of the modulo operation, represents the largest integer not exceeding z;

[0035] In a nonlinear transceiver, v k The calculation expression is:

[0036]

[0037] in, Represents the lower triangular matrix with unit diagonal elements, u k =s k +i k Represents a valid signal vector; when B k = 0, the nonlinear transceiver is simplified to a linear transceiver,

[0038] Preferably, the calculation expression for establishing the received signal of the transceiver of the multi-carrier MIMO wireless energy-carrying communication system based on the signal vector is:

[0039]

[0040] Among them, y k Indicates receiving signal, represents the frequency domain precoding matrix in the transmitter, represents the antenna noise vector, which is an additive white Gaussian noise vector.

[0041] Preferably, the receiver divides the received signal into a received signal power of a β-proportional part and a received signal power of a 1-β-proportional part through a power divider, and the received signal power of the β-proportional part is used for energy collection, and the frequency domain signal y sent to the energy collector is kEH The calculation expression is:

[0042]

[0043] The 1-β ratio of the received signal power is used for information decoding, and the frequency domain signal sent to the information decoder is It is expressed as:

[0044]

[0045] in, represents the decoding noise efficiency and is the additive white Gaussian noise vector introduced by the digital circuit in the information decoder.

[0046] Preferably, the transceiver optimization problem is established as follows:

[0047]

[0048] in, represents the equalization matrix of the kth subcarrier, M k represents the MSE matrix of the kth subcarrier, Tr(.) represents the trace of the matrix, P th Represents the maximum transmit power of the system, E in represents the input power of the energy harvester, E th represents the energy harvesting power requirement of the system design, Indicates the average energy harvesting power threshold.

[0049] Preferably, the MSE matrix M of the k-th subcarrier k The calculation method is as follows:

[0050] The frequency domain signal Equalization into balanced signal The calculation expression is:

[0051]

[0052] Among them, based on the equalized signal The MSE matrix of the kth subcarrier is calculated as Equalize the signal The calculation expression is converted into:

[0053]

[0054] in, The calculation expression is:

[0055]

[0056] in, represents the decoding noise variance, Ψ k (P k ,G k ,C k ) is calculated as:

[0057]

[0058] in, Indicates equivalent to .

[0059] Preferably, the input power E of the energy harvester is in The calculation method is as follows:

[0060]

[0061] in, represents the antenna noise variance, and the superscript H represents the conjugate device of the matrix.

[0062] Preferably, the average energy harvesting power threshold The calculation method is as follows:

[0063] Based on input power E in , calculate the harvested power γ(E in )for:

[0064]

[0065] Considering the energy harvesting power E required by the system design th , let the energy collection constraint be γ(E in )≥E th , the energy harvesting constraints are transformed into equivalent constraints as follows:

[0066]

[0067] in, The calculation expression is as follows:

[0068]

[0069] Among them, E m and E 0 represent the saturation power and activation power of the energy harvester respectively, and τ and ν characterize the circuit characteristics of the energy harvester.

[0070] Preferably, in S4, an alternating optimization algorithm is used to solve the transceiver optimization problem, including: using an alternating method to solve {G k}、{C k}、{P k,β} and Φ are optimized until the objective function converges, and the transceiver optimization solution is obtained.

[0071] Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

[0072] The present invention proposes a transceiver optimization method for an IRS-assisted multi-carrier MIMO wireless power-carrying communication system. First, linear precoding or Tomlinson-Harashima nonlinear precoding processing is performed on symbol vectors to be carried by different subcarriers, and a signal vector corresponding to the linear transceiver or the nonlinear transceiver is obtained. Then, based on the signal vector, a receiving signal of the transceiver of the multi-carrier MIMO wireless power-carrying communication system is established. By establishing and solving a transceiver optimization problem, a locally optimal transceiver optimization solution is obtained, thereby providing a unified linear transceiver or nonlinear transceiver design framework to adapt to different usage scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 A flowchart showing a transceiver optimization method for an IRS-assisted multi-carrier MIMO wireless energy-carrying communication system proposed in an embodiment of the present invention;

[0074] Figure 2 A schematic diagram showing the structure of a multi-carrier MIMO wireless energy-carrying communication system proposed in an embodiment of the present invention;

[0075] Figure 3 It represents the average sumMSE versus transmit power distribution diagram proposed in the embodiment of the present invention;

[0076] Figure 4 It represents the distribution diagram of average sumMSE versus number of reflection units proposed in the embodiment of the present invention;

[0077] Figure 5 A distribution diagram of average achievable rate versus transmit power proposed in an embodiment of the present invention is shown; DETAILED DESCRIPTION

[0078] The drawings are for illustrative purposes only and should not be construed as limiting the present patent;

[0079] It is understandable to those skilled in the art that some well-known contents may be omitted in the drawings;

[0080] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0081] Example 1

[0082] like Figure 1As shown, this embodiment provides a transceiver optimization method for an IRS-assisted multi-carrier MIMO wireless energy-carrying communication system, comprising the following steps:

[0083] S1. Linear precoding or Tomlinson-Harashima nonlinear precoding of symbol vectors to be carried by different subcarriers to obtain a signal vector corresponding to a nonlinear transceiver or a linear transceiver;

[0084] S2 based on the signal vector, establish a multi-carrier MIMO wireless communication system transceiver receiving signal;

[0085] S3. Under the premise of satisfying the given transmit power constraint and collection power constraint, establish a transceiver optimization problem with minimizing the sum of the mean square error of the received signal as the objective function;

[0086] S4. Solve the transceiver optimization problem to obtain a locally optimal transceiver optimization solution.

[0087] See also Figure 2 The multi-carrier MIMO wireless energy communication system includes N t A transmitter with N antennas r Receiver with N antennas and s The multi-carrier MIMO wireless energy communication system has a total bandwidth of W, which is evenly distributed to N subcarriers.

[0088] The equivalent frequency domain channel between transceivers mainly consists of two parts: direct link and reflection link, where the reflection link is dynamically adjusted by the IRS controller. Therefore, the calculation expression of the equivalent frequency domain channel matrix of the kth subcarrier is:

[0089]

[0090] in, represents the equivalent frequency domain channel matrix of the kth subcarrier, represents the equivalent frequency domain channel matrix of the k-th subcarrier transmitter-receiver link, represents the equivalent frequency domain channel matrix of the k-th subcarrier transmitter-intelligent reflecting surface IRS link, represents the equivalent frequency domain channel matrix of the k-th subcarrier intelligent reflecting surface IRS-receiver link; represents the reflection coefficient matrix of the intelligent reflective surface IRS, diag(.) represents the diagonal matrix, φ m represents the reflection coefficient of the mth passive reflection unit, φ m The adjustment range of the phase is [0,2π).

[0091] The performing linear precoding or Tomlinson-Harashima nonlinear precoding processing on symbol vectors to be carried by different subcarriers includes:

[0092] The symbol vector carried by the kth subcarrier is defined as L k ≤min{N t ,N r},s k All elements of are modulated by M-QAM and have a mean of 0 and a variance of 1. Before transmission, s k First, it is processed by the THP unit; the THP processing unit is through the modulus operator MOD and the lower triangular feedback matrix The symbol vectors to be carried by different subcarriers are processed nonlinearly. In particular, when B k = 0, the nonlinear transceiver is simplified to a linear transceiver, and the calculation expression of the signal vector corresponding to the nonlinear transceiver is obtained as follows:

[0093] The calculation expression is:

[0094]

[0095] Among them, v k (n) represents the signal vector v k The nth element in represents a complex vector, v k The real and imaginary parts of all elements are limited to The modulo operation expression is defined as:

[0096]

[0097] Among them, MOD M (.) represents the modulo operation, x represents the input of the modulo operation, represents the largest integer not exceeding z;

[0098] Based on the above analysis, in the nonlinear transceiver, k The calculation expression is:

[0099]

[0100] in, Represents the lower triangular matrix with unit diagonal elements, u k =s k +i k represents a valid signal vector; if M is large enough, then Established. When B k = 0, the nonlinear transceiver is simplified to a linear transceiver. For the linear transceiver, Ck =I Lk , v k =u k =s k .

[0101] Next, we use a frequency domain precoding matrix Used to process v k , and obtain the transmitted signal, which is converted into a received signal after propagating through the wireless channel. That is, based on the signal vector, the calculation expression of the received signal of the transceiver of the multi-carrier MIMO wireless energy communication system is established as follows:

[0102]

[0103] Among them, y k Indicates receiving signal, represents the frequency domain precoding matrix in the transmitter, represents the antenna noise vector, which is an additive white Gaussian noise vector, It means that it has zero mean and a variance of independent complex Gaussian random distributions.

[0104] For energy collection, all antennas are set to use a unified energy collection ratio β∈(0,1) for power division. The receiver divides the received signal into the received signal power of the β-proportion part and the received signal power of the 1-β-proportion part through a power divider. The received signal power of the β-proportion part is used for energy collection, and the frequency domain signal sent to the energy collector is The calculation expression is:

[0105]

[0106] According to Parsval’s theorem, the input power of the energy harvester is It can be deduced that the input power E of the energy harvester is in The calculation method is as follows:

[0107]

[0108] in, represents the antenna noise variance, and the superscript H represents the conjugate device of the matrix.

[0109] The average energy harvesting power threshold The calculation method is as follows:

[0110] Based on input power E in , a nonlinear EH model is used to calculate the collected power γ(E in )for:

[0111]

[0112] Considering the energy harvesting requirements of the system design th , let the energy collection constraint be γ(E in )≥E th , the energy harvesting constraints are transformed into equivalent constraints as follows:

[0113]

[0114] in, The calculation expression is as follows:

[0115]

[0116] Among them, E m and E 0 represent the saturation power and activation power of the energy harvester respectively, and τ and ν characterize the circuit characteristics of the energy harvester.

[0117] The 1-β ratio of the received signal power is used for information decoding, and the frequency domain signal sent to the information decoder is It is expressed as:

[0118]

[0119] in, represents the decoding noise vector, which is the additive white Gaussian noise vector introduced by the digital circuit in the information decoder. It means that it has zero mean and a variance of independent complex Gaussian random distributions.

[0120] The MSE matrix M of the kth subcarrier k The calculation method is as follows:

[0121] The frequency domain signal Equalization into balanced signal The calculation expression is:

[0122]

[0123] Among them, based on the equalized signal The MSE matrix of the kth subcarrier is calculated as Equalize the signal The calculation expression is converted into:

[0124]

[0125] in, The calculation expression is:

[0126]

[0127] in, represents the second variance, Ψ k (P k ,G k ,C k ) is calculated as

[0128]

[0129] in, Indicates equivalent to .

[0130] The transceiver optimization problem is:

[0131]

[0132] in, represents the equalization matrix of the kth subcarrier, M k represents the MSE matrix of the kth subcarrier, Tr(.) represents the trace of the matrix, P th Represents the maximum transmit power of the system, E in represents the input power of the energy harvester, E th represents the energy harvesting requirements of the system design, Indicates the average energy harvesting power threshold.

[0133] In S4, It is a highly non-convex optimization problem because: the complex coupling relationship between different optimization variables exists in the objective function and the constraint condition C 2 At the same time, the constraint C 4 The unit module constraint also brings additional challenges to solving this problem.

[0134] Therefore, the alternating optimization (AO) algorithm is used to solve the non-convex problem. Solving the problem includes: using an alternating method to sequentially solve {G k}、{C k}、{P k ,β} and Φ are optimized until the objective function converges to obtain the transceiver optimization solution. The specific optimization method is as follows:

[0135] For the matrix {G k}, including:

[0136] Because {G k} only appears in the question In the objective function, this means that {G k The optimization of} can be simplified into an unconstrained optimization problem, About G k The first-order derivative of is zero, and the optimal closed-form expression of the equilibrium matrix can be derived:

[0137]

[0138] Substituting equation (1.12) into equation (1.10) and using the matrix inversion lemma, we can further obtain:

[0139]

[0140] in, as well as

[0141] For the matrix {C k}, including:

[0142] As mentioned above: For linear transceivers, Therefore, the following mainly derives the {C k According to formula (1.13), we can get:

[0143]

[0144] in, represent The i-th column of . From formula (51), we can see that [M k ] i,i Just depends on This means that the MSE of different symbols can be minimized independently. Let T k The Cholesky decomposition of You can get:

[0145]

[0146] Among them, Q k is a lower triangular matrix, is a unit diagonal lower triangular matrix, and has If [M k ] i,i The minimum value of can be reached if and only if the summation term on the right side of the second equal sign in equation (1.15) is zero, which requires:

[0147]

[0148] Then we can get the optimal expression:

[0149]

[0150] Substituting equation (1.16) into equation (1.13), the MSE matrix of THP can be rewritten as So we can get:

[0151]

[0152] For the matrix {P k Joint optimization of} and β ratios, including:

[0153] For a given {G k},{C k} and Φ, {P k The joint optimization problem of} and β can be described as:

[0154]

[0155] Using the following proposition, we can It is further simplified into a scalar optimization problem.

[0156] Proposition 1: For a given β, the problem The optimal precoding matrix satisfies the following structure:

[0157]

[0158] in, Depend on Front L k The eigenvectors corresponding to the large eigenvalues ​​are composed of z k,i is the transmit power allocated to the i-th symbol of the k-th subcarrier, is a unitary matrix, and makes the MSE matrix of THP The diagonal elements of are equal to:

[0159]

[0160] Among them, λ k,i yes The i-th largest eigenvalue of .

[0161] Proof: Given β, if the constraint C 2 is satisfied, then the problem It is converted into minimizing the sum MSE of the received signal under the condition of given transmit power. Substituting equation (1.19) into equation (1.18) yields:

[0162]

[0163] Further, substitute equation (1.19) and equation (1.21) into This results in an equivalent scalar optimization problem:

[0164]

[0165] Among them, F 0 is defined as:

[0166]

[0167] as well as

[0168] However It is still a non-convex optimization problem, and the SCA technique is used to transform it into an approximate convex form. Introduce auxiliary variables {s k,i} and {p k,i}, we can get the following equivalent problem:

[0169]

[0170] Where F is defined as:

[0171]

[0172] The convexity of F can be verified by showing that its Hessian matrix is ​​positive semidefinite. In addition, the problem The constraint condition C 3 , C 5 and C 6 is affine, C 7 is convex, the only non-convex term is C 8 To obtain the problem The approximate convex form can be used to 8 Perform a first-order Taylor expansion. For a convex function f(x), its first-order Taylor expansion is in represents the gradient of f(·). Therefore, we can get:

[0173]

[0174] in, and is the optimal solution for the nth iteration of the SCA method. It is a convex problem. For detailed proof, please refer to Example 2 below. Therefore, it can be solved using existing convex optimization toolboxes such as CVX.

[0175] According to the above analysis, the problem The solution process is shown in Algorithm 1:

[0176]

[0177] The optimization of the reflection coefficient matrix Φ includes:

[0178] For a given {P k}, {G k} and {C k}, the optimization problem of the reflection coefficient matrix Φ can be expressed as:

[0179]

[0180] It should be noted that the objective function and C 2 is non-convex with respect to Φ, and C 4 The unit module constraint of is also non-convex. To solve these difficulties, Decompose it into a series of subproblems and then use relaxation techniques to solve these subproblems.

[0181] Let r k,m and Respectively represent R k The mth column and T k The channel matrix of formula (1.1) can be rewritten into the following equivalent form:

[0182]

[0183] make Indicates T k P k The mth row of Further rewritten as

[0184]

[0185] Among them J k,m and K k,m is defined as

[0186]

[0187] The derivation in equation (1.25) applies It can be observed that φ m Did not appear in J k,m and K k,m Based on the above analysis and discarding some irrelevant items, It can be decomposed into a series of sub-problems. The mth reflection coefficient can be optimized by solving the following problems:

[0188]

[0189] in, Represents T k Pk C k The mth row of Now the question The only non-convex constraint is C 11 , C 11 Relaxation is |φ m |≤1 we can get:

[0190]

[0191] Apparently is a question about φ m is a convex problem and can be solved by CVX. For each iteration of the AO algorithm, we can solve a series of Complete the optimization of the reflection coefficient matrix. If the final optimized Φ ＊ Satisfy the constraint C 11 , which means that the relaxation is tight, which means that Φ ＊ yes A local optimal solution of ; if it is not satisfied, it is necessary to In addition, φ m , Normalization will not violate the EH constraint because the collected energy will increase with the increase of the reflection coefficient amplitude.

[0192] Based on the above analysis, the process of the AO algorithm is summarized in Algorithm 2:

[0193]

[0194]

[0195] Further, according to the system parameter settings in Table 1, the transceiver optimization algorithm of the multi-carrier MIMO wireless energy communication system assisted by the IRS proposed in this embodiment is used to perform simulation experiments, and all simulation results are obtained after 2000 independent channel realizations. In order to verify the effectiveness of the algorithm proposed in this embodiment, this embodiment also performs corresponding simulations on the schemes without IRS and random phase IRS, and serves as a control group.

[0196] Table 1 System simulation parameters

[0197]

[0198] Figure 3 shows the relationship between the average sum MSE performance and the transmit power, where N s =20, E th=-5dBm; It can be seen that with the increase of transmission power, the average sum MSE corresponding to all transceiver algorithms is constantly decreasing. Regardless of THP or linear transceiver, all schemes using IRS are better than those without IRS, and the scheme proposed in this embodiment can always achieve the minimum sum MSE performance. In addition, for the same transceiver algorithm, the sumMSE performance of THP is always better than that of linear transceiver. It is worth noting that using the transceiver design scheme proposed in this embodiment, the performance difference between THP and linear transceiver is more obvious than other schemes, which further verifies that THP has a more obvious performance advantage in the IRS-assisted MIMOSWIPT system.

[0199] Figure 4 shows the relationship between the average sum MSE performance and the number of reflection units, where P th =28dBm, E th =-5dBm; It can be seen that, regardless of THP or linear transceiver, the average sumMSE performance of the transceiver algorithm proposed in this embodiment is always better than other solutions. For the same transceiver algorithm, the performance of THP is always better than that of linear transceiver. At the same time, under the premise of using the transceiver solution proposed in this embodiment, as the number of reflection units increases, the performance of linear transceiver and THP is also increasing. This is because: more reflection units can improve the quality of the wireless channel, which is very beneficial to the communication performance of the MIMO system.

[0200] Figure 5 shows the relationship between the average achievable rate and the transmit power, where P th =28dBm, E th =-5dBm; It can be observed that the achievable rates of all schemes increase with the increase of transmit power. Regardless of whether it is a linear transceiver or THP, the scheme proposed in this embodiment can achieve optimal performance. For the same transceiver scheme, the performance of THP is always better than that of a linear transceiver. The reason is: the design goal of this embodiment is to minimize sumMSE rather than maximize the information rate. For THP, its optimal precoding matrix can not only minimize the sum MSE of each subcarrier, but also maximize the achievable rate of each subcarrier. For a linear transceiver, its optimal precoding matrix can only minimize the sumMSE of each subcarrier, which leads to the performance difference between a linear transceiver and THP.

[0201] The simulation results show that the AO algorithm proposed in this embodiment can achieve a smaller sumMSE than the traditional benchmark solution. In addition, whether it is sum MSE or average achievable rate, the performance of THP is always better than that of linear transceiver, which further proves the performance advantage of THP.

[0202] In this embodiment, a transceiver optimization scheme for applying nonlinear / linear transceivers to a multi-carrier MIMO wireless energy communication system assisted by a smart reflective surface is first designed, including a transceiver model, a transceiver optimization algorithm and simulation results. Then, a THP-based IRS-assisted multi-carrier MIMO is first modeled. SWIPT system, on this basis, only the feedback matrix of the transmitter of the system needs to be changed to an all-zero matrix to simplify it into a linear transceiver case; and the transmitter model is characterized in that the transmitter adopts THP technology, wherein the THP processing includes modulo operation, feedback matrix and precoding matrix, and when the feedback matrix is ​​an all-zero matrix, the transceiver model is equivalent to a linear transceiver model; the receiver model is characterized in that the receiver uses a power divider to divide the signal into two parts according to a certain ratio, one part is used for information decoding, and the other part is used for energy collection, wherein the energy collector is characterized by its nonlinear characteristics, which can better characterize the physical characteristics of the real energy collection model and has strong practical significance; this embodiment also jointly optimizes the transmit precoding matrix, transmit feedback matrix, receive equalization matrix, power allocation ratio and reflection coefficient matrix of the multi-carrier MIMOSWIPT system, so that the system minimizes the sum of the MSEs of the received signals under the premise of satisfying the given transmit power constraints and collection power constraints. Since the optimization problem established is highly non-convex, an alternating optimization algorithm is proposed for solving it, which mainly includes the following two steps: First, for a given IRS reflection coefficient matrix, the optimal closed-form expressions of the precoding matrix, feedback matrix and equalization matrix are used to convert the original optimization problem into a joint scalar optimization problem about power allocation and power division ratio, and then the problem is solved with the help of the successive convex approximation (SCA) technology; then, the optimization problem about the reflection coefficient matrix is ​​decomposed into a series of sub-problems using the precoding matrix, feedback matrix, equalization matrix and power allocation ratio obtained in the previous step. By solving these sub-problems in turn, the optimization of the reflection coefficient matrix is ​​completed. The alternating iteration of the above two steps can find the local optimal solution of the system design. The simulation results show that whether it is a nonlinear transceiver or a linear transceiver, compared with the schemes without IRS and random phase IRS, the transceiver design scheme proposed in this embodiment can significantly improve the sunMSE performance of the system.

[0203] Example 2

[0204] This example proves in detail that problem P5 is a convex problem as follows:

[0205] Due to the constraint C of problem P5 3 , C 5 and C 6 is affine, the constraint C 7 and C 9 Therefore, to prove that problem P5 is convex, we only need to prove that the Hessian matrix of the objective function F is semi-positive definite.

[0206] For THP, we can get So we only need to prove that the Hessian matrix is positive semidefinite, that is:

[0207]

[0208] in, It can be proved And have:

[0209]

[0210] therefore, It can be expressed as:

[0211]

[0212] in, t=1,…,L k .because So we only need to prove:

[0213]

[0214] Since the following always holds true:

[0215]

[0216] So equations (1.31) and (1.28) can be easily proved.

[0217] For a linear transceiver, the same method can be used to prove that is convex. Specifically, it can be shown that the Hessian matrix Among them have:

[0218]

[0219] Using (1.33) it is easy to get is a convex function.

[0220] Combining the above two situations, It is a convex problem.

[0221] Obviously, the above embodiments of the present invention are only examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the embodiments here. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A transceiver optimization method for an IRS-assisted multi-carrier MIMO wireless energy-carrying communication system, characterized in that: The following steps are involved: S1. Linear precoding or Tomlinson-Harashima nonlinear precoding of symbol vectors to be carried by different subcarriers to obtain a signal vector corresponding to a nonlinear transceiver or a linear transceiver; S2 based on the signal vector, establish a multi-carrier MIMO wireless communication system transceiver receiving signal; S3. Under the premise of satisfying the given transmit power constraint and collection power constraint, establish a transceiver optimization problem with minimizing the sum of the mean square error of the received signal as the objective function; S4. Solve the transceiver optimization problem to obtain a locally optimal transceiver optimization solution.

2. The transceiver optimization method of the IRS-assisted multi-carrier MIMO wireless energy-carrying communication system according to claim 1, characterized in that: The multi-carrier MIMO wireless energy communication system includes an N t A transmitter with N antennas r Receiver with N antennas and s An intelligent reflection surface IRS with passive reflection units; wherein the total bandwidth of the multi-carrier MIMO wireless energy-carrying communication system is W, the total bandwidth is evenly distributed to N subcarriers, and the calculation expression of the equivalent frequency domain channel matrix of the kth subcarrier is: in, represents the equivalent frequency domain channel matrix of the kth subcarrier, represents the equivalent frequency domain channel matrix of the k-th subcarrier transmitter-receiver link, represents the equivalent frequency domain channel matrix of the k-th subcarrier transmitter-intelligent reflecting surface IRS link, represents the equivalent frequency domain channel matrix of the k-th subcarrier intelligent reflecting surface IRS-receiver link; represents the reflection coefficient matrix of the intelligent reflective surface IRS, diag(.) represents the diagonal matrix, φ m represents the reflection coefficient of the mth passive reflection unit, φ m The adjustment range of the phase is [0,2π).

3. The transceiver optimization method of the IRS-assisted multi-carrier MIMO wireless energy-carrying communication system according to claim 2, characterized in that: The performing linear precoding or Tomlinson-Harashima nonlinear precoding processing on symbol vectors to be carried by different subcarriers includes: The symbol vector carried by the kth subcarrier is defined as L k ≤min{N t ,N r }, through the modulus operator MOD and the lower triangular feedback matrix The symbol vectors to be carried by different subcarriers are processed nonlinearly, and the calculation expression of the signal vector corresponding to the nonlinear transceiver is obtained as follows: Among them, v k (n) represents the signal vector v k The nth element in represents a complex vector, v k The real and imaginary parts of all elements are limited to The modulo operation expression is defined as: Among them, MOD M (.) represents the modulo operation, x represents the input of the modulo operation, represents the largest integer not exceeding z; In a nonlinear transceiver, v k The calculation expression is: Among them, C k =B k +I Lk Represents the lower triangular matrix with unit diagonal elements, u k =s k +i k Represents a valid signal vector; when B k = 0, the nonlinear transceiver is simplified to a linear transceiver, C k =I Lk , v k =u k =s k .

4. The transceiver optimization method of the IRS-assisted multi-carrier MIMO wireless energy-carrying communication system according to claim 3, characterized in that: The calculation expression for establishing the received signal of the transceiver of the multi-carrier MIMO wireless energy communication system based on the signal vector is: Among them, y k Indicates receiving signal, represents the frequency domain precoding matrix in the transmitter, represents the antenna noise vector, which is an additive white Gaussian noise vector.

5. The transceiver optimization method of the IRS-assisted multi-carrier MIMO wireless energy-carrying communication system according to claim 4, characterized in that: The receiver divides the received signal into a β-proportional part of the received signal power and a 1-β-proportional part of the received signal power through a power divider. The β-proportional part of the received signal power is used for energy collection, and the frequency domain signal sent to the energy collector is The calculation expression is: The 1-β ratio of the received signal power is used for information decoding, and the frequency domain signal sent to the information decoder is It is expressed as: in, represents the decoding noise vector, which is the additive white Gaussian noise vector introduced by the digital circuit in the information decoder.

6. The transceiver optimization method of the IRS-assisted multi-carrier MIMO wireless energy-carrying communication system according to claim 5, characterized in that: The transceiver optimization problem is: C3:0＜β＜1, in, represents the equalization matrix of the kth subcarrier, M k represents the MSE matrix of the kth subcarrier, Tr(.) represents the trace of the matrix, P th Represents the maximum transmit power of the system, E in represents the input power of the energy harvester, E th represents the energy harvesting power requirement of the system design, Indicates the average energy harvesting power threshold.

7. The transceiver optimization method of the IRS-assisted multi-carrier MIMO wireless energy-carrying communication system according to claim 6, characterized in that: The MSE matrix M of the k-th subcarrier k The calculation method is as follows: The frequency domain signal Equalization into balanced signal The calculation expression is: Among them, based on the equalized signal The MSE matrix of the kth subcarrier is calculated as Equalize the signal The calculation expression is converted into: in, The calculation expression is: in, represents the decoding noise variance, Ψ k (P k ,G k ,C k ) is calculated as: in, Indicates equivalent to .

8. The transceiver optimization method of the IRS-assisted multi-carrier MIMO wireless energy-carrying communication system according to claim 7, characterized in that: The energy harvester has an input power E in The calculation method is as follows: in, represents the antenna noise variance, and the superscript H represents the conjugate device of the matrix.

9. The transceiver optimization method of the IRS-assisted multi-carrier MIMO wireless energy-carrying communication system according to claim 8, characterized in that: The average energy harvesting power threshold The calculation method is as follows: Based on input power E in , calculate the harvested power γ(E in )for: Considering the energy harvesting requirements of the system design th , let the energy collection constraint be γ(E in )≥E th , the energy harvesting constraints are transformed into equivalent constraints as follows: in, The calculation expression is as follows: Among them, E m and E0 represent the saturation power and activation power of the energy harvester respectively, and τ and ν characterize the circuit characteristics of the energy harvester.

10. The transceiver optimization method of the IRS-assisted multi-carrier MIMO wireless energy-carrying communication system according to claim 9, characterized in that: In S4, an alternating optimization algorithm is used to solve the transceiver optimization problem, including: using an alternating method to solve {G k }、{C k }、{P k ,β} and Φ are optimized until the objective function converges, and the transceiver optimization solution is obtained.

Images