Communication method, device and system based on reconfigurable intelligent surface

Abstract

The invention discloses a communication method based on a reconfigurable intelligent surface, a communication device based on the reconfigurable intelligent surface and a communication system based on the reconfigurable intelligent surface, and belongs to the technical field of wireless communication. Jointly optimizing a current beam forming vector and a current RIS reflection parameter of the sender according to a channel estimation result obtained in real time through a depth deterministic strategy gradient algorithm; a joint optimization model of beam forming and RIS parameters of a sender is established, a non-convex optimization problem is converted into a Markov decision process, efficient solution is realized through a depth deterministic policy gradient algorithm, and the transmission capability of communication can be improved. Furthermore, a reconfigurable intelligent surface technology is introduced, a reflection unit with independent amplitude-phase regulation and control capability is designed, the amplitude fixed limitation of a traditional passive RIS is broken through, double-fading path loss can be effectively overcome, and then the anti-interference performance of a communication system in complex electromagnetic environments such as high spectral density and high interference power can be improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless communications, and more specifically, relates to a communication method, device and system based on a reconfigurable smart surface. Background Art

[0002] With the rapid development of mobile communications technology, challenges in communication security and privacy are becoming key issues that the next generation of wireless mobile communications (6th Generation Mobile Networks, or 6G) systems aim to address. Traditional solutions proposed by the industry include frequency hopping, spread spectrum technology, and multi-antenna techniques.

[0003] Frequency hopping technology achieves interference mitigation by rapidly changing the carrier frequency of the transmitted signal, making it difficult for interferers to predict and track. However, this method places high demands on the management and allocation of frequency resources and is limited in high-density spectrum environments. Spread spectrum technology spreads the signal over a wider frequency band, dispersing the energy of the interfering signal over a wider frequency band, thereby reducing the impact of interference. Although spread spectrum technology excels at interference mitigation, it has a low transmission rate and can fail in the presence of high-power interference. Multi-antenna technology utilizes multiple antennas for signal transmission and reception, improving the system's interference mitigation capabilities through spatial diversity and beamforming. However, multi-antenna systems require high hardware and computing resources and are susceptible to multipath effects in complex electromagnetic environments.

[0004] That is, the existing communication system has poor anti-interference performance in complex electromagnetic environments such as high spectrum density and high interference power. Summary of the Invention

[0005] In response to the above-mentioned defects or improvement needs of the prior art, the present invention provides a communication method, device and system based on a reconfigurable smart surface, the purpose of which is to solve the technical problem of poor anti-interference performance of existing communication systems in complex electromagnetic environments such as high spectrum density and high interference power.

[0006] To achieve the above objectives, according to one aspect of the present invention, a communication method based on a reconfigurable smart surface is provided, comprising:

[0007] S1: Set the initial reflection coefficient matrix of the reconfigurable smart surface RIS and the initial beamforming vector of the transmitter;

[0008] S2: Control the sender to send a first communication signal via a first direct link and a first RIS auxiliary link corresponding to the initial reflection coefficient matrix using the initial beamforming vector; the interference source to send a first interference signal via a second direct link and a second RIS auxiliary link corresponding to the initial reflection coefficient matrix; and use the receiver to measure, in real time, the signal-to-interference-and-noise ratio of the received signal corresponding to the first communication signal and the first interference signal;

[0009] S3: Jointly optimizing the sender's current beamforming vector and current RIS reflection parameters based on the real-time channel estimation results using a deep deterministic policy gradient algorithm; the deep deterministic policy gradient algorithm includes: a reward function centered on maximizing the signal-to-interference-and-noise ratio of the received signal; a Markov decision process state space containing channel state information, power constraints, and historical performance indicators; and a continuous action space containing beamforming vectors, RIS phase offsets, and amplitude gains;

[0010] S4: Control the sender to send a second communication signal via the first direct link and a third RIS auxiliary link corresponding to the current reflection coefficient matrix using the initial beamforming vector; the interference source sends a second interference signal via the second direct link and a fourth RIS auxiliary link corresponding to the current reflection coefficient matrix; and use the receiver to measure the signal to interference and noise ratio of the received signal corresponding to the second communication signal and the second interference signal in real time;

[0011] S5: When the adjustment period has not been reached, directly proceed to S3; when the adjustment period has been reached, dynamically update the reflection coefficient matrix obtained in S4 according to the channel change rate, and then proceed to S3.

[0012] Furthermore, the reward function r t Designed for: t =10log 10 (γ t )-λ1Δp t -λ2D t ; Among them, γ t is the signal-to-interference-noise ratio of the received signal at time t, ΔP t is the power constraint violation at time t, D t is the service quality deviation at time t, λ1 and λ2 are two penalty coefficients.

[0013] Furthermore, the state space s t Means: s t =[vec(G l ),h j ,h,P t ,P j ,γ t-1 ]; the action space a tIndicates: a t =[w t ,θ1,…,θ N ,β1,…,β N ]; Among them, vec(G l ) is the vector corresponding to the channel matrix of the sender-RIS, h j is the channel matrix from the interference source to the RIS, h is the channel matrix from the RIS to the receiver, P t is the transmit power of the sender, P j is the transmission power of the interference source, γ t-1 is the signal-to-interference-noise ratio of the received signal at time t-1, w t is the square beamforming vector sent at time t, θ1,…,θ N is the phase shift of each unit of RIS at time t, β1,…,β N is the amplitude change of RIS units at time t.

[0014] Furthermore, the optimization process of the current RIS reflection parameters in the deep deterministic policy gradient algorithm is: first optimizing the amplitude gain parameters based on the gradient descent method, and then implementing the amplitude-phase joint optimization under the power constraint condition.

[0015] Furthermore, the amplitude-phase joint optimization satisfies the following constraints: Where N is the total number of reflection units, β n ∈[1,β max ] is the amplitude gain coefficient of the nth reflection unit, is the incident power of the nth reflector unit, P c is the circuit power consumption, is the maximum available power of RIS.

[0016] Furthermore, the re-implementation of the amplitude-phase joint optimization under the power constraint condition includes: taking the current optimal phase A dynamic search window is established as the center; the window width Δθ is adjusted according to the change direction of the signal to interference noise ratio of adjacent subframes to obtain the interval When the window width is less than π / 12, a local fine search is triggered.

[0017] Furthermore, when the adjustment period is reached, the reflection coefficient matrix obtained by dynamically updating S4 according to the channel change rate includes: pre-correcting the phase deviation caused by Doppler frequency shift; dynamically adjusting the gain coefficient in the reflection coefficient matrix according to the received signal strength; and for non-cooperative interference, the transmitting beam actively detects channel state information.

[0018] According to another aspect of the present invention, there is provided a communication device based on a reconfigurable smart surface, comprising:

[0019] An initial setting module is used to set the initial reflection coefficient matrix of the reconfigurable smart surface RIS and the initial beamforming vector of the transmitter;

[0020] A first control module is configured to control the sender to send a first communication signal via a first direct link and a first RIS auxiliary link corresponding to the initial reflection coefficient matrix using the initial beamforming vector; the interference source to send a first interference signal via a second direct link and a second RIS auxiliary link corresponding to the initial reflection coefficient matrix; and to use the receiver to measure in real time the signal-to-interference-and-noise ratio of the received signal corresponding to the first communication signal and the first interference signal;

[0021] a loop optimization module for jointly optimizing the current beamforming vector and the current RIS reflection parameter of the transmitter based on the channel estimation results obtained in real time using a deep deterministic policy gradient algorithm; the deep deterministic policy gradient algorithm includes: a reward function centered on maximizing the signal-to-interference-and-noise ratio of the received signal; a Markov decision process state space containing channel state information, power constraints, and historical performance indicators; and a continuous action space containing the beamforming vector, RIS phase offset, and amplitude gain;

[0022] A second control module is configured to control the sender to send a second communication signal via the first direct link and a third RIS auxiliary link corresponding to the current reflection coefficient matrix using the initial beamforming vector; the interference source to send a second interference signal via the second direct link and a fourth RIS auxiliary link corresponding to the current reflection coefficient matrix; and to use the receiver to measure in real time the signal-to-interference-and-noise ratio of the received signal corresponding to the second communication signal and the second interference signal;

[0023] An iterative judgment module is used to directly enter the loop optimization module when the adjustment period has not been reached; when the adjustment period is reached, the reflection coefficient matrix output by the second control module is dynamically updated according to the channel change rate, and then enter the loop optimization module.

[0024] According to another aspect of the present invention, a communication system based on a reconfigurable smart surface is provided, comprising: a sender, a reconfigurable smart surface (RIS), an interference source, a receiver, a memory, and a processor, wherein the sender sends a communication signal to the receiver via its corresponding RIS auxiliary link and direct link, and simultaneously the interference source sends an interference signal to the receiver via its corresponding RIS auxiliary link and direct link; the receiver measures, in real time, the signal-to-interference-and-noise ratio of the communication signal and the received signal corresponding to the interference signal; the memory stores a computer program, and when the processor executes the computer program, the steps of the communication method based on the reconfigurable smart surface are implemented.

[0025] According to another aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the communication method based on the reconfigurable smart surface are implemented.

[0026] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects compared with the prior art:

[0027] The present invention provides a communication method based on a reconfigurable smart surface. This method uses a deep deterministic policy gradient algorithm to jointly optimize the sender's current beamforming vector and current RIS reflection parameters based on real-time channel estimation results. A joint optimization model for the sender's beamforming and RIS parameters is established, transforming the non-convex optimization problem into a Markov decision process. This method achieves an efficient solution using a deep deterministic policy gradient algorithm, thereby improving communication transmission capacity. Furthermore, the reconfigurable smart surface technology is introduced, with the design featuring reflective units with independent amplitude-phase control capabilities, overcoming the fixed amplitude limitations of traditional passive RIS. Each reflective unit has an adjustable gain range of 1 to 3 times, effectively overcoming double-fading path loss and improving the communication system's anti-interference performance in complex electromagnetic environments such as those with high spectral density and high interference power. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 Flowchart of the communication method based on the reconfigurable smart surface provided in Example 1 of the present invention;

[0029] Figure 2 A schematic diagram of the results of the reconfigurable smart surface-assisted communication system provided in Example 1 of the present invention;

[0030] Figure 3 Flowchart of the heterogeneous resource collaborative scheduling method based on fog node collaboration provided in Example 1 of the present invention;

[0031] Figure 4 A flowchart of the ADMM-based cache strategy formulation provided in Example 1 of the present invention;

[0032] Figure 5 Schematic diagram of a heterogeneous resource collaborative scheduling system based on fog node collaboration provided in Example 1 of the present invention;

[0033] Figure 6 A schematic diagram of a heterogeneous resource collaborative scheduling system based on fog node collaboration providing services to users provided in Example 1 of the present invention;

[0034] Figure 7 This is a simulation diagram of the convergence performance of the ADMM-based cache algorithm provided in Example 1 of the present invention.

[0035] Figure 8This is a graph showing the relationship between the optimal average delay and network heterogeneous resources provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0036] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely for the purpose of explaining the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0037] Example 1

[0038] This embodiment provides a communication method based on a reconfigurable smart surface, such as Figure 1 As shown, it includes: S1-S5. The communication method is applied to a RIS-assisted communication system, such as Figure 2 As shown in Figure 1, it includes a transmitter, a RIS, an interference source, a receiver, a memory, and a processor. The RIS can be active or passive. Taking an active RIS as an example, this system consists of N independently controllable reflector units, each with dual phase and amplitude adjustment capabilities. The active RIS reflection model: Reflector units with independent amplitude and phase control capabilities are designed to overcome the fixed amplitude limitation of traditional passive RIS. The gain of each reflector unit can be adjusted from 1 to 3 times, effectively overcoming the dual-path fading loss of traditional passive RIS. Furthermore, the active RIS, a reconfigurable smart surface with integrated active amplification circuits, overcomes the dual-path fading flaw of traditional passive RIS by independently controlling the phase (θ∈[0,2π)) and amplitude (β∈[1,3]) of the reflected signal. This provides signal amplification capabilities and can dynamically adapt to high-interference environments. The transmitter transmits a communication signal to the receiver via its corresponding RIS auxiliary link and direct link. Simultaneously, the interference source transmits an interference signal to the receiver via its corresponding RIS auxiliary link and direct link. The receiver measures the signal-to-interference-noise ratio (SIR) of the received signal corresponding to the communication signal and the interference signal in real time. Its advantage lies in the intelligent anti-interference mechanism: building a closed-loop system including interference feature extraction, reflection beam forming, and signal cancellation to achieve real-time perception and dynamic cancellation of interference signals.

[0039] S1: Set the initial reflection coefficient matrix of the reconfigurable smart surface (RIS) and the initial beamforming vector of the transmitter. Specifically, initialize the reflection coefficient matrix of the active RIS, set the initial phase of each reflector to be uniformly distributed in the interval [0, 2π], and set the initial amplitude value to the preset maximum gain. Furthermore, the reflection coefficient matrix Φ of the active RIS is expressed as: where β n ∈[1,β max ] is the amplitude gain coefficient of the nth reflection unit, θ n∈[0,2π) is the phase offset, β max The preset maximum gain value.

[0040] S2: Control the sender to use the initial beamforming vector to send the first communication signal via the first direct link and the first RIS auxiliary link corresponding to the initial reflection coefficient matrix; the interference source sends the first interference signal via the second direct link and the second RIS auxiliary link corresponding to the initial reflection coefficient matrix; use the receiver to measure the signal-to-interference-and-noise ratio of the received signal corresponding to the first communication signal and the first interference signal in real time.

[0041] S3: Jointly optimize the sender's current beamforming vector and current RIS reflection parameters based on real-time channel estimation results using a deep deterministic policy gradient algorithm. The deep deterministic policy gradient algorithm includes: a reward function centered on maximizing the signal-to-interference-and-noise ratio of the received signal; a Markov decision process state space containing channel state information, power constraints, and historical performance indicators; and a continuous action space containing beamforming vectors, RIS phase offsets, and amplitude gains.

[0042] Among them, Deep Deterministic Policy Gradient (DDPG), a reinforcement learning algorithm for continuous action spaces, consists of a dual actor-critic network. The actor network outputs deterministic policies (such as RIS phase / amplitude parameters), while the critic network evaluates Q-values. Through experience replay and soft update mechanisms, it achieves real-time optimization in dynamic channel environments, significantly improving convergence speed compared to traditional algorithms.

[0043] Specifically, the deep deterministic policy gradient algorithm adopts a dual-network architecture, consisting of a policy network that takes as input a state vector of (M+3N+2) dimensions and outputs an action vector of (M+2N) dimensions; and a comment network that takes as input a joint state-action vector of (M+3N+2+M+2N) dimensions and outputs a Q-value evaluation. M is the number of transmitter antennas, and N is the number of RIS reflection units. Furthermore, parameter adaptation is achieved through a target network soft-update mechanism, including setting up an experience replay buffer to store historical state transition data, using Ornstein-Uhlenbeck noise to enhance exploration capabilities, and performing periodic parameter synchronization to ensure algorithm stability.

[0044] S4: Control the sender to use the initial beamforming vector to send the second communication signal via the first direct link and the third RIS auxiliary link corresponding to the current reflection coefficient matrix; the interference source sends the second interference signal via the second direct link and the fourth RIS auxiliary link corresponding to the current reflection coefficient matrix; use the receiver to measure the signal-to-interference-and-noise ratio of the received signal corresponding to the second communication signal and the second interference signal in real time.

[0045] S5: If the adjustment period has not been reached, directly proceed to S3; if the adjustment period is reached, dynamically update the reflection coefficient matrix obtained in S4 according to the channel change rate, and then proceed to S3. Specifically, the reflection coefficient matrix output by the joint optimization module can be dynamically updated according to the channel change rate ρ, and the adjustment period T satisfies:

[0046] Furthermore, the following algorithm can be used to determine the key performance indicators of the communication method, including: Anti-interference gain: Δγ = γ RIS -γ direct Power efficiency: Convergence time: The time required to reach 95% of the maximum SINR.

[0047] Furthermore, the reward function r t Designed for: t =10log 10 (γ t )-λ1ΔP t -λ2D t ; γ t is the signal-to-interference-noise ratio of the received signal at time t, ΔP t is the power constraint violation at time t, D t is the service quality deviation at time t, λ1,λ2 are two penalty coefficients. t Means: s t =[vec(G l ),h j ,h,P t ,P j ,γ t-1 ]; action space a t Indicates: a t =[w t ,θ1,…,θ N ,β1,…,β N ]. Among them, vec(G l ) is the vector corresponding to the channel matrix of the sender-RIS, h j is the channel matrix from the interference source to the RIS, h is the channel matrix from the RIS to the receiver, P t is the transmit power of the sender, P j is the transmission power of the interference source, γ t-1 is the signal-to-interference-noise ratio of the received signal at time t-1, w t is the square beamforming vector sent at time t, θ1,…,θ N is the phase shift of each unit of RIS at time t, β1,…,β N is the amplitude change of RIS units at time t.

[0048] Furthermore, the optimization process of the current RIS reflection parameters in the deep deterministic policy gradient algorithm is: first optimize the amplitude gain parameters based on the gradient descent method, and then implement the amplitude-phase joint optimization under the power constraint. Furthermore, the amplitude-phase joint optimization satisfies the following constraints: N is the total number of reflection units, β n ∈[1,β max ] is the amplitude gain coefficient of the nth reflection unit, is the incident power of the first n reflection units, P c is the circuit power consumption, is the maximum available power of RIS. Further, the amplitude-phase joint optimization under power constraint is implemented, including: using the improved golden section method to take the current optimal phase A dynamic search window is established as the center; the window width Δθ is adjusted according to the change direction of the signal to interference noise ratio of adjacent subframes to obtain the interval When the window width is less than π / 12, a local fine search is triggered. Its advantages lie in the design of a parameter update strategy based on the channel change rate, the use of an improved golden section method for fast phase search, and the optimization of amplitude parameters combined with gradient descent.

[0049] Furthermore, when the adjustment period is reached, the reflection coefficient matrix obtained by S4 is dynamically updated according to the channel change rate, including: pre-correction of the phase deviation caused by Doppler frequency shift; dynamic adjustment of the gain coefficient in the reflection coefficient matrix according to the received signal strength; and active detection of channel state information by the transmitting beam for non-cooperative interference.

[0050] The following describes a process of collaborative scheduling of heterogeneous resources based on fog node collaboration using the communication method provided in this embodiment:

[0051] First, consider a RIS-assisted multiple-input single-output (MISO) communication anti-interference scenario. In this scenario, a transmitter equipped with M antennas performs downlink communication with a single-antenna receiver, while a malicious interference source equipped with L antennas conducts full-band interference on the receiver. The RIS consists of N passive reflector units, controlled by the transmitter via a RIS controller. This paper comprehensively considers a fully connected link propagation scenario in which both direct and reflected links coexist. This scenario involves both legitimate and interfering signals reaching the receiver via both the direct link and the RIS-assisted reflected link, respectively.

[0052] Since path loss has a significant impact on signal strength, it is assumed that the signal power reaching the RIS after multiple reflections can be ignored. In order to improve the performance gain of communication anti-interference, it is assumed that the channel state information (CSI) at each communication node can be accurately detected and obtained. For the CSI of the legitimate signal propagation channel, it is obtained through the conventional method of pilot estimation; and for the CSI of the non-cooperative interference signal propagation channel, it is obtained by deploying a small number of active detection sensors at the RIS and the receiver. The effective baseband channel models from the sender to the RIS, from the interference source to the RIS, from the sender to the receiver, from the interference source to the receiver, and from the RIS to the receiver are respectively expressed as: G l =g L {h nm} n∈[1,N],m∈[1,M] ∈C N×M , G j =g J {h nl} n∈[1,N],l∈[1,L] ∈C L×N ,

[0053] Without loss of generality, assume that all channels H obey the Rician channel model, that is, k is the Rice factor (K-factor), which represents the power ratio of the LoS path to the scattering path. LOS is the deterministic line-of-sight path component; H NLOS is the scattered component of Rayleigh distribution, which obeys complex Gaussian distribution. Excluding RIS units used for other functions, assuming that all RIS reflection units are used to counter interference, the reflection coefficient matrix of RIS can be expressed as Φ l =(Γ l,1 ,Γ l,2 ,……,Γ l,N )∈C N×N , Φ j =(Γ j,1 ,Γ j,2 ,……,Γ j,N )∈C N×N ,in and They represent the reflection coefficients of the nth reflection unit to the legitimate signal and the interference signal, and the amplitude changes are α l,n and α j,n , the phase shifts are and The transmit beamforming vectors of the transmitter and the interference source are respectively. By adjusting the antenna phase and amplitude by weight, the beamforming vectors of the transmitter and the interference source must meet the power constraint w l <P l and w j <P l , through the power P l and P j To achieve normalization,

[0054] Combined with the actual transmission process of the signal, the received signal at the receiver can be expressed as:

[0055]

[0056] Indicates the receiver's legitimate signal, represents the interference signal at the receiver, and n is the variance of the Gaussian white noise at the receiver. Therefore, the SINR of the legitimate signal at the receiver can be expressed as:

[0057]

[0058] In order to improve the anti-interference capability of the communication system under consideration, under multiple constraints such as transmit power and RIS reflection phase shift, an optimization problem of maximizing the legal signal SINR for the transmit beamforming vector and RIS reflection coefficient matrix is ​​constructed as follows:

[0059]

[0060] stC1:||w l || 2 ≤P l,max

[0061]

[0062] C3:S U ≥S min

[0063] Among them, constraint C1 is the transmit power constraint on the sender's beamforming vector. Specifically, the sender's transmit power must not exceed P l,max C2 is the phase shift constraint on the RIS reflection coefficient matrix. Assuming that the signal amplitude remains unchanged after reflection, the continuous phase shift is adjusted to be between (0, 2π]. C3 is the receiver sensitivity requirement, which ensures that the signal quality of the receiver meets the relevant Quality of Service (QoS) requirements.

[0064] Since the interference source emits interference in the entire frequency band, while the legitimate signal is only in a certain frequency band, it can be considered that the interference and the legitimate signal are always in the same frequency band, and the reflection matrix of RIS for both is the same. Therefore, assuming Φ l =Φ j Obviously, from the optimization problem, it can be seen that the optimization variable w l and Φ l The dimension is high and the two are tightly coupled, while the constraint on the RIS phase shift is a non-convex rank-1 constraint, and the objective function is a non-convex fractional-order function.

[0065] The second step is to solve the aforementioned optimization problem to obtain the optimal signal-to-interference-and-noise ratio. This problem is a fractional nonlinear problem and is typically NP-hard. The highly nonlinear coupling between the beamforming vector and the RIS reflection matrix can make the problem inherently nonconvex. Channel parameters exhibit slow temporal variations (at a rate of α), necessitating a real-time adaptive strategy. However, each channel variation requires solving the optimization problem, resulting in computational delays that violate real-time operational constraints. Given the non-convexity of the problem and the dynamic nature of the channel, traditional convex optimization methods exhibit significant limitations in solving optimization problems. DDPG, with its ability to handle continuous actions, dynamic adaptability, and non-convex optimization, significantly outperforms convex optimization in active RIS interference mitigation systems. In particular, DDPG can achieve optimization with lower latency and higher interference mitigation gain in high-dimensional RIS configurations, rapidly varying channels, and hardware constraints.

[0066] The MDP framework for the anti-interference problem consists of four tuples: state space, action space, state transition probability, and reward function:

[0067] 1) State space s: The state space needs to comprehensively represent dynamic parameters, including: CSI; RIS configuration status; jammer characteristics and location; system resources, including system power supply, RIS energy efficiency, etc.

[0068] 2) Action space a: The action needs to implement RIS reflection parameter adjustment Φ l and BS beamforming vector w l Joint optimization.

[0069] 3) State transition probability p: State transition requires modeling the randomness of the communication environment and RIS actions. Based on the Rice fading model, the time-varying characteristics of the channel gain are taken into account.

[0070] 4) Reward function r: The reward function needs to balance communication performance and resource consumption to maximize anti-interference performance. There is a dynamic penalty mechanism for applying large negative rewards for exceeding the constraints of the optimization problem.

[0071] like Figure 3As shown, DDPG combines deep neural networks with deterministic policy gradients and is specifically designed for continuous action space optimization, making it suitable for the dynamic adjustment of high-dimensional parameters in interference-resistant communications. In the proposed system, the action space involves continuous parameters such as phase / amplitude adjustment of RIS reflection units, optimization of beamforming vectors, and power control. Traditional discretization methods can lead to the curse of dimensionality, while DDPG directly outputs deterministic actions through the participant network. In addition, DDPG combines a replay buffer and a target network to effectively address non-stationarity such as time-varying channel characteristics in dynamic interference environments and improves sample utilization through offline policy learning.

[0072] The present invention is described in detail below with reference to simulation: The simulation parameters are set as follows: The channel large-scale fading model is composed of Given, γ0-30dB represents the path loss factor at the reference distance d0=1m, d represents the distance, and α represents the path loss exponent. Without loss of generality, we assume that α BR =2.2,α RU =αRE, k = 2.5, and α BU =α BE =3.6. G l and G j Obey Rayleigh fading. Unless otherwise specified, we assume M = 8, N = 100, L = 8, P l =50dBm, P j =50dBm, P Rm =40dBm, ρ=0.1.

[0073] like Figure 4 As shown in Figure 1, the learning rate plays a key role in DDPG optimization of active RIS systems. A learning rate of 0.001 achieves a near-optimal balance between convergence speed, stability, and anti-interference performance. However, a high learning rate may increase the gradient norm of the critic network's loss function, leading to oscillation.

[0074] Figure 5 The nonlinear relationship between the channel variation rate (ρ) and the signal-to-interference-plus-noise ratio (SINR) is described, revealing a key system design trade-off: RIS significantly improves SINR in quasi-static channels (ρ < 0.1), but its effectiveness diminishes in rapidly changing scenarios. This highlights the need for adaptive algorithms and hardware acceleration to maintain optimization agility in real-world, time-sensitive application deployments. This nonlinearity stems from the competitive relationship between channel coherence time and the latency optimized by RIS.

[0075] Figure 6 、 Figure 7 、 Figure 8The results show that the proposed DDPG optimization scheme approaches the genetic algorithm in terms of anti-interference capability and outperforms other optimization schemes. Active RIS enhances the auxiliary link's support capabilities compared to passive RIS. Furthermore, since the RIS is located closer to the interference source, it plays a major role in suppressing interference. Therefore, minimizing interference is a more effective solution than maximizing legitimate signals. Figure 6 The results show that as the number of RIS units N increases, the system's anti-interference capability gradually improves, thanks to its better utilization of multipath. The growth of active RIS is limited due to its limited power supply. Passive RIS has theoretically great growth potential, but is actually limited by factors such as hardware costs. Figure 7 The results show that the anti-interference capability of the active RIS-assisted system first increases and then saturates as the number of transmit antennas M increases. Its ultimate diversity gain limit is determined by channel independence, hardware defects, and interference dynamics. When the number of transmit antennas exceeds the channel degrees of freedom or the RIS control dimension, the spatial correlation of multipath propagation leads to a decrease in the marginal benefit of diversity gain. Figure 8 Display SINR with legal transmit power P l Increasing legitimate power directly increases the transmission energy of the desired signal, and RIS more effectively converts this energy into signal gain at the receiver through beamforming and interference suppression. From a practical perspective, the sustainability of this linear gain is limited by the accuracy of RIS hardware reciprocity calibration and the saturation threshold of nonlinear components.

[0076] Based on these experimental results, DDPG demonstrates unique advantages in optimizing active RIS-assisted anti-interference systems, particularly under dynamically changing channels. DDPG avoids the "curse of dimensionality" caused by combinatorial explosion in high-dimensional action spaces, thereby reducing algorithmic complexity. Through its deterministic policy network, it directly outputs continuous RIS parameters, enabling gradient-based optimization of non-convex SINR surfaces.

[0077] Example 2

[0078] This embodiment provides a communication device based on a reconfigurable smart surface, including:

[0079] An initial setting module is used to set the initial reflection coefficient matrix of the reconfigurable smart surface RIS and the initial beamforming vector of the transmitter;

[0080] The first control module is configured to control the transmitter to transmit a first communication signal via a first direct link and a first RIS auxiliary link corresponding to an initial reflection coefficient matrix using an initial beamforming vector; the interference source to transmit a first interference signal via a second direct link and a second RIS auxiliary link corresponding to the initial reflection coefficient matrix; and to use the receiver to measure in real time the signal-to-interference-and-noise ratio of a received signal corresponding to the first communication signal and the first interference signal;

[0081] A recurrent optimization module is used to jointly optimize the sender's current beamforming vector and current RIS reflection parameters based on real-time channel estimation results using a deep deterministic policy gradient algorithm. The deep deterministic policy gradient algorithm includes: a reward function centered on maximizing the signal-to-interference-and-noise ratio of the received signal; a Markov decision process state space containing channel state information, power constraints, and historical performance indicators; and a continuous action space containing beamforming vectors, RIS phase offsets, and amplitude gains.

[0082] The second control module is used to control the sender to send the second communication signal via the first direct link and the third RIS auxiliary link corresponding to the current reflection coefficient matrix using the initial beamforming vector; the interference source sends the second interference signal via the second direct link and the fourth RIS auxiliary link corresponding to the current reflection coefficient matrix; and use the receiver to measure the signal-to-interference-and-noise ratio of the received signal corresponding to the second communication signal and the second interference signal in real time;

[0083] The iterative judgment module is used to directly enter the loop optimization module when the adjustment period has not been reached; when the adjustment period is reached, the reflection coefficient matrix output by the second control module is dynamically updated according to the channel change rate, and then enter the loop optimization module.

[0084] Example 3

[0085] This embodiment provides a communication system based on a reconfigurable smart surface, including: a transmitter, a reconfigurable smart surface (RIS), an interference source, a receiver, a memory, and a processor. The transmitter sends a communication signal to the receiver via its corresponding RIS auxiliary link and direct link, and the interference source simultaneously sends an interference signal to the receiver via its corresponding RIS auxiliary link and direct link. The receiver measures the signal-to-interference-and-noise ratio of the communication signal and the received signal corresponding to the interference signal in real time. The memory stores a computer program, and when the processor executes the computer program, the steps of the communication method based on the reconfigurable smart surface are implemented.

[0086] Example 4

[0087] This embodiment provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the communication method based on the reconfigurable smart surface are implemented.

[0088] It will be easily understood by those skilled in the art that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A communication method based on a reconfigurable smart surface, characterized in that: include: S1: Set the initial reflection coefficient matrix of the reconfigurable smart surface RIS and the initial beamforming vector of the transmitter; S2: Control the sender to send a first communication signal via a first direct link and a first RIS auxiliary link corresponding to the initial reflection coefficient matrix using the initial beamforming vector; an interference source to send a first interference signal via a second direct link and a second RIS auxiliary link corresponding to the initial reflection coefficient matrix; and use a receiver to measure, in real time, a signal-to-interference-and-noise ratio of a received signal corresponding to the first communication signal and the first interference signal; S3: Jointly optimizing the sender's current beamforming vector and current RIS reflection parameters based on the real-time channel estimation results using a deep deterministic policy gradient algorithm; the deep deterministic policy gradient algorithm includes: a reward function centered on maximizing the signal-to-interference-and-noise ratio of the received signal; a Markov decision process state space containing channel state information, power constraints, and historical performance indicators; and a continuous action space containing beamforming vectors, RIS phase offsets, and amplitude gains; S4: Control the sender to send a second communication signal via the first direct link and a third RIS auxiliary link corresponding to the current reflection coefficient matrix using the initial beamforming vector; the interference source sends a second interference signal via the second direct link and a fourth RIS auxiliary link corresponding to the current reflection coefficient matrix; and use the receiver to measure the signal to interference and noise ratio of the received signal corresponding to the second communication signal and the second interference signal in real time; S5: When the adjustment period has not been reached, directly proceed to S3; when the adjustment period has been reached, dynamically update the reflection coefficient matrix obtained in S4 according to the channel change rate, and then proceed to S3.

2. The communication method based on the reconfigurable smart surface according to claim 1, characterized in that: The reward function r t Designed for: t =10log 10 (γ t )-λ1ΔP t -λ2D t ; Among them, γ t is the signal-to-interference-noise ratio of the received signal at time t, ΔP t is the power constraint violation at time t, D t is the service quality deviation at time t, λ1 and λ2 are two penalty coefficients.

3. The communication method based on the reconfigurable smart surface according to claim 2, characterized in that: The state space s t Means: s t =[vec(G l ),h j ,h,P t ,P j ,γ t-1 ]; The action space a t Indicates: a t =[w t ,θ1,…,θ N ,β1,…,β N ]; Among them, vec(G l ) is the vector corresponding to the channel matrix of the sender-RIS, h j is the channel matrix from the interference source to the RIS, h is the channel matrix from the RIS to the receiver, P t is the transmit power of the sender, P j is the transmission power of the interference source, γ t-1 is the signal-to-interference-noise ratio of the received signal at time t-1, w t is the square beamforming vector sent at time t, θ1,…,θ N is the phase shift of each unit of RIS at time t, β1,…,β N is the amplitude change of RIS units at time t.

4. The communication method based on the reconfigurable smart surface according to claim 1, characterized in that: The optimization process of the current RIS reflection parameters in the deep deterministic policy gradient algorithm is as follows: first, the amplitude gain parameters are optimized based on the gradient descent method, and then the amplitude-phase joint optimization is implemented under the power constraint condition.

5. The communication method based on the reconfigurable smart surface according to claim 4, characterized in that: The amplitude-phase joint optimization satisfies the following constraints: Where N is the total number of reflection units, β n ∈[1,β max ] is the amplitude gain coefficient of the nth reflection unit, is the incident power of the nth reflector unit, P c is the circuit power consumption, is the maximum available power of RIS.

6. The communication method based on the reconfigurable smart surface according to claim 4, characterized in that: The re-implementation of the amplitude-phase joint optimization under the power constraint condition includes: With the current optimal phase A dynamic search window is established as the center; the window width Δθ is adjusted according to the change direction of the signal to interference noise ratio of adjacent subframes to obtain the interval When the window width is less than π / 12, a local fine search is triggered.

7. The communication method based on a reconfigurable smart surface according to any one of claims 1 to 6, characterized in that: The step of dynamically updating the reflection coefficient matrix obtained in S4 according to the channel change rate when the adjustment period is reached includes: Pre-correction is performed on the phase deviation caused by Doppler frequency shift; the gain coefficient in the reflection coefficient matrix is ​​dynamically adjusted according to the received signal strength; and for non-cooperative interference, the transmitting beam actively detects channel state information.

8. A communication device based on a reconfigurable smart surface, characterized in that: include: An initial setting module is used to set the initial reflection coefficient matrix of the reconfigurable smart surface RIS and the initial beamforming vector of the transmitter; A first control module is configured to control the sender to send a first communication signal via a first direct link and a first RIS auxiliary link corresponding to the initial reflection coefficient matrix using the initial beamforming vector; and an interference source to send a first interference signal via a second direct link and a second RIS auxiliary link corresponding to the initial reflection coefficient matrix; Using a receiving party to measure in real time the signal-to-interference-and-noise ratio of a received signal corresponding to the first communication signal and the first interference signal; A loop optimization module, configured to jointly optimize the current beamforming vector and the current RIS reflection parameter of the transmitter according to the channel estimation result obtained in real time by using a deep deterministic policy gradient algorithm; The deep deterministic policy gradient algorithm includes: a reward function centered on maximizing the signal-to-interference-and-noise ratio of the received signal; a Markov decision process state space containing channel state information, power constraints, and historical performance indicators; and a continuous action space containing beamforming vectors, RIS phase offsets, and amplitude gains. A second control module is configured to control the sender to send a second communication signal via the first direct link and a third RIS auxiliary link corresponding to the current reflection coefficient matrix using the initial beamforming vector; the interference source to send a second interference signal via the second direct link and a fourth RIS auxiliary link corresponding to the current reflection coefficient matrix; and to use the receiver to measure in real time the signal-to-interference-and-noise ratio of the received signal corresponding to the second communication signal and the second interference signal; An iterative judgment module is used to directly enter the loop optimization module when the adjustment period has not been reached; when the adjustment period is reached, the reflection coefficient matrix output by the second control module is dynamically updated according to the channel change rate, and then enter the loop optimization module.

9. A communication system based on a reconfigurable smart surface, characterized in that: include: A sender, a RIS, an interference source, a receiver, a memory, and a processor, wherein the sender sends a communication signal to the receiver via its corresponding RIS auxiliary link and direct link, and the interference source sends an interference signal to the receiver via its corresponding RIS auxiliary link and direct link; the receiver measures the signal-to-interference-and-noise ratio of the communication signal and the received signal corresponding to the interference signal in real time; the memory stores a computer program, wherein the processor implements the steps of the method described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images