Qos-driven robust resource optimization method for ris-aided backscattering system

Abstract

The present application relates to a kind of QoS drive-based RIS auxiliary backscattering system robust resource optimization method, belong to backscattering communication field.It includes: constructing RIS auxiliary backscattering communication system model;Considering receiver time delay QoS demand, power station transmitting power constraint, backscattering node transmitting power constraint, backscattering node device throughput constraint, channel uncertainty constraint, RIS phase shift constraint and energy collection constraint, and based on channel uncertainty, establish the RIS auxiliary backscattering communication system robust resource allocation problem with total energy efficiency maximization as optimization target;The original problem is decomposed into equivalent beamforming, backscattering node transmitting power, RIS phase shift and time optimization sub-problem;And multiple convex optimization problems are solved using CVX, to obtain optimal resource allocation strategy.The present application can effectively improve system effective capacity while reducing communication interruption probability under the guarantee of information receiver different time delay QoS demand.

Description

Technical Field

[0001] The invention belongs to the field of backscatter communications and relates to a QoS-driven RIS-assisted backscatter system robust resource optimization method. Background Art

[0002] Backscatter communication is one of the new generation of green and low-cost IoT technologies, significantly reducing system energy consumption, enabling spectrum resource sharing, and increasing system capacity. However, at some point, the transmission performance of a backscatter system may degrade due to environmental obstacles. Therefore, reconfigurable smart surfaces (RISs) with the ability to reconfigure transmission paths have been introduced into backscatter communication systems. In a RIS-assisted backscatter communication system, backscatter nodes can transmit information to receivers via direct links and RIS-assisted indirect links. Furthermore, electromagnetic waves and wireless signals can be precisely transmitted to their destinations via reconfigurable channels, effectively improving the signal transmission quality and coverage performance of wireless networks.

[0003] Most existing RIS-assisted wireless power backscatter communication systems assume perfect channel state information at the transmitter and receiver ends. However, due to the influence of estimation errors, feedback delay, and dual-path loss in the actual physical channel, obtaining true channel state information is difficult. Furthermore, with the rapid development of latency-sensitive applications such as smart living and instant consumption, most existing systems and algorithms fail to consider the receiver's latency QoS requirements, leading to limited application scenarios and incompatibility issues. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a robust resource optimization method for a RIS-assisted backscatter system based on QoS. Taking into account the information receiver delay QoS requirements, the power station maximum transmission power constraints, the RIS phase shift constraints, the backscatter node transmission power constraints, the energy collection constraints, the transmission time constraints and the channel uncertainty constraints, a model is established for a RIS-assisted wireless power backscatter communication system with the optimization goal of maximizing the effective capacity of the system. The optimization problem with uncertain parameters is converted into a deterministic problem using the S process and variable substitution; the non-convex optimization problem is converted into an equivalent convex optimization problem using the continuous convex approximation method, the semi-definite relaxation method and the alternating optimization method; and the optimal resource allocation strategy is obtained using the CVX toolbox, eigenvalue decomposition and Gaussian randomization method.

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

[0006] A QoS-driven robust resource optimization method for a RIS-assisted backscatter system includes the following steps:

[0007] S1: Construct a RIS-assisted wireless power backscatter communication system model;

[0008] S2: Considering the receiver delay QoS requirements, power station transmit power constraints, backscatter node transmit power constraints, backscatter node device throughput constraints, channel uncertainty constraints, RIS phase shift constraints, and energy harvesting constraints, and based on channel uncertainty, a robust resource allocation problem for the RIS-assisted wireless power backscatter communication system is established with the total energy efficiency maximization as the optimization goal;

[0009] S3: Using continuous convex approximation, S-process, semidefinite relaxation method and alternating optimization method, the original problem is decomposed into equivalent beamforming, backscatter node transmit power, RIS phase shift and time optimization subproblems. CVX is used to solve multiple convex optimization problems and obtain the optimal resource allocation policy.

[0010] Furthermore, in step S1, the RIS-assisted backscatter system includes: at least one power station equipped with M antennas, K backscatter nodes equipped with energy harvesting circuits and active transmission circuits, a RIS with N reflective elements, and a single-antenna information receiver; the RIS is deployed between the backscatter node and the information receiver, and the signal transmission direction is changed by adjusting the phase shift of the RIS to thereby improve the channel gain;

[0011] The continuous time frame T is divided into K+1 time slots, namely {t0, t1,…, t k}, where t0 represents the energy collection phase, the power station sends energy signals for all backscattering nodes to collect energy, t k Indicates the active transmission phase of the kth backscatter node. The backscatter node sends signals to the receiver through the direct link and the RIS reflection link. The backscatter node set is defined as and RIS array element collection

[0012] Furthermore, in step S2, the total effective capacity of the system is expressed as:

[0013]

[0014] Among them, C k represents the effective capacity of the kth backscatter node:

[0015]

[0016] Where θ is the QoS requirement of the information receiver, t k represents the active transmission time of the kth backscattering node, B represents the bandwidth, and R k is the instantaneous rate of the kth backscattering node, which is expressed as:

[0017]

[0018] Among them, p k represents the transmission power of the kth backscattering node, ψ=[diag{Φ}1], represents the cascade channel between the kth backscatter node and the information receiver; is the RIS phase shift diagonal matrix, is the channel coefficient between the kth backscattering node and the information receiver, is the channel vector between RIS and the information receiver, is the channel vector between the kth backscatter node and RIS, (·) H is the conjugate transpose of the matrix.

[0019] Furthermore, in step S2, the robust resource allocation problem of the RIS-assisted wireless power backscatter communication system with guaranteed receiver delay QoS and the optimization goal of maximizing the total effective capacity of the system is expressed as:

[0020]

[0021] Among them, w k is the beamforming vector from the power station to the kth backscattering node, P max is the maximum transmission power threshold of the power station; represents the phase of the nth reflection unit of RIS, α k is the energy loss coefficient of the kth backscattering node, represents the energy collected by the kth backscattering node within t0, 0≤ρ k ≤1 represents the energy conversion efficiency of the kth backscattering node, E min 、E max They represent the lower limit of energy required for active transmission of the reflector node and the upper limit of battery capacity; t0 represents the energy collection time, T represents the frame length; Δh k represents the channel estimation error from the power station to the kth backscattering node, ΔH k represents the channel estimation error from the kth backscattering node to the receiver through the RIS;

[0022] C1 represents the maximum transmit power constraint of the power station; C2 represents the RIS phase shift constraint; C3 represents the transmit power constraint of the backscatter device; C4 represents the energy collection constraint; C5 represents the time constraint; C6 represents the channel uncertainty constraint; and in the channel uncertainty constraint, the channel uncertainty is expressed as:

[0023]

[0024] in, and respectively represent the set of channel uncertainty from the power station to the k-th backscattering node and the k-th backscattering node to the receiver via RIS, and respectively represent the estimated values of the channel from the power station to the k-th backscattering node and the k-th backscattering node to the receiver via RIS, δ k and respectively represent the upper bound of the estimation error of the channel from the power station to the k-th backscattering node and the k-th backscattering node to the receiver via RIS.

[0025] Further, in step S3, the S procedure is used to transform the uncertainty constraints C3 and C4 in the system effective capacity maximization robust resource allocation problem into deterministic linear matrix inequalities, where is expanded as

[0026]

[0027] where, and satisfy W k ≥ 0, Rank(W k ) = 1;

[0028] Combining the signal uncertainty sets and the constraints C3 and C4 are transformed into the following linear matrix inequalities:

[0029]

[0030]

[0031] where λ1, λ2, λ3 are the slack variables, I M denotes the M x M dimensional identity diagonal matrix,

[0032] By treating the uncertainty in the objective function with the continuous convex approximation method, the slack variable ζ k is introduced, and the objective function is relaxed as

[0033]

[0034] where the slack variable ζ k satisfies the constraint:

[0035]

[0036] Combining the signal uncertainty sets and the constraints are further transformed into the following linear matrix inequalities:

[0037]

[0038] where λ4is a slack variable, Ψ = ψ H ψ, and satisfies Ψ ≥ 0, Rank(Ψ) = 1;

[0039] Based on the above transformation, the following deterministic optimization problem is obtained:

[0040]

[0041]

[0042]

[0043] C7: Ψ ≥ 0, Rank(Ψ) = 1

[0044] C8: W k ≥ 0, Rank(W k ) = 1

[0045]

[0046] where, The objective function and the constraint condition of the optimization problem still have coupled variables, and it is still a non-convex optimization problem.

[0047] Further, by fixing {p k , Ψ, t0, t k} in the deterministic optimization problem, the following power station beamforming sub-optimization problem is obtained:

[0048]

[0049]

[0050]

[0051]

[0052] C8: W k ≥ 0, Rank(W k ) = 1

[0053]

[0054] The solution process of the power station beamforming sub-optimization problem is as follows: based on the semi-positive relaxation method, the rank-one constraint is first relaxed, and the upper bound of the above problem is obtained by using the CVX toolbox. If the obtained beamforming matrix satisfies Rank(W k ) = 1, then the beamforming vector w k can be obtained by eigenvalue decomposition; otherwise, the approximate solution is obtained by using the Gaussian randomization method.

[0055] Further, by fixing {w k ,Ψ,t0,t k} in the deterministic optimization problem, we obtain the following backscattering node transmit power sub-optimization problem:

[0056]

[0057]

[0058]

[0059]

[0060] where, The backscattering node transmit power sub-optimization problem is a convex optimization problem, which can be solved directly by using the CVX toolbox.

[0061] Further, by fixing {w k ,p k ,Ψ,t0,t k} in the deterministic optimization problem, we obtain the following RIS phase shift sub-optimization problem:

[0062]

[0063]

[0064]

[0065]

[0066] C8:Ψ≥0,Rank(Ψ)=1

[0067]

[0068] where,Ψ=[diag{Φ} 1] H [diag{Φ} 1];

[0069] The solution process of the phase shift sub-optimization problem is as follows: first, the Rank(Ψ)=1 constraint is relaxed by using the semi-definite relaxation method, then it is judged whether the solution obtained by using the CVX toolbox satisfies the rank-one constraint, if it satisfies, the phase shift matrix is obtained by using eigenvalue decomposition, otherwise, the approximate solution is obtained by using Gaussian randomization.

[0070] Further, by fixing {w k ,p k ,Ψ} in the deterministic optimization problem, we obtain the following time sub-optimization problem:

[0071]

[0072]

[0073]

[0074]

[0075]

[0076] where the time sub-optimization problem is a joint convex problem with respect to t0 and t k and is solved directly using the CVX toolbox.

[0077] The present application has the beneficial effects of:

[0078] The method considers the delay QoS requirement of the receiver, and dynamically allocates the energy collection and information transmission time of the backscattering node, so as to guarantee the delay QoS requirement of the receiver and ensure sufficient energy supply for the backscattering node to perform reliable information transmission.

[0079] The present application considers imperfect channel state information in the system design stage, so that the system has higher robustness and can perform information transmission under various channel conditions.

[0080] The method is a resource allocation method with the optimization target of maximizing the effective capacity considering the delay QoS requirement, and realizes the trade-off between system capacity and robustness.

[0081] Other advantages, objects, and features of the present application will be in part apparent and in part pointed out hereinafter in the specification, and will be observed in the practice of the present application. The objects and other advantages of the present application can be realized and attained by the structure particularly pointed out in the specification as follows. BRIEF DESCRIPTION OF DRAWINGS

[0082] In order to make the purposes, technical solutions and advantages of the present application clearer, the preferred embodiments of the present application will be described in detail below with reference to the accompanying drawings, in which:

[0083] Figure 1 is a whole flow chart of the QoS-driven RIS-assisted backscattering system robust resource optimization method of the present application;

[0084] Figure 2 is a structure schematic diagram of the RIS-assisted backscattering system of the present application;

[0085] Figure 3 is an effective capacity convergence diagram under an embodiment of the present application;

[0086] Figure 4 A robust graph for an embodiment of the present application. DETAILED DESCRIPTION

[0087] Other advantages and benefits of the present application will become apparent to those skilled in the art upon consideration of the disclosure or can be learned by practice of the application. The present application can be realized and achieved by means of the structures and combinations as specifically described herein and by equivalents thereof. Various modifications of the described modes of carrying out the application which are obvious to those skilled in the art are intended to be within the scope of the following claims. It is specifically intended that changes in form, alternative constructions, and equivalents come within the scope of the following claims. The following examples are provided by way of illustration and are not meant to be limiting of the present application.

[0088] The accompanying drawings are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the principles of the application. In the drawings:

[0089] The same or similar components in the drawings of the embodiments of the present application correspond to the same or similar components; in the description of the present application, it should be understood that if the terms "upper", "lower", "left", "right", "front", "back" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, therefore the terms describing the positional relationship in the drawings are only for illustrative purposes, and cannot be understood as a limitation of the present application, for those skilled in the art, the specific meanings of the above terms can be understood according to the specific circumstances.

[0090] Please refer to Figures 1 to 4 A QoS-driven RIS-aided backscattering system robust resource optimization method.

[0091] The present application provides a QoS-driven RIS-aided backscattering system robust resource optimization method, as shown in Figure 1 The method comprises the following steps:

[0092] S1: Constructing a RIS-aided wireless power supply backscattering communication system model;

[0093] S2: Considering the receiver delay QoS requirements, power station transmit power constraints, backscatter node transmit power constraints, backscatter node device throughput constraints, channel uncertainty constraints, RIS phase shift constraints and energy harvesting constraints, a robust resource allocation model for a dual-RIS-assisted wireless power backscatter communication system is established based on channel uncertainty with the optimization goal of maximizing total energy efficiency.

[0094] S3: The original problem is decomposed into multiple equivalent convex optimization subproblems using continuous convex approximation, S-process, semidefinite relaxation method, and alternating optimization method. CVX is then used to solve these multiple convex optimization problems to obtain the optimal beamforming vector, backscatter node transmit power, RIS phase shift, and time allocation factor, i.e., the optimal resource allocation policy.

[0095] In step S1 of this embodiment, Figure 2 As shown in Figure 1, the system consists of a power station equipped with M antennas, K single-antenna passive backscatter nodes equipped with energy harvesting circuits and active transmission circuits, a RIS with N reflective array elements, and a single-antenna information receiver. The continuous time frame T is divided into K+1 time slots, namely {t0, t1, …, t k}, where t0 represents the energy collection phase, the power station sends energy signals for all backscattering nodes to collect energy, t k Indicates the active transmission phase of the kth backscatter node, the backscatter node sends signals to the receiver through the direct link and the RIS reflection link. Define the backscatter node set and RIS array element collection

[0096] In the energy collection phase, the signal received by the kth backscattering node is:

[0097]

[0098] in, represents the channel vector between the power station and the kth backscattering node, and s k represents the beamforming vector and symbol sent by the power station to the kth backscattering node, and satisfies E[|s k | 2 ]=1,n k Indicates that at the kth backscattering node, the mean is 0 and the variance is Additive white Gaussian noise.

[0099] According to the linear energy collection model, the energy collected by the kth backscattering node at time slot t0 is:

[0100]

[0101] where 0≤ρ k ≤1 denotes the energy conversion efficiency of the k-th backscatter node.

[0102] In the active transmission phase, the information receiver receives the signal at time t k

[0103]

[0104] where p k denotes the transmit power of the k-th backscatter node, satisfying α k ∈(0,1) is the energy loss factor, is the channel coefficient between the k-th backscatter node and the information receiver, is the channel vector between the RIS and the information receiver, is the phase-shift diagonal matrix of the RIS, is the channel vector between the k-th backscatter node and the RIS, x k denotes the information symbol transmitted by the k-th backscatter node, satisfying n r,k denotes the additive white Gaussian noise (AWGN) at the information receiver with mean 0 and variance

[0105] The instantaneous rate of the k-th backscatter node is denoted as

[0106]

[0107] where B denotes the bandwidth, ψ = [diag{Φ}1], denotes the cascade channel between the k-th backscatter node and the information receiver.

[0108] According to the effective capacity theory, assuming that the queue service process satisfies the theorem, the effective capacity of the k-th backscatter node can be denoted as

[0109]

[0110] where θ is the QoS index, and its size represents the delay tolerance of the information receiver. Then the total effective capacity of the system can be denoted as

[0111]

[0112] ​​In step S2, this embodiment is based on the channel uncertainty set, considers the power station transmit power constraint, information receiver delay QoS requirement, backscatter node transmit power constraint, RIS phase shift constraint, time allocation constraint and energy collection constraint, and maximizes the system effective capacity by jointly optimizing the power station transmit beamforming vector, backscatter node transmit power, RIS phase shift, backscatter node energy collection time and active transmission time.

[0113] Since feedback delay and dual path loss lead to channel uncertainty, the channel uncertainty factor is taken into account in advance in the above optimization problem. According to robust optimization theory, the channel uncertainty problem can be described as:

[0114]

[0115] in, and represents the channel uncertainty set, and represents the channel estimation value, ΔH k and Δh k represents the channel estimation error, δ k and represents the upper bound of the estimation error.

[0116] Therefore, the robust resource allocation problem of maximizing the system effective capacity can be modeled as:

[0117]

[0118] Among them, C1 represents the maximum transmit power constraint of the power station; C2 represents the RIS phase shift constraint; C3 represents the backscatter device transmit power constraint; C4 represents the energy collection constraint; C5 represents the time constraint; and C6 represents the channel uncertainty constraint.

[0119] In step S3, the uncertain constraints C3 and C4 in the robust resource allocation problem for maximizing the system effective capacity can be transformed into deterministic linear matrix inequalities using the S procedure. First, expand (2) as follows:

[0120]

[0121] in, And satisfy W k ≥0,Rank(W k )=1.

[0122] Then, combined with (7), constraints C3 and C4 can be transformed into the following linear matrix inequality

[0123]

[0124] where λ1, λ2, λ3 are the slack variables, I M denotes the M x M dimensional identity matrix,

[0125] To handle the uncertainty in the objective function, the slack variables ζ k are introduced by the continuous convex approximation method, and the relaxed objective function is

[0126]

[0127] where the slack variables ζ k satisfy the constraints

[0128]

[0129] Combining (7) and (12), (12) can be transformed into the following linear matrix inequality

[0130]

[0131] where λ4 is the slack variable, Ψ = ψ H , and satisfies Ψ ≥ 0, Rank(Ψ) = 1.

[0132] Based on the above transformation, the following deterministic optimization problem can be obtained:

[0133]

[0134] where The optimization problem (14) is still a non-convex optimization problem, and there are still coupled variables in the objective function and the constraint conditions, so it is still difficult to solve.

[0135] Based on the research on the alternating optimization theory, the optimization problem (14) is decomposed into four sub-problems: 1) power station beamforming optimization problem; 2) backscattering node transmit power optimization problem; 3) RIS phase shift optimization problem; 4) time optimization problem.

[0136] By fixing {p k , Ψ, t0, t k}, the following power station beamforming sub-optimization problem can be obtained:

[0137]

[0138] Due to the existence of the constraint Rank(W k ) = 1, the optimization problem (15) is a non-convex problem. Based on the semi-definite relaxation method, the rank-one constraint is first relaxed, and the upper bound of the above problem is obtained by using the CVX toolbox. If the obtained beamforming matrix satisfies Rank(Wk )=1, the beamforming vector w can be obtained by eigenvalue decomposition k ; Otherwise, the Gaussian randomization method is used to obtain an approximate solution.

[0139] By fixing {w k ,Ψ,t0,t k}, we can get the following backscatter node transmit power sub-optimization problem:

[0140]

[0141] in, The optimization problem (16) is a convex optimization problem and can be solved directly using the CVX toolbox.

[0142] By fixing {ω k ,p k ,t0,t k}, we can get the following RIS phase shift sub-optimization problem:

[0143]

[0144] where Ψ = [diag{Φ}1] H [diag{Φ}1], similar to the optimization problem (15), first use the semi-definite relaxation method to relax Rank(Ψ) = 1, and then determine whether the solution obtained using the CVX toolbox satisfies the rank-one constraint. If so, the phase shift matrix is ​​obtained using eigenvalue decomposition, otherwise an approximate solution is obtained using Gaussian randomization.

[0145] By fixing {ω k ,p k ,Ψ}, we can get the following time sub-optimization problem:

[0146]

[0147] Among them, problem (18) is a question about t0 and t k The joint convex problem of can be solved using the CVX toolbox.

[0148] The application effect of the present invention is described in detail below with reference to simulation.

[0149] 1) Simulation conditions

[0150] This section demonstrates the convergence and effectiveness of the proposed method through simulation results. Assume that the system has a power station located at (0,0), a RIS at (5,0), an information receiver at (8,0), and backscatter nodes randomly distributed within a circle with a radius of 2 meters centered at (5,0). The channel model uses the Nakagami-m small-scale fading model with m = 2. Other simulation parameters are shown in Table 1.

[0151] Table 1

[0152]

[0153] 2) Simulation results

[0154] In this embodiment, Figure 3 The effective capacity convergence diagram of the iterative method in this example is given. Figure 4 The robustness diagram of the iterative method of this example is given. Figure 3 It is shown that the method of the present invention can quickly achieve convergence under different delay QoS requirements of the information receiver, thereby proving that the method of the present invention can well guarantee the communication quality of the system and has real-time performance. Figure 4 Display along with channel error As the channel error increases, the satisfaction probability (1-interruption probability) of the method of the present invention decreases slowly and always remains above 92%, while the satisfaction probability of other methods decreases with the increase of channel error. It decreases rapidly with the increase of , which proves that the method of the present invention has strong robustness. Figure 3 and Figure 4 The experimental results show that the method of the present invention not only guarantees the receiver's delay QoS requirements, but also ensures real-time performance and high robustness.

[0155] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solutions, which should all be included in the scope of the claims of the present invention.

Claims

1. A QoS-driven RIS-assisted backscatter system robust resource optimization method, characterized by: It includes the following steps: S1: Construct a RIS-assisted wireless power backscatter communication system model; S2: Considering the receiver delay QoS requirements, power station transmit power constraints, backscatter node transmit power constraints, backscatter node device throughput constraints, channel uncertainty constraints, RIS phase shift constraints, and energy harvesting constraints, and based on channel uncertainty, a robust resource allocation problem for the RIS-assisted wireless power backscatter communication system is established with the total energy efficiency maximization as the optimization goal; S3: Using continuous convex approximation, S-process, semi-definite relaxation method and alternating optimization method, the original problem is decomposed into equivalent beamforming, backscattering node transmit power, RIS phase shift and time optimization sub-problems; And use CVX to solve multiple convex optimization problems to obtain the optimal resource allocation strategy; In step S2, the total effective capacity of the system is expressed as: Among them, C k represents the effective capacity of the kth backscatter node: Where θ is the QoS requirement of the information receiver, t k represents the active transmission time of the kth backscattering node, B represents the bandwidth, and R k is the instantaneous rate of the kth backscattering node, which is expressed as: Among them, p k represents the transmission power of the kth backscattering node, ψ=[diag{Φ}1], represents the cascade channel between the kth backscatter node and the information receiver; is the RIS phase shift diagonal matrix, is the channel coefficient between the kth backscattering node and the information receiver, is the channel vector between RIS and the information receiver, is the channel vector between the kth backscatter node and RIS, (·) H is the conjugate transpose of the matrix; In step S2, the robust resource allocation problem of the RIS-assisted wireless power backscatter communication system with guaranteed receiver delay QoS and the optimization goal of maximizing the total effective capacity of the system is expressed as: s.t.C1:||w k || 2 ≤P max Among them, w k is the beamforming vector from the power station to the kth backscattering node, P max is the maximum transmission power threshold of the power station; represents the phase of the nth reflection unit of RIS, α k is the energy loss coefficient of the kth backscattering node, represents the energy collected by the kth backscattering node within t0, 0≤ρ k ≤1 represents the energy conversion efficiency of the kth backscattering node, E min 、E max They represent the lower limit of energy required for active transmission of the reflector node and the upper limit of battery capacity; t0 represents the energy collection time, T represents the frame length; Δh k represents the channel estimation error from the power station to the kth backscattering node, ΔH k represents the channel estimation error from the kth backscattering node to the receiver through the RIS; C1 represents the maximum transmit power constraint of the power station; C2 represents the RIS phase shift constraint; C3 represents the transmit power constraint of the backscatter device; C4 represents the energy collection constraint; C5 represents the time constraint; C6 represents the channel uncertainty constraint; and in the channel uncertainty constraint, the channel uncertainty is expressed as: in, and They represent the channel uncertainty sets from the power station to the kth backscatter node and from the kth backscatter node to the receiver via the RIS, and represents the channel estimation value from the power station to the kth backscatter node and from the kth backscatter node to the receiver through the RIS, δ k and It represents the upper bound of the estimation error from the power station to the kth backscattering node and from the kth backscattering node to the receiver through the RIS.

2. The QoS-driven RIS-assisted backscatter system robust resource optimization method according to claim 1, characterized in that: In step S1, the RIS-assisted backscatter system includes: at least one power station equipped with M antennas, K backscatter nodes equipped with energy harvesting circuits and active transmission circuits, a RIS with N reflective elements, and a single-antenna information receiver. The RIS is deployed between the backscatter node and the information receiver, and the signal transmission direction is changed by adjusting the phase shift of the RIS to improve the channel gain. The continuous time frame T is divided into K+1 time slots, namely {t0, t1,…, t k }, where t0 represents the energy collection phase, the power station sends energy signals for all backscattering nodes to collect energy, t k Indicates the active transmission phase of the kth backscatter node. The backscatter node sends signals to the receiver through the direct link and the RIS reflection link. The backscatter node set is defined as and RIS array element collection 3. The QoS-driven RIS-assisted backscatter system robust resource optimization method according to claim 1, characterized in that: In step S3, the uncertainty constraints C3 and C4 in the robust resource allocation problem of maximizing the system effective capacity are transformed into deterministic linear matrix inequalities using the S process, where The expanded representation is: in, And satisfy W k ≥0,Rank(W k )=1; Combined signal uncertainty set and Constraints C3 and C4 are transformed into the following linear matrix inequalities: Among them, λ1, λ2, λ3 are slack variables, I M represents the M×M dimensional unit diagonal matrix, The uncertainty in the objective function is handled by the continuous convex approximation method, and the slack variable ζ is introduced. k , the objective function is relaxed to: Among them, the slack variable ζ k Satisfy the constraints: Combined signal uncertainty set and The constraints are further transformed into the following linear matrix inequalities: Among them, λ4 is the slack variable, Ψ=ψ H ψ, and satisfy Ψ≥0, Rank(Ψ)=1; Based on the above transformation, the following deterministic optimization problem is obtained: C2, C5 C7:Ψ≥0,Rank(Ψ)=1 C8:W k ≥0,Rank(W k )=1 in, There are still coupled variables in the objective function and constraints of this optimization problem, and it is still a non-convex optimization problem.

4. The QoS-driven RIS-assisted backscatter system robust resource optimization method according to claim 3, characterized in that: By fixing {p k ,Ψ,t0,t k }, we get the following power station beamforming sub-optimization problem: C8:W k ≥0,Rank(W k )=1 The solution process of the power station beamforming sub-optimization problem is as follows: based on the semi-positive relaxation method, the rank-one constraint is first relaxed, and the upper bound of the above problem is obtained by using the CVX toolbox. For the obtained beamforming matrix, if Rank(W k )=1, the beamforming vector w can be obtained by eigenvalue decomposition k ; Otherwise, the Gaussian randomization method is used to obtain an approximate solution.

5. The QoS-driven RIS-assisted backscatter system robust resource optimization method according to claim 3, characterized in that: By fixing {w k ,Ψ,t0,t k }, we get the following backscatter node transmit power sub-optimization problem: in, The backscatter node transmit power sub-optimization problem is a convex optimization problem that is directly solved using the CVX toolbox.

6. The QoS-driven RIS-assisted backscatter system robust resource optimization method according to claim 3, characterized in that: By fixing {ω k ,p k ,t0,t k }, we get the following RIS phase shift sub-optimization problem: C8:Ψ≥0,Rank(Ψ)=1 where, Ψ = [diag{Φ}1] H [diag{Φ}1]; The solution process of the phase shift sub-optimization problem is as follows: first, Rank(Ψ)=1 is relaxed using the semi-definite relaxation method, and then it is determined whether the solution obtained using the CVX toolbox satisfies the rank-one constraint. If so, the phase shift matrix is ​​obtained using eigenvalue decomposition, otherwise an approximate solution is obtained using Gaussian randomization.

7. The QoS-driven RIS-assisted backscatter system robust resource optimization method according to claim 3, characterized in that: By fixing {ω k ,p k ,Ψ}, we get the following time sub-optimization problem: Among them, the time sub-optimization problem is about t0 and t k The joint convex problem of is solved directly using the CVX toolbox.

Images