Hybrid vehicle system ISAC optimization method based on unmanned aerial vehicle intelligent reflecting surface

Abstract

The invention relates to the field of wireless communication between unmanned aerial vehicles and the Internet of Vehicles, and discloses a hybrid vehicle system ISAC optimization method based on an intelligent reflecting surface of an unmanned aerial vehicle. According to the method, firstly, vehicles in a road are clustered, the CAV serves as a cluster head vehicle, and the HDV serves as a cluster member vehicle; the cluster head vehicle can send a sensing signal to sense the state of the cluster member while communicating with the cluster member; meanwhile, the CAV can also communicate with other CAVs through a virtual channel constructed by the intelligent reflecting surface on the unmanned aerial vehicle; in order to improve the performance of the ISAC system and solve the problem of variable coupling, a joint optimization method of the CAV and the intelligent reflecting surface of the unmanned aerial vehicle is designed; according to the method, intra-cluster and inter-cluster communication rates and sensing beam gains are maximized by jointly optimizing a precoding matrix, the position of an unmanned aerial vehicle and a reflecting surface phase; according to the invention, the ISAC performance of the unmanned aerial vehicle intelligent reflector hybrid vehicle system can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of wireless communications, and in particular to an ISAC optimization method based on a hybrid intelligent vehicle system assisted by an intelligent reflective surface of an unmanned aerial vehicle. Background Art

[0002] For a considerable period of time in the future, there will be two types of vehicles on the road: intelligent, communicating, and autonomous vehicles (CAVs), and human-driven vehicles (HDVs). The CAVs, equipped with more advanced technology, are far more intelligent than the HDVs, capable of handling more communication and computing tasks and leveraging their advanced hardware to perceive the road environment. However, the HDVs, as a significant component of the transportation system, also require road information. However, base stations lack coverage of many road details, so obtaining this information from CAVs is a promising solution. To ensure that HDVs receive timely information from CAVs, it is possible to cluster them, with CAVs serving as cluster heads and HDVs as cluster members. CAVs need to send a communication-associated sensing signal (ISAC) to each cluster member, allowing them to communicate with and perceive their status.

[0003] However, relying solely on a single cluster head for road perception is insufficient, so multiple cluster head vehicles are required to collaborate on the road. However, because the LOS channels between cluster head vehicles are blocked, the channel capacity and reliability of the NLOS channels cannot meet the requirements. This makes collaboration between cluster head vehicles difficult.

[0004] Thanks to the development of smart reflective surfaces, users can communicate with each other through the virtual LOS generated by these surfaces, significantly improving information transmission speed and stability. However, fixed reflective surface locations often fail to adapt to the highly dynamic IoV environment. Drones, however, are highly flexible and maneuverable, allowing them to operate over most land areas. Therefore, deploying smart reflective surfaces on drones is a promising solution. CAVs can communicate with other CAVs using the virtual LOS generated by these surfaces. Therefore, CAVs on the road need to generate intra-cluster ISAC beams and inter-cluster communication beams. Furthermore, the drones need to adjust their positions and the phase of the reflective surfaces according to the changing road conditions.

[0005] However, there is currently no solution proposed for optimizing ISAC performance of combined CAV and UAV smart reflective surfaces in hybrid vehicle systems. Summary of the Invention

[0006] In order to overcome the shortcomings of the existing technology, the present invention provides an effective method for optimizing the performance of CAV and UAV intelligent reflective surface combined with ISAC in a hybrid vehicle system.

[0007] The technical solution adopted by the present invention to solve the technical problem is: This invention is based on a hybrid intelligent vehicle system assisted by a UAV and intelligent reflective surfaces. The system consists of a UAV equipped with an intelligent reflective surface (RIS). On a road served by the UAV, multiple CAVs and HDVs are present. The CAV uses its own sensing signals to detect nearby HDVs, grouping them into its own cluster and then notifying the roadside unit (RSU) of the clustering results. Furthermore, the first CAV on the road communicates unknown information ahead to other CAVs via a wireless channel.

[0008] The ISAC optimization method for the UAV intelligent reflector-assisted hybrid intelligent vehicle system includes the following steps: 1) When CAVs and HDVs enter a road, each CAV groups nearby HDVs into its own cluster and notifies the RSU of the clustering results. The drone acquires the positions of all CAVs and calculates the CAV-to-drone communication channel, while maintaining continuous motion within a specified period to facilitate inter-CAV communication.

[0009] 2) The first CAV on the road will serve as the lead vehicle, transmitting road information to the remaining CAVs. Each time the first CAV leaves the road, a new lead vehicle will be selected from the remaining CAVs. Simultaneously, the drone will calculate the communication channel with each CAV based on their position and adjust its own position in real time, thereby assisting the lead vehicle in communicating with the other CAVs.

[0010] 3) The first CAV on the road generates an ISAC beam within the cluster and an inter-cluster communication beam, while the remaining CAVs generate ISAC beams within their own cluster. In addition, different frequency bands are used for intra-cluster and inter-cluster communication. In addition, to measure the ISAC performance, the intra-cluster communication rate and , inter-cluster communication rate and , intra-cluster sensing beam gain and As a criterion for judgment.

[0011] 4) Optimize the precoding matrix, UAV position, and smart reflector phase based on the intra-cluster communication rate, inter-cluster communication rate, and intra-cluster beam pattern gain.

[0012] Furthermore, in each time slot, in step 4), the intra-cluster and inter-cluster beamforming of the lead vehicle, as well as the position and reflector phase of the UAV, are first optimized. Then, the inter-cluster beamforming of the remaining CAV clusters is optimized.

[0013] Furthermore, the optimization problem of the head vehicle cluster and the UAV is modeled as follows: ; Similarly, the optimization problem for the remaining CAV clusters is modeled as follows: ; In each cluster, HDVs, and there are a total of CAV. They represent the intra-cluster communication rate, inter-cluster communication rate and sensing beam gain respectively. The weights representing the intra-cluster and inter-cluster communication rates and the perceived beam gain. They represent the maximum transmission power of the vehicle, the precoding matrix of the intra-cluster communication signal, the precoding matrix of the inter-cluster communication signal, and the precoding matrix of the sensing signal. represents the phase shift unit of RIS, correspondingly, represents the RIS phase shift matrix. Indicates the current position of the drone. Indicates the position of the drone at the last moment.

[0014] In summary, Equations (2) and (6) represent the vehicle's transmission power limits. Equation (3) represents the RIS's phase shift range limits. Equation (4) represents the drone's speed limits.

[0015] Taking the optimization problem of the head vehicle cluster as an example, the original optimization problem (1) is split into two sub-problems: First, optimize the precoding matrix Then optimize the drone position according to the precoding matrix optimization results And smart reflector phase shift matrix ; For the optimization problem (5) of other clusters, the precoding matrix of the inter-cluster communication signal, the position of the drone, and the RIS phase shift are not considered, so only the precoding matrix needs to be optimized .

[0016] Furthermore, the intra-cluster communication rate and inter-cluster communication rate in the optimization problem (1) are expressed as ; in ; Respectively represent the The signal-to-interference-noise ratio of HDV and the inter-cluster The signal to interference noise ratio at CAV. Indicates CAV to Channel gain for each HDV. and They represent the precoding vectors of the signals sent by the head vehicle to the HDV and other CAVs in its cluster, so the precoding matrix , . represents the covariance matrix of the radar. and Respectively represent HDV and The noise at each CAV. Indicates that the drone has arrived The channel gain of each CAV. represents the smart reflection surface phase shift matrix, Indicates the number of smart reflective surface elements. Indicates the channel gain from the lead vehicle to the UAV.

[0017] In addition, the optimization problem (1) The perceptual beam gain in each HDV direction is expressed as: in represents the azimuth, Indicates the pitch angle.

[0018] Furthermore, the optimization of the lead vehicle and the drone in step 4) specifically includes: 4.1、Derivation about The derivative of .

[0019] 4.2. Construct a Deep Unfolding model based on Projected Gradient Ascent (PGA). This model first unfolds each iteration of PGA into a fully connected layer in the neural network and sets the iteration step size in PGA as a trainable parameter in the neural network. The training loss function is set to: The Deep Unfolding model after training can be solved within a limited number of PGA iterations. .

[0020] 4.3. The Proximal Policy Gradient (PPO) algorithm is used to solve the UAV position deployment and reflector optimization problem. Specifically, the UAV obtains the channels between itself and all CAVs and uses them as input to the PPO. The PPO then uses the Deep Unfolding model to output the UAV's next position deployment, the reflector phase shift, and the precoding matrix for the signal sent by the lead vehicle.

[0021] For the remaining CAV optimization problems, we only need to use the Deep Unfolding model to solve the precoding matrix of the intra-cluster communication signal, without considering the precoding matrix of the inter-cluster communication signal and the phase shift optimization of the drone position and reflection surface.

[0022] The technical concept of this invention is that existing technology still lacks ISAC performance optimization for hybrid vehicle systems that combine CAVs and drones with intelligent reflective surfaces. Therefore, this invention focuses on this area and provides a solution that considers beamforming, drone positioning, and intelligent reflective surface optimization. This method improves ISAC system performance by optimizing intra-cluster and inter-cluster communication rates and intra-cluster sensing beam gain.

[0023] The beneficial effects of the present invention are mainly manifested in: filling the gap in the existing technology in the optimization of ISAC performance of the combined CAV and UAV intelligent reflective surface in the hybrid vehicle system, and improving the communication rate between clusters and the perception beam gain within the cluster by combining beamforming, UAV position and intelligent reflective surface. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 is a schematic diagram of a hybrid intelligent vehicle system assisted by an intelligent reflective surface of a UAV according to the method of the present invention; Figure 2 Schematic diagram of the performance of the method of the present invention at different time slots; Figure 3 Schematic diagram of the performance of the intra-cluster and inter-cluster communication rates of the method of the present invention at different time slots. DETAILED DESCRIPTION

[0025] The present invention will be further described below with reference to the accompanying drawings.

[0026] The present invention first groups vehicles on the road into clusters, with CAVs serving as cluster heads and HDVs serving as cluster members. The cluster head can send inter-cluster signals to communicate with cluster members while sensing their status. CAVs can also communicate with other CAVs via virtual channels created by the intelligent reflective surfaces on drones. To improve the performance of the ISAC system and address variable coupling, a joint optimization method for CAVs and drone intelligent reflective surfaces was designed. This method maximizes intra-cluster and inter-cluster communication rates and sensing beam gain by jointly optimizing the precoding matrix, drone position, and reflective surface phase.

[0027] Reference Figures 1 to 3 ,The ISAC optimization method of the hybrid vehicle system based on the ,UAV intelligent reflective surface is implemented based on the existing wireless ,information transmission system, that is, the UAV intelligent reflective surface assisted hybrid ,intelligent vehicle system is composed of a UAV equipped with an intelligent ,reflective surface, multiple HDVs and multiple CAVs.

[0028] In this embodiment, the method provides services to CAVs on the road within a specified period. The drone uses its onboard intelligent reflective surface to create a virtual LOS, providing a communication channel between the lead vehicle and other CAVs. The intelligent reflective surface also reconfigures the channel between CAVs by changing the phase of its own reflected signal. Simultaneously, CAVs on the road also transmit inter-cluster signals to HDVs within their cluster, allowing them to simultaneously communicate with the HDVs and sense their status. Furthermore, CAVs within a cluster and HDVs communicating with inter-cluster CAVs operate on different frequency bands.

[0029] In this embodiment, the intra-cluster communication rate and the inter-cluster communication rate are expressed as ; in ; Respectively represent the The signal-to-interference-noise ratio of HDV and the inter-cluster The signal to interference noise ratio at CAV. Indicates CAV to Channel gain for each HDV. and They represent the precoding vectors of the signals sent by the head vehicle to the HDV and other CAVs in its cluster, so the precoding matrix , . represents the covariance matrix of the radar. and Respectively represent HDV and The noise at each CAV. Indicates that the drone has arrived The channel gain of each CAV. represents the smart reflection surface phase shift matrix, Indicates the number of smart reflective surface elements. Indicates the channel gain from the lead vehicle to the UAV.

[0030] In addition, the cluster The perceptual beam gain in each HDV direction is expressed as ; in represents the azimuth, Indicates the pitch angle.

[0031] The precoding matrix, UAV position, and smart reflector phase optimization of the lead vehicle cluster in this embodiment can be expressed as: ; The precoding matrix optimization for other clusters can be expressed as: ; The optimization problems of other clusters can be solved directly, but the optimization problem of the head cluster needs to be decomposed into two sub-problems. First, optimize the precoding matrix Then optimize the drone position according to the precoding matrix optimization results And smart reflector phase shift matrix .

[0032] The ISAC performance optimization method for a hybrid intelligent vehicle system assisted by a drone's intelligent reflective surface in this embodiment addresses existing ISAC performance optimization methods that fail to consider combining CAV and drone intelligent reflective surfaces. This method improves the ISAC performance of the hybrid intelligent vehicle system by jointly optimizing the precoding matrix, drone position, and intelligent reflective surface phase.

[0033] In this embodiment, CAVs and HDVs are randomly distributed on the road and continuously moving, forming multiple clusters. A drone flexibly moves 50 meters above the road to facilitate communication between CAVs. Each CAV also adjusts its precoding matrix settings in real time based on the HDV's position.

[0034] The method of the present invention uses a 20-layer Deep Unfolding algorithm to solve the precoding matrix and utilizes PPO to optimize the reflector phase shift and drone position. Comparative Method 1 uses a 20-layer Deep Unfolding algorithm to solve the precoding matrix and utilizes PPO to optimize the drone position, while the reflector phase shift units are randomly set. Comparative Method 2 uses a zero-forcing (ZF) algorithm to solve the precoding matrix and utilizes PPO to optimize the reflector phase shift and drone position. Comparative Method 3 uses a 10-layer Deep Unfolding algorithm to solve the precoding matrix and utilizes PPO to optimize the reflector phase shift and drone position.

[0035] Figure 2 The performance of the ISAC system in each time slot using the proposed method is demonstrated. This performance is evaluated using the weighted sum of intra-cluster and inter-cluster communication rates and intra-cluster perceptual beam gain. As can be seen, the proposed method maintains superior performance compared to other methods for 30 consecutive time slots, with no significant fluctuations.

[0036] Figure 3 The sum of the intra-cluster and inter-cluster communication rates for each time slot in the method of the present invention is shown. This demonstrates that the method of the present invention can effectively improve intra-cluster and inter-cluster communication rates, achieving superior performance compared to other comparative methods. Therefore, the present invention possesses certain advantages.

[0037] The contents described in the embodiments of this specification are merely an enumeration of the implementation forms of the inventive concept and are for illustrative purposes only; the scope of protection of the present invention should not be regarded as limited to the specific forms described in this embodiment, and the scope of protection of the present invention is also based on the equivalent technical means that ordinary technicians in this field can think of based on the inventive concept.

Claims

1. The ISAC optimization method for hybrid vehicle systems based on intelligent reflective surfaces of UAVs is characterized by: The system includes an unmanned aerial vehicle (UAV) equipped with a smart reflective surface (RIS). Multiple intelligent communicating autonomous vehicles (CAVs) and ordinary human-driven vehicles (HDVs) are present on the road served by the UAV. The CAV uses its own sensing signals to detect nearby HDVs and group them into its own cluster. It then notifies the roadside unit (RSU) of the clustering results. In addition, the first CAV on the road notifies other CAVs of unknown information ahead via a wireless channel. The method includes the following steps: 1) When CAVs and HDVs enter the road, each CAV divides nearby HDVs into its own cluster and notifies the RSU of the clustering result. The drone obtains the positions of all CAVs and calculates the channel between the CAV and the drone, while continuously moving within a set period to assist in communication between CAVs. 2) The first CAV on the road will serve as the lead vehicle, transmitting road information to other CAVs. Whenever the first CAV leaves the road, a new lead vehicle will be selected from the remaining CAVs. Simultaneously, the drone will calculate the communication channel with each CAV based on its position and adjust its own position in real time, thereby assisting the lead vehicle in communicating with other CAVs. 3) The first CAV on the road generates ISAC beams for intra-cluster communication and perception and inter-cluster communication. The remaining CAVs generate ISAC beams for their own clusters. In addition, different frequency bands are used for intra-cluster and inter-cluster communication. In order to measure the performance of ISAC, the intra-cluster communication rate and , inter-cluster communication rate and , intra-cluster sensing beam gain and As a criterion for judging; 4) Optimize the precoding matrix, UAV position, and smart reflector phase based on the intra-cluster communication rate, inter-cluster communication rate, and intra-cluster beam pattern gain.

2. The ISAC optimization method for a hybrid vehicle system based on an intelligent reflective surface of a UAV as claimed in claim 1, characterized in that: In each time slot, in step 4), the intra-cluster and inter-cluster beamforming of the lead vehicle and the position and reflector phase of the UAV are optimized first, and then the synaesthesia beamforming of the remaining CAVs is optimized; The modeling of the two optimization problems is as follows: The optimization problem of the lead vehicle and the UAV is modeled as follows: ； The optimization problem for the remaining CAVs is modeled as follows: ； There are K HDVs in each cluster, and there are C CAVs in total on the road excluding the lead vehicle. They represent intra-cluster communication rate, inter-cluster communication rate and sensing beam gain respectively; represents the azimuth, Indicates the pitch angle; The weights representing the intra-cluster and inter-cluster communication rates and the perceived beam gain; They represent the maximum transmission power of the vehicle, the precoding matrix of the intra-cluster communication signal, the precoding matrix of the inter-cluster communication signal, and the precoding matrix of the sensing signal respectively; represents the phase shift unit of RIS, correspondingly, represents the RIS phase shift matrix; Indicates the position of the drone; Indicates the position of the drone at the last moment; When the optimization problem of the head vehicle cluster is considered, the original optimization problem (1) is split into two sub-problems: First, optimize the precoding matrix Then optimize the drone position according to the precoding matrix optimization results And smart reflector phase shift matrix ; For the optimization problem (5) of other clusters, the precoding matrix of the inter-cluster communication signal, the position of the drone, and the RIS phase shift are not considered. Therefore, only the precoding matrix needs to be optimized. .

3. The ISAC optimization method for a hybrid vehicle system based on an intelligent reflective surface of a UAV as claimed in claim 1, characterized in that: In step 3), in each time slot, the intra-cluster communication rate and inter-cluster communication rate are expressed as: ； in ； Respectively represent the The signal-to-interference-noise ratio of HDV and the inter-cluster Signal to Interference and Noise Ratio at each CAV; in, Indicates CAV to Channel gain of each HDV; and They represent the precoding vectors of the signals sent by the head vehicle to the HDV and other CAVs in its cluster, so the precoding matrix , ; represents the covariance matrix of the radar; and Respectively represent HDV and Noise at each CAV; Indicates that the drone has arrived The channel gain of each CAV; represents the smart reflection surface phase shift matrix, Indicates the number of smart reflective surface elements; Indicates the channel gain from the lead vehicle to the UAV.

4. The ISAC optimization method for a hybrid vehicle system based on an intelligent reflective surface of a UAV as claimed in claim 1, characterized in that: The optimization of the lead vehicle and the drone in step 4) specifically includes: Step 4.1: Derivation about The derivative of Step 4.2: Build a Deep Unfolding model based on Projected Gradient Ascent (PGA). First, expand each iteration of PGA into a fully connected layer in the neural network, and set the iteration step size in PGA as a trainable parameter in the neural network. Set the training loss function to: ; The Deep Unfolding model after training is solved within a limited number of PGA iterations ; Step 4.3: Use the proximal policy gradient algorithm (PPO) to solve the UAV position deployment and reflection surface optimization problem. Specifically, the UAV obtains the channels between itself and all CAVs and uses them as the input of the PPO. The PPO then combines the Deep Unfolding model to output the UAV's next position deployment, the reflection surface phase shift, and the precoding matrix of the signal sent by the lead vehicle. Step 4.4: For the remaining CAV optimization problems, the Deep Unfolding model is used to solve the precoding matrix without considering the optimization of the UAV position and reflective surface.

Images