Unmanned aerial vehicle physical layer transmission method for overall security
By introducing interfering drones and improved deep reinforcement learning algorithms in the physical layer transmission of the UAV, the security performance optimization problem caused by the unknown location of potential eavesdroppers in UAV communication is solved, and continuous security guarantee and efficient learning of the UAV communication process are achieved.
Patent Information
- Application Number
- CN202510246622.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-04
- Publication Date
- 2025-04-01
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing UAV physical layer security transmission method is difficult to optimize the cumulative security performance of the entire transmission process when the Eves location of the potential eavesdropper is unknown. When the deep reinforcement learning method deals with continuous decision-making problems in the UAV communication environment, it has problems such as slow convergence speed and low learning efficiency.
A drone physical layer transmission method for overall security is adopted to obtain the location information of communication drones and users, calculate the threat area of potential eavesdropper Eves and discrete it to construct optimization problems. Using the long-term and short-term memory network LSTM and priority experience playback PER mechanism, combined with the deep reinforcement learning algorithm DDQN, the interference strategy of interfering with the drone is optimized to achieve continuous security guarantee for the communication process of the drone.
By introducing interference drones and improved deep reinforcement learning algorithms, continuous security guarantees for the drone communication process are achieved, and the learning efficiency and convergence performance of the algorithm are improved, so that it can better adapt to the dynamically changing drone communication environment.
Smart Images
Figure CN119743230A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of unmanned aerial vehicle communication, and in particular relates to a physical layer transmission method of an unmanned aerial vehicle oriented to overall security. Background Art
[0002] The UAV communication system is vulnerable to security threats such as eavesdropping and interference due to its open wireless transmission environment. To address this issue, physical layer security technology has become an important security measure by utilizing the physical characteristics of wireless channels to ensure communication security. Most of the previous research on the physical layer security of UAVs is based on the situation where the position of the potential eavesdropper Eves is known, and focuses on the instantaneous security performance of a single time slot, such as maximizing the instantaneous confidentiality capacity or instantaneous security rate. However, in the actual UAV communication process, the lack of information on the position of the potential eavesdropper Eves makes it impossible to calculate the eavesdropping channel, resulting in a lack of system security measurement methods for this situation. Therefore, it is more practical to optimize the cumulative security performance of the entire transmission process when the position of the potential eavesdropper Eves is unknown.
[0003] In addition, due to the dynamics and complexity of the UAV communication environment, traditional optimization methods are difficult to effectively handle high-dimensional decision spaces. Although deep reinforcement learning methods have shown advantages in solving complex decision problems, existing deep Q-network algorithms have problems such as low learning efficiency and poor convergence performance when processing time-related decision sequences. Therefore, there is an urgent need for a UAV physical layer security transmission method that can optimize overall security performance and has efficient learning capabilities. Summary of the invention
[0004] In view of the above-mentioned deficiencies in the prior art, the present invention provides a drone physical layer transmission method for overall security, which solves the following technical problems: First, the existing drone physical layer security transmission method only focuses on the instantaneous security performance of a single time slot, ignoring the cumulative security performance of the entire transmission process, and it is difficult to ensure the continuous security of the drone communication process; second, the current research focuses on simple and easy-to-handle scenarios, ignoring the consideration of the potential eavesdropper Eves' position and drone movement. Third, the existing deep reinforcement learning method has the problems of slow convergence speed and low learning efficiency when dealing with continuous decision-making problems in drone communication environments.
[0005] In order to achieve the above objectives, the technical solution adopted by the present invention is: a physical layer transmission method of a drone for overall security, comprising the following steps: S1. Obtain the location and power information of the communication drone at each moment, as well as the location of the user at each moment; S2. According to the current position and power information of the communication drone, the Euclidean distance between the communication drone and the ground node is calculated when the instantaneous signal-to-noise ratio of the potential eavesdropper Eves is equal to the threshold of the signal-to-noise ratio of the potential eavesdropper Eves, and the threat area with the communication drone on the ground as the center and R as the radius is obtained, where the Euclidean distance is recorded as R; S3, discretize the threat area in a grid manner and construct a vector of length n, where each element in the vector corresponds to a position, indicating whether the position is a safe area; S4. Establish an optimization problem for the safety of elements and positions in the vector at each moment, and sum the optimization problems of all time periods, and use the sum result as the final optimization problem. The objective function is to maximize the sum of the safe area ratios of all time periods, and the position is the threatened area. The position of the discretized point in represents the position of the potential eavesdropper Eves; S5. According to the final optimization problem, the long short-term memory network LSTM and the priority experience replay PER mechanism are used in combination with the deep reinforcement learning algorithm DDQN to obtain the interference strategy for jamming drones and complete the drone physical layer transmission for overall security.
[0006] The beneficial effects of the present invention are as follows: the present invention uses an additional jamming drone as an auxiliary to interfere with the potential eavesdropper Eves in the environment by sending artificial noise to ensure the safe transmission of information between legitimate drones and users; the security performance of multiple time slots is taken as the overall optimization goal, and combined with the improved dual-depth Q network algorithm, continuous security protection of the drone communication process is achieved. This method improves the learning efficiency and convergence performance of the algorithm by introducing the long short-term memory network LSTM and the priority experience replay PER mechanism, so that it can better adapt to the dynamically changing drone communication environment.
[0007] Furthermore, the expression of the Euclidean distance is as follows: ; ; ;in, Indicates drone k With ground node u The average path loss between represents the line-of-sight transmission probability between the UAV and the ground unit, Indicates drone k With ground node u The Euclidean distance between and Represent the path loss index of LoS and NLoS respectively. NLoS represents non-line-of-sight link, and LoS represents line-of-sight link. and are attenuation coefficients, represents the transmission probability of NLoS, and All represent constants, Indicates communication drone k The vertical distance to the ground, Indicates ground user u With communication drone k The projected distance between them on the ground plane.
[0008] Furthermore, the instantaneous signal-to-noise ratio of the potential eavesdropper Eves is expressed as follows: ;in, represents the instantaneous signal-to-noise ratio of the potential eavesdropper Eves, and They represent the transmission power of the communication UAV and the transmission power of the friendly interference UAV respectively. and They represent the path losses from the communication UAV and the jamming UAV to the potential eavesdropper Eves, Represents the noise power of the eavesdropper Eves.
[0009] Furthermore, the expression of the vector is as follows: ;in, Indicates time t The vector of Indicates the number of i elements, Indicates n elements, T Represents the transpose of a vector.
[0010] The beneficial effect of the above further solution is that the vector Used to indicate threat areas Specifically, the vector Each element of Corresponding to a location in the threat area, >0 means there is a potential eavesdropping threat at this location. <0 means that the location is safe. Through this quantitative representation, the distribution of threat areas can be clearly described, providing clear input for subsequent optimization problems. This not only simplifies the complexity of the problem, but also enables the optimization problem to be efficiently solved through numerical methods (such as convex optimization, reinforcement learning, etc.).
[0011] Furthermore, the final optimization problem is expressed as follows: ; ; ; ; ; ; ; ; ; ; ; ; where a represents the acceleration of the friendly jammer drone, P Indicates the power of friendly interference drones, Indicates time t In the vector i elements, express t The threat zone at all times, T Indicates the total time, Indicates time t The vector of Indicates time t The length of the vector, represents the instantaneous signal-to-noise ratio of the potential eavesdropper Eves, and Respectively represent the friendly interference drone at time t of x Coordinates and y coordinate, and Respectively represent the friendly interference drone at time t The power and height, and Indicates the potential eavesdropper Eves x Coordinates and y coordinate, represents the threshold of Eves signal-to-noise ratio of potential eavesdroppers, represents the user signal-to-noise ratio threshold, represents the user signal-to-noise ratio, Indicates friendly jamming of drones at all times t The horizontal acceleration vector, Indicates friendly jamming of drones at all times t The vertical acceleration vector, and denote the upper limits of acceleration and total energy respectively, Indicates friendly interference drone at all times t The horizontal instantaneous velocity, Indicates friendly interference drone at all times t The vertical instantaneous velocity, Indicates the maximum speed of the friendly jammer drone. Represents the three-dimensional UAV propulsion energy model, , and Respectively indicate time t Communication drone x coordinate,y Coordinates and height, Indicates the minimum distance between communicating UAVs to avoid collision. , , , , and Indicates the upper and lower limits of the flight range of friendly interference drones. Indicates the time slot size, and They represent the profile power and induced power of the blade in the hovering state, Indicates the continuous power during the entire descent and ascent process, Indicates friendly jamming of drones at all times t The average speed, is the tip speed of the rotor blade, is the drag coefficient of the fuselage, represents the air density, Indicates the strength of the rotor. represents the area of the rotor disk, represents the average induced velocity of the rotor in hover, Indicates friendly interference drone at all times t The average vertical velocity.
[0012] The beneficial effect of the above further scheme is that the time slot optimization only focuses on the optimal solution at a single time point, while the overall time optimization takes a global perspective and considers the safety of the entire mission cycle. This global optimization can avoid local optimal solutions and ensure that the drone always adopts the optimal safety strategy during the entire mission, thereby improving the overall safety performance of the system.
[0013] Furthermore, the S5 comprises the following steps: Initialize the parameters of the current long short-term memory network LSTM; Get the initial observation state and buffer; Use the long short-term memory network LSTM to take the observed state as input and solve the Q value of all actions; Select actions using the greedy strategy method e-greedy ; Execution Action , and receive the next state respectively , end identifier and rewards ; Will Store into the buffer according to priority, where Indicates the current action; According to the sample optimization level, unexperienced samples are extracted from the buffer, where the expression of sampling weight is as follows: ; ;in, represents the sampling weight, represents the total number of samples in the experience pool, represents the probability of selecting a sample, represents the hyperparameter used to control the magnitude of the correction, k represents the index of the sample, Indicates k The priority of each sample; According to the final optimization problem, combined with the extracted unexperienced samples, the target value is calculated and the loss function is used Update the long short-term memory network parameters θ; Through continuous iterative optimization, the jamming strategy for jamming drones is obtained, and the drone physical layer transmission for overall security is completed.
[0014] Furthermore, the expression of the priority is as follows: ;in, Indicates priority, Indicates the time difference, Represents a constant used to adjust the importance of the optimization level; Time difference The update expression is as follows: .
[0015] The beneficial effect of the above further scheme is that the Prioritized Experience Replay PER mechanism assigns a priority to each experience sample according to the time difference error (TDerror), so that samples with greater impact on policy optimization are sampled and trained more frequently. This mechanism can accelerate model convergence, avoid inefficient learning caused by uniform sampling, and correct the deviation caused by non-uniform sampling through importance sampling weights to ensure the stability and accuracy of training. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] Figure 1 The figure is a flow chart of the method of the present invention.
[0017] Figure 2 This is a model diagram of the drone communication system. DETAILED DESCRIPTION
[0018] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.
[0019] Example like Figure 1 As shown, the present invention provides a physical layer transmission method for drones oriented to overall security, and the implementation method thereof is as follows: S1. Obtain the position and power information of the communication drone at each moment, as well as the position of the user at each moment, wherein the position of the user at each moment is used to calculate the user signal-to-noise ratio, ensuring that the user signal-to-noise ratio is greater than the user signal-to-noise ratio threshold, which is reflected in the second constraint condition of the optimization problem; S2. According to the current position and power information of the communication drone, the Euclidean distance between the communication drone and the ground node is calculated when the instantaneous signal-to-noise ratio of the potential eavesdropper Eves is equal to the threshold of the signal-to-noise ratio of the potential eavesdropper Eves, and the threat area with the communication drone on the ground as the center and R as the radius is obtained, where the Euclidean distance is recorded as R; S3, discretize the threat area in a grid manner and construct a vector of length n, where each element in the vector corresponds to a position, indicating whether the position is a safe area; S4. Establish an optimization problem for the safety of elements and positions in the vector at each moment, and sum the optimization problems of all time periods, and use the sum result as the final optimization problem. The objective function is to maximize the sum of the safe area ratios of all time periods, and the position is the threatened area. The position of the discretized point in represents the position of the potential eavesdropper Eves; S5. According to the final optimization problem, the long short-term memory network LSTM and the priority experience replay PER mechanism are combined with the deep reinforcement learning algorithm DDQN to obtain the interference strategy for the UAV and complete the UAV physical layer transmission for overall security. The implementation method is as follows: Initialize the parameters of the current long short-term memory network LSTM; Get the initial observation state and buffer; Use the long short-term memory network LSTM to take the observed state as input and solve the Q value of all actions; Select actions using the greedy strategy method e-greedy ; Execution Action , and receive the next state respectively , end identifier and rewards ; Will Store into the buffer according to priority, where Indicates the current action; According to the sample optimization level, extract the unexperienced samples from the buffer area; According to the final optimization problem, combined with the extracted unexperienced samples, the target value is calculated and the loss function is used Update the long short-term memory network parameters θ; Through continuous iterative optimization, the jamming strategy for jamming drones is obtained, and the drone physical layer transmission for overall security is completed.
[0020] In this embodiment, Figure 2 As shown in the figure, the system is a UAV communication system composed of a communication UAV (S-UAV), a user (Bob), a friendly jammer UAV (J-UAV) and a potential eavesdropper Eves (Eve). In this system, the communication UAV is responsible for data collection and transmitting confidential information to ground users, ground users are responsible for performing tasks according to predetermined trajectories, and the friendly jammer UAV interferes with the potential eavesdropper Eves by sending artificial noise.
[0021] Communication model: The line-of-sight (LOS) transmission probability between the drone and the ground unit is calculated as follows: .
[0022] Accordingly, the NLoS transmission probability is: .
[0023] The average path loss is a key indicator for evaluating the link quality between the UAV and the ground equipment. Its mathematical expression is: ;in, Indicates drone k With ground node u The average path loss between represents the line-of-sight transmission probability between the UAV and the ground unit, Indicates drone k With ground node u The Euclidean distance between and Represent the path loss index of LoS and NLoS respectively. NLoS represents non-line-of-sight link, and LoS represents line-of-sight link. and are attenuation coefficients, represents the transmission probability of NLoS, and All represent constants, Indicates communication drone k The vertical distance to the ground, Indicates ground user u With communication drone k The projected distance between them on the ground plane.
[0024] The instantaneous signal-to-noise ratio of the user and the potential eavesdropper Eves is: ; ;in, represents the user signal-to-noise ratio, represents the instantaneous signal-to-noise ratio of the potential eavesdropper Eves, and They represent the transmission power of the communication UAV and the transmission power of the friendly interference UAV respectively. and They represent the path losses from the communication UAV and the jamming UAV to the potential eavesdropper Eves, Represents the noise power of the eavesdropper Eves.
[0025] Security metrics: In order to accurately measure the security of the system, two thresholds are set and To evaluate the reception performance of the user and potential eavesdropper Eves. Specifically, if the signal-to-noise ratio of the receiver is higher than the threshold, it means that the receiver can decode the information; otherwise, it means that the receiver cannot obtain the content of the information. Therefore, in the absence of friendly jamming drones, Location collection It is considered a threat area. Under the interference of friendly jammer drones, the location collection Into The location is called the safe region SR; the number and location set of grid points in the SR at the current moment The ratio of the number of grid points in is regarded as the safe area ratio at that moment.
[0026] Friendly jammer J-UAV kinematic description: Will the friendly jammer J-UAV be at the moment t The horizontal acceleration vector is defined as , which can be expressed as: ;in, and Separate moments t hour x and y The acceleration component in the vertical direction is defined as . Similarly, friendly interference drones t The instantaneous velocity at time is defined as the horizontal vector and vertically Then the next moment t The speed of +1 can be calculated by the following formula: ; ;in, Indicates friendly interference with drones next moment The horizontal instantaneous velocity vector, Indicates the time slot size, Indicates the next moment Vertical speed.
[0027] time t The average speed is recorded as: ; ;in, and represent the average velocities in the horizontal and vertical directions respectively.
[0028] According to the friendly jammer drone t The position at the moment and the above motion state can be calculated by the following formula: t +1 position: ; ;in, and Respectively represent the friendly interference drone at time t +1 and time t location, Indicates time t +1 Jammer Drone Altitude, and Respectively represent the friendly interference drone at time t +1 and time t Height, Indicates friendly interference drone at all times t The vertical instantaneous velocity, Indicates friendly interference drone at all times t The vertical acceleration vector, Indicates the time slot size, q Refers to x or y, used to simplify x , y The coordinate position calculation formula is: Indicates friendly interference drone moment t exist x or y The component of velocity in the direction.
[0029] Combined with the above kinematic equations, it can be seen that by adjusting the acceleration of the friendly jamming UAV in each specified time period, the expected results of controlling the speed and trajectory of the friendly jamming UAV can be achieved.
[0030] In order to represent the energy consumption of the entire communication UAV flight, the present invention adopts a three-dimensional UAV propulsion energy model: .
[0031] In a given space and time range, let the acceleration of the friendly jammer be a and the power be p , both of which are optimizable decision variables. The optimization goal of the present invention is to maximize the total cumulative safe area ratio in the time dimension by finely regulating the acceleration parameters and power output of the friendly interference drone. The specific optimization problem is expressed as follows: ; in, Indicates the upper and lower limits of the flight range of friendly interference drones. c Indicates the minimum distance between communicating UAVs to avoid collision.
[0032] In order to improve the efficiency of the optimization procedure, the present invention introduces a vector: ;in, Indicates time t The vector of Indicates the number of i elements, Indicates n elements, T Represents the transpose of a vector.
[0033] Use this vector to show the discrete threat locations at time t Safety performance on each element express Neidi the safety of the potential location, express t The number of non-zero elements in the vector indicates the number of locations that do not meet the safety condition.
[0034] In summary, the present invention introduces a new objective function, which makes the vector The number of non-zero elements in represents SR, thus transforming the original problem into the following problem: ; ; ; ; ; ; ; ; ; ; ; where a represents the acceleration of the friendly jammer drone, P Indicates the power of friendly interference drones, Indicates time t In the vector i elements, express t The threat zone at all times, TIndicates the total time, Indicates time t The vector of Indicates time t The length of the vector, represents the instantaneous signal-to-noise ratio of the potential eavesdropper Eves, and Respectively represent the friendly interference drone at time t of x Coordinates and y coordinate, and Respectively represent the friendly interference drone at time t The power and height, and Indicates the potential eavesdropper Eves x Coordinates and y coordinate, represents the threshold of Eves signal-to-noise ratio of potential eavesdroppers, represents the user signal-to-noise ratio threshold, represents the user signal-to-noise ratio, Indicates friendly interference drone at all times t The horizontal acceleration vector, Indicates friendly interference drone at all times t The vertical acceleration vector, and denote the upper limits of acceleration and total energy respectively, Indicates friendly interference drone at all times t The horizontal instantaneous velocity, Indicates friendly interference drone at all times t The vertical instantaneous velocity, Indicates the maximum speed of the friendly jammer drone. Represents the three-dimensional UAV propulsion energy model, , and Respectively indicate time t Communication drone x coordinate, y Coordinates and height, Indicates the minimum distance between communicating UAVs to avoid collision. , , , , and Indicates the upper and lower limits of the flight range of friendly interference drones. Indicates the time slot size, and They represent the profile power and induced power of the blade in the hovering state, Indicates the continuous power during the entire descent and ascent process, Indicates friendly interference drone at all times t The average speed, is the tip speed of the rotor blade, is the drag coefficient of the fuselage, represents the air density, Indicates the strength of the rotor. represents the area of the rotor disk, represents the average induced velocity of the rotor in hover, Indicates friendly interference drone at all times t The average vertical velocity.
[0035] Since this problem has long-term goals and non-convex constraints, traditional optimization algorithms are difficult to handle. Therefore, the present invention uses a deep reinforcement learning algorithm, which can autonomously learn complex strategies and manage high-dimensional, nonlinear features to solve complex problems in dynamic environments. Although the training process of the deep reinforcement learning DRL algorithm takes a certain amount of time, the trained model can quickly generate optimized or feasible solutions using existing data, thereby quickly adapting and providing reliable results in real-time or dynamic environments.
[0036] In view of the fact that the problem of the present invention is to consider the overall time security, the convolutional neural network (CNN) in the deep reinforcement learning algorithm DDQN cannot capture long-term dependencies when processing time series problems. Therefore, the present invention replaces the original convolutional neural network with a long short-term memory network (LSTM). The long short-term memory network is a special recursive neural network (RNN) that aims to solve the long-term dependency problem faced by the standard recursive neural network when processing continuous data. Standard recursive neural networks have difficulty capturing long-term dependencies and often lead to gradient vanishing or exploding problems when processing long sequences. The long short-term memory network solves this problem through a gating mechanism (including input gates, forget gates, and output gates). This mechanism enables the long short-term memory network to effectively capture long-term dependencies and avoid the gradient vanishing or exploding problems that occur in traditional recursive neural networks when processing long sequences. The core state update process of the long short-term memory network can be expressed as: ; ;in, , and They represent the forget gate, input gate, and output gate that control the update of the storage unit. The current storage state and candidate storage state calculated according to the current input state and the previous state are respectively represented as and , represents the hidden state of the current time slot, express t -1 The storage status at time.
[0037] LSTM networks can effectively manage information flow and ensure that the model captures important long-term dependencies. When used in conjunction with the deep reinforcement learning algorithm DDQN, LSTM networks can improve the model's ability to capture temporal relationships and make accurate decisions based on long-term dependencies, especially when the model needs to handle states with high uncertainty during the decision-making process.
[0038] In the traditional experience replay mechanism of the deep reinforcement learning algorithm DDQN, all experience samples are selected for training with the same probability. However, this uniform sampling cannot distinguish the importance of experience samples, which may lead to ignoring key experience, reducing training efficiency and prolonging convergence time. The Prioritized Experience Replay PER mechanism is an improved experience replay technique that assigns a priority to each experience sample based on the time difference (TD error) to measure its potential to improve the strategy. The higher the priority of the sample, the greater the probability of being selected.
[0039] The priority of each experience is calculated as follows: ;in, Indicates priority, Indicates the time difference, Represents a constant used to adjust the importance of the optimization level.
[0040] Since the priority-based sampling method causes an imbalance in sample selection, it can no longer be assumed that all samples contribute equally to policy optimization. This deviation may reduce the representativeness of the samples, thereby affecting the accuracy of gradient updates and learning effects. Therefore, it is necessary to correct the errors in gradient updates. In this case, importance sampling weights need to be introduced into the strategy to adjust the deviations caused by non-uniform sampling. The importance sampling weight is defined as: ;in, N represents the total number of samples in the experience pool, represents the probability of selecting a sample, represents a hyperparameter controlling the magnitude of the correction. Therefore, when updating the gradient, the TD error is updated according to the following formula: .
[0041] Then, the priority of the samples is re-evaluated according to the new TD error. This update mechanism enables the model to dynamically adjust the sample selection strategy. From the above process, it can be seen that the priority experience replay strategy PER can effectively increase the sampling frequency of key experiences and make the training process of deep reinforcement learning more efficient.
[0042] In summary, compared with the prior art, the present invention has the following advantages: By taking the security performance of multiple time slots as the overall optimization target, the present invention can achieve higher long-term security. Specifically, the present invention not only considers the instantaneous confidentiality capacity of a single time slot, but also ensures the continuous security of the entire transmission process through cumulative security optimization, effectively preventing the risk of information leakage caused by instantaneous security performance fluctuations; The present invention adopts an improved dual-depth Q network algorithm and introduces a long short-term memory network to enable the system to effectively capture the time correlation characteristics in the UAV communication environment. This improvement enables the algorithm to make full use of historical information when making transmission parameter decisions, thereby improving the accuracy of decision-making; The present invention innovatively introduces a priority experience replay mechanism, which significantly improves the convergence efficiency of the algorithm by giving priority to learning important samples. This mechanism enables the system to learn effective security transmission strategies more quickly and better respond to sudden security threats. The method proposed in the present invention has strong practicability and adaptability. By comprehensively considering the cumulative safety performance and learning efficiency, the method can quickly establish a safe transmission strategy in various complex UAV communication scenarios and has low computational complexity.
Claims
1. A physical layer transmission method for drones with overall security, characterized in that: The following steps are involved: S1. Obtain the location and power information of the communication drone at each moment, as well as the location of the user at each moment; S2. According to the current position and power information of the communication drone, the Euclidean distance between the communication drone and the ground node is calculated when the instantaneous signal-to-noise ratio of the potential eavesdropper Eves is equal to the threshold of the signal-to-noise ratio of the potential eavesdropper Eves, and the threat area with the communication drone on the ground as the center and R as the radius is obtained, where the Euclidean distance is recorded as R; S3, discretize the threat area in a grid manner and construct a vector of length n, where each element in the vector corresponds to a position, indicating whether the position is a safe area; S4. Establish an optimization problem for the safety of elements and positions in the vector at each moment, and sum the optimization problems of all time periods, and use the sum result as the final optimization problem. The objective function is to maximize the sum of the safe area ratios of all time periods, and the position is the threatened area. The position of the discretized point in represents the position of the potential eavesdropper Eves; S5. According to the final optimization problem, the long short-term memory network LSTM and the priority experience replay PER mechanism are used in combination with the deep reinforcement learning algorithm DDQN to obtain the interference strategy for jamming drones and complete the drone physical layer transmission for overall security.
2. The overall security-oriented UAV physical layer transmission method according to claim 1 is characterized in that: The expression of the Euclidean distance is as follows: in, Indicates drone k With ground node u The average path loss between represents the line-of-sight transmission probability between the UAV and the ground unit, Indicates drone k With ground node u The Euclidean distance between and Represent the path loss index of LoS and NLoS respectively. NLoS represents non-line-of-sight link, and LoS represents line-of-sight link. and are attenuation coefficients, represents the transmission probability of NLoS, and All represent constants, Indicates communication drone k The vertical distance to the ground, Indicates ground user u With communication drone k The projected distance between them on the ground plane.
3. The overall safety-oriented UAV physical layer transmission method according to claim 1 is characterized in that: The expression of the instantaneous signal-to-noise ratio of the potential eavesdropper Eves is as follows: in, represents the instantaneous signal-to-noise ratio of the potential eavesdropper Eves, and They represent the transmission power of the communication UAV and the transmission power of the friendly interference UAV respectively. and They represent the path losses from the communication UAV and the jamming UAV to the potential eavesdropper Eves, Represents the noise power of the eavesdropper Eves.
4. The overall safety-oriented UAV physical layer transmission method according to claim 1, characterized in that: The expression of the vector is as follows: in, Indicates time t The vector of Indicates the number of i elements, Indicates n elements, T Represents the transpose of a vector.
5. The overall safety-oriented UAV physical layer transmission method according to claim 1, characterized in that: The final optimization problem is expressed as follows: Where a represents the acceleration of the friendly jammer drone, P Indicates the power of friendly interference drones, Indicates time t In the vector i elements, express t The threat zone at all times, T Indicates the total time, Indicates time t The vector of Indicates time t The length of the vector, represents the instantaneous signal-to-noise ratio of the potential eavesdropper Eves, and Respectively represent the friendly interference drone at time t of x Coordinates and y coordinate, and Respectively represent the friendly interference drone at time t The power and height, and Indicates the potential eavesdropper Eves x Coordinates and y coordinate, represents the threshold of Eves signal-to-noise ratio of potential eavesdroppers, represents the user signal-to-noise ratio threshold, represents the user signal-to-noise ratio, Indicates friendly jamming of drones at all times t The horizontal acceleration vector, Indicates friendly interference drone at all times t The vertical acceleration vector, and denote the upper limits of acceleration and total energy respectively, Indicates friendly jamming of drones at all times t The horizontal instantaneous velocity, Indicates friendly jamming of drones at all times t The vertical instantaneous velocity, Indicates the maximum speed of the friendly jammer drone. Represents the three-dimensional UAV propulsion energy model, , and Respectively indicate time t Communication drone x coordinate, y Coordinates and height, Indicates the minimum distance between communicating UAVs to avoid collision. , , , , and Indicates the upper and lower limits of the flight range of friendly interference drones. Indicates the time slot size, and They represent the profile power and induced power of the blade in the hovering state, Indicates the continuous power during the entire descent and ascent process, Indicates friendly jamming of drones at all times t The average speed, is the tip speed of the rotor blade, is the drag coefficient of the fuselage, represents the air density, Indicates the strength of the rotor. represents the area of the rotor disk, represents the average induced velocity of the rotor in hover, Indicates friendly interference drone at all times t The average vertical velocity.
6. The overall safety-oriented UAV physical layer transmission method according to claim 1 is characterized in that: The S5 comprises the following steps: Initialize the parameters of the current long short-term memory network LSTM; Get the initial observation state and buffer; Use the long short-term memory network LSTM to take the observed state as input and solve the Q value of all actions; Select actions using the greedy strategy method e-greedy ; Execution Action , and receive the next state respectively , end identifier and rewards ; Will Store into the buffer according to priority, where Indicates the current action; According to the sample optimization level, unexperienced samples are extracted from the buffer, where the expression of sampling weight is as follows: in, represents the sampling weight, represents the total number of samples in the experience pool, represents the probability of selecting a sample, represents the hyperparameter used to control the magnitude of the correction, k represents the index of the sample, Indicates k The priority of each sample; According to the final optimization problem, combined with the extracted unexperienced samples, the target value is calculated and the loss function is used Update the long short-term memory network parameters θ; Through continuous iterative optimization, the jamming strategy for jamming drones is obtained, and the drone physical layer transmission for overall security is completed.
7. The overall safety-oriented UAV physical layer transmission method according to claim 6, characterized in that: The priority expression is as follows: in, Indicates priority, Indicates the time difference, Represents a constant used to adjust the importance of the optimization level; Time difference The update expression is as follows: 。