CN112243252B - Safety transmission enhancement method for relay network of unmanned aerial vehicle - Google Patents

Abstract

The invention belongs to the technical field of unmanned aerial vehicle communication, and discloses a safe transmission enhancing method for an unmanned aerial vehicle relay network, which comprises the steps of establishing a communication system from a ground source node to a destination node by taking an unmanned aerial vehicle as a relay and a channel model, and calculating the private rate of a transmission link from the source node to the destination node according to signals received by the unmanned aerial vehicle and the destination node in two time slots, namely a half-duplex mode; constructing an optimization model which takes the maximum privacy rate as a target function and designs the position constraint condition of the interference power unmanned aerial vehicle; traversing the destination node D under the condition that the position of the unmanned aerial vehicle is fixed, and optimizing an interference power distribution scheme by the coordinate position of the eavesdropper E so as to maximize the privacy rate; and obtaining an optimal position generation data set of the unmanned aerial vehicle by using an exhaustive search method, constructing and training a DNN model, and finding the optimal position of the unmanned aerial vehicle by using the high calculation efficiency of the DNN. The unmanned aerial vehicle is convenient to deploy and is not limited by complex terrains and obstacles; the communication applicability is strong, and the information transmission quality is high.

Description

A security transmission enhancement method for UAV relay network

technical field

The invention belongs to the technical field of unmanned aerial vehicle communication, and in particular relates to a safety transmission enhancement method for an unmanned aerial vehicle relay network.

Background technique

Present: The rapid development of wireless communication technology has brought various conveniences to people. At the same time, more and more attention is paid to wireless information security. The wireless communication security problem is essentially the broadcast nature of wireless communication. Therefore, how to ensure the security of wireless communication from the physical layer is the key to solving this problem. Physical layer security is formulated under the eavesdropping channel model, which achieves perfect secrecy in information theory when the performance of the eavesdropping channel model is low compared to the legitimate channel.

With the rapid development and deployment of 5G networks, academia and industry have shown great interest in unmanned aerial vehicles (UAVs). UAV has the advantages of low cost, convenient deployment, high flexibility and adaptability, so it has a wide range of applications in military and civilian fields. In particular, it can be used as a wireless sensor node, relay station or mobile base station, etc. Drones are often used to assist communications in complex and changing environments. However, due to the uncertain factors in the flight environment, the reliability of the UAV communication system may be reduced, thereby affecting the communication quality. Therefore, it is of great significance to improve the network performance involving UAVs.

On the other hand, deep neural networks (DNNs) have also made significant progress while advancing various applications. Deep learning techniques have also been applied in wireless communication research, such as transmit power control and channel estimation, while bringing significant performance improvements. DNNs enable us to formulate effective policies for communication systems based on data and knowledge. Furthermore, using a well-trained DNN model can significantly reduce the complexity of the actual implementation. Due to the significant advantages of DNNs, deep learning has been widely used in UAV communication to facilitate design and improve performance.

Through the above analysis, the existing problems and defects in the prior art are:

(1) UAVs are often used to assist communication in complex and changing environments. Considering the complex outdoor environment, due to uncertain factors in the flight environment, such as: high-rise buildings may block the LOS communication link between the ground user and the BS, the reliability of the UAV communication system may be reduced, thereby affecting the communication quality.

(2) Existing physical layer security In the traditional point-to-point information transmission technology, the main security problem is that the nodes at both ends of the communication are eavesdropped. Compared with the traditional point-to-point information transmission technology, the cooperative relay network will face more severe security problems, because in the process of multi-path transmission through the relay node, the expansion of the information diffusion surface will increase the probability of eavesdroppers to eavesdrop on confidential information. .

(3) The existing security schemes lack consideration of the potential uncertainty of the actual network.

The difficulty of solving the above problems and defects is: the existing security mechanism is based on the key system, and the physical layer security is a brand-new solution. In the scenario considered by the present invention, the UAV relay network adopts a physical layer security mechanism, and it is necessary to reasonably allocate the interference signal transmitted by the legitimate receiver. In addition, a deep neural network model is designed to improve network performance.

The significance of solving the above problems and defects is as follows: the present invention adopts a physical layer security scheme, which does not require a key and has low complexity; The privacy rate of transmission; also, consider legitimate receivers transmitting a jamming signal independent of the source signal to counter eavesdropping. In response to this problem, it is first decomposed into two sub-problems to solve the problem of jamming strategy and UAV position deployment respectively. Then, based on the effective binary search method, the interference problem is solved, and then the UAV deployment is solved with the DNN framework. Finally, the simulation results are given to verify the effectiveness of the proposed scheme.

SUMMARY OF THE INVENTION

Aiming at the problems existing in the prior art, the present invention provides a security transmission enhancement method for the UAV relay network.

The present invention is implemented in this way, a method for enhancing the safety transmission of the UAV relay network, and the method for enhancing the safety transmission for the UAV relay network includes:

Establish a UAV relay communication system, the goal of which is to jointly optimize the jamming power and the UAV relay position to maximize the privacy rate of the system;

Establish the channel model of the ground source node-to-destination node communication system with the UAV as the relay, and in the two-slot-half-duplex mode, according to the UAV received signal and the received signal of the destination node, calculate the source node to The privacy rate of the destination node transmission link;

Build an optimization model with the objective function when the privacy rate reaches the maximum, and design the interference power and UAV position constraints;

Under the condition that the position of the UAV is fixed, traverse the destination node D and the coordinate position of the eavesdropper E, and use the binary search algorithm to optimize the interference power allocation scheme to maximize the privacy rate;

Under the condition of optimal interference power, use the exhaustive search method to obtain the optimal position of the UAV, generate a data set, build and train the DNN model, apply it to the test set, and use the high computational efficiency of DNN to find the optimal position of the UAV. The best location to maximize the privacy rate and achieve secure transmission.

Further, the UAV relay network-oriented security transmission enhancement method establishes a UAV relay communication system, the goal of which is to jointly optimize the interference power and the UAV relay position to maximize the privacy rate of the system; A UAV relay communication system composed of source node S, destination node D, UAV relay R and eavesdropper E, S sends a signal to the UAV relay, and R amplifies and forwards the signal to D.

Further, the security transmission enhancement method for the UAV relay network establishes the channel model of the ground source node to destination node communication system with the UAV as the relay and in the two time slot-half duplex mode, according to The UAV receives the signal and the received signal of the destination node, and calculates the privacy rate of the transmission link from the source node to the destination node; the channel model of the communication system from the ground source node to the destination node:

where c is the speed of light, α _G is the path loss index of the ground communication link, f _c is the carrier frequency, d _eg is the distance between the eavesdropper and the ground user, and σ _G is the shadow fading variable of the channel;

The channels related to the UAV include the line-of-sight LoS group and the non-line-of-sight NLoS group, and the probability of having a LOS connection between the ground user and the UAV is:

where A and B are constants depending on the environment, (x _u , y _u ) is the position of the drone in the horizontal dimension, h is the height of the drone, (x _g , y _g ) is the position of the ground user, The probability of NLoS is P _NLoS =1-P _LoS , and the path loss models of LoS and NLoS links are:

α_L和α_N是LoS和NLoS信道的路径损耗指数，η_LoS和η_NLoS分别是LoS和NLoS的平均额外损失；概率平均路径损耗是在LoS和NLoS条件下平均得出：where d is the transmission distance,

α _L and α _N are the path loss exponents for LoS and NLoS channels, η _LoS and η _NLoS are the average additional losses for LoS and NLoS, respectively; the probabilistic average path loss is averaged under LoS and NLoS conditions:

h _ij =P _LoS L _LoS +P _NLoS L _NLoS , i∈{R,D}, j∈{R,E,D};

In the first time slot, S transmits its signal to the drone relay R, which is also eavesdropped by the eavesdropper E; at the same time, D emits artificial noise, which confuses the eavesdropper; in the second time slot, S is silent state, the drone relay amplifies the received signal and sends it to D, D is also eavesdropped by eavesdroppers, denoted by z _S and z _J for confidential signals and cooperative jamming signals from S and D, both z _S and z _J are normalized powers, ie |z _S | ² =1 and |z _J | ² =1, where |·| represents the absolute value, the signals received at the first time slot, UAV and E are:

是R和D处的复杂加性白高斯噪声AWGN，遵循均值为零且方差为

的复高斯分布；where P _S and P _D are the transmit powers from S and D, respectively, n _R and

is the complex additive white Gaussian noise AWGN at R and D, following zero mean and variance

The complex Gaussian distribution of ;

In the second time slot, R amplifies and forwards the received signal to D with an amplification factor β, where P _R is the transmit power of R, and β is expressed as:

Then, the signals received at D and E are:

是D和E处的复杂加性高斯白噪声，D有效地去除该项，并且得到D处的接收信号为：where n _D and

is the complex additive white Gaussian noise at D and E, D effectively removes this term, and the received signal at D is obtained as:

Calculate the privacy rate of the transmission link from the source node to the destination node, and redefine the channel-to-noise ratio:

In the first time slot, the SINR of the eavesdropping link is:

The instantaneous SINRs of D and E are:

The eavesdropper adopts the maximum proportion combined MRC method, and the signal-to-interference-noise ratio SINR at the eavesdropping node E is:

In the relay system based on physical layer security, the private rate is obtained as follows:

Further, the security transmission enhancement method for the UAV relay network constructs an optimization model that takes the privacy rate reaching the maximum as the objective function, and designs the interference power and UAV position constraints:

where (x _u , y _u ) represents the horizontal position of the UAV in the considered area A.

Further, the security transmission enhancement method for the UAV relay network is determined under the condition that the UAV position is fixed, traverse the destination node D, the coordinate position of the eavesdropper E, and use the binary search algorithm to optimize the interference power allocation scheme, To maximize privacy rate: In the case of fixed drone placement, the optimization problem of jamming power is as follows:

Further, the security transmission enhancement method for the UAV relay network uses the binary search algorithm to obtain the power obtained by solving the power distribution scheme of the current fixed UAV position, and the optimal interference power is

Step 1: Initialization: set the minimum and maximum values of _PD , given as P _min and P _max , where P _min =0; define a sufficiently small threshold ε;

R_S由第二步给出，如果

则P_max＝P^*，否则P_min＝P^*；Step 2: Order

R _S is given by the second step, if

Then P _max =P ^* , otherwise P _min =P ^* ;

Step 3: If |P _max -P _min |<ε, obtain the optimal interference power P ^* , and complete the power allocation strategy of the UAV relay communication system; otherwise, repeat step 2.

Further, under the condition of optimal interference power, the security transmission enhancement method for UAV relay network uses an exhaustive search method to obtain the optimal position of UAV, generates a data set, constructs and trains a DNN model, and is applied to. The test set uses the high computing efficiency of DNN to find the best position of the drone to maximize the privacy rate and achieve safe transmission;

To obtain the optimal jamming power, the optimization problem of UAV position deployment is obtained as follows:

Further, the security transmission enhancement method for the UAV relay network uses a neural network model with multiple hidden layers and a large amount of training data to learn network characteristics to solve the optimization problem of the UAV position deployment, and realize the DNN-based wireless network. Human-machine deployment scheme; build a fully connected DNN model with one input layer, two hidden layers and one output layer, learn its input/output relationship by training the DNN, apply it to the test set, take advantage of the high computational efficiency of DNN, find The optimal position of the UAV maximizes the privacy rate and realizes the safe transmission of the system. Consider a system located in the range of 200 meters × 200 meters, the fixed height h of the UAV is 150 meters, and the source node has a fixed position (0 ,0,0), first of all, the positions of S, D, E are fixed, the horizontal position of the UAV is traversed, and the optimal interference power P ^* is obtained by using the binary search algorithm. Under the condition of the optimal interference power, the maximum privacy rate Corresponding to the optimal position of the UAV; then, by traversing the coordinate positions of D and E, find the optimal position of the UAV corresponding to different D and E, and construct a data set;

无人机最优位置的坐标(x_u,y_u)作为标签输出，表示为

其中i∈{1,2,···,N}，把Q＝[q₁,q₂,···,q_N]作为DNN的输入，q_i中的每一项对应一个输入神经元，取无人机的最优位置U＝[u₁,u₂,···,u_N]作为DNN模型的输出，u_i中的每一项对应一个输出神经元，最后，将经过良好训练的DNN模型用于测试集，快速有效的找到无人机的最优位置，求解无人机位置优化的问题。In the input layer of the DNN model, the coordinates (x _d , y _d ) and (x _e , y _e ) of the target node and the eavesdropper are reshaped into a 4×1 data sample, denoted as

The coordinates (x _u , y _u ) of the optimal position of the UAV are output as labels, which are expressed as

where i∈{1,2,...,N}, take Q=[q ₁ ,q ₂ ,...,q _N ] as the input of DNN, each item in q _i corresponds to an input neuron, Take the optimal position U=[u ₁ , u ₂ , . . . , u _N ] as the output of the DNN model, each item in _ui corresponds to an output neuron, and finally, the well-trained The DNN model is used in the test set to quickly and effectively find the optimal position of the UAV and solve the problem of UAV position optimization.

Another object of the present invention is to provide a computer device, the computer device includes a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the following step:

Establish a UAV relay communication system, the goal of which is to jointly optimize the jamming power and the UAV relay position to maximize the privacy rate of the system;

Establish the channel model of the ground source node-to-destination node communication system with the UAV as the relay, and in the two-slot-half-duplex mode, according to the UAV received signal and the received signal of the destination node, calculate the source node to The privacy rate of the destination node transmission link;

Build an optimization model with the objective function when the privacy rate reaches the maximum, and design the interference power and UAV position constraints;

Under the condition that the position of the UAV is fixed, traverse the destination node D and the coordinate position of the eavesdropper E, and use the binary search algorithm to optimize the interference power allocation scheme to maximize the privacy rate;

Under the condition of optimal interference power, use the exhaustive search method to obtain the optimal position of the UAV, generate a data set, build and train the DNN model, apply it to the test set, and use the high computational efficiency of DNN to find the optimal position of the UAV. The best location to maximize the privacy rate and achieve secure transmission.

Another object of the present invention is to provide a UAV relay network-oriented security transmission enhancement system implementing the UAV relay network-oriented security transmission enhancement method. Transmission enhancement systems include:

The UAV relay communication system establishment module is used to establish the UAV relay communication system, and its goal is to jointly optimize the interference power and the UAV relay position to maximize the privacy rate of the system;

The channel model and slot establishment module are used to establish the channel model of the ground source node-to-destination node communication system with the UAV as the relay, and in the two-slot-half-duplex mode, according to the UAV received signal and the purpose The received signal of the node calculates the privacy rate of the transmission link from the source node to the destination node;

The objective function building module is used to construct an optimization model that takes the privacy rate reaching the maximum as the objective function, design interference power, and UAV position constraints;

The privacy rate maximization module is used to traverse the coordinate position of the destination node D and the eavesdropper E under the condition that the position of the drone is fixed, and use the binary search algorithm to optimize the interference power allocation scheme to maximize the privacy rate;

The safe transmission module is used to obtain the optimal position of the UAV using the exhaustive search method under the condition of optimal interference power, generate a data set, build and train a DNN model, apply it to the test set, and utilize the high computational efficiency of DNN, Find the best location for the drone to maximize privacy rates for secure transfers.

Combined with all the above technical solutions, the advantages and positive effects of the present invention are as follows: the security transmission enhancement method for the UAV relay network designed by the present invention, first of all, when the UAV position is fixed, the receiver The power of the transmitted interference signal is allocated to maximize the privacy rate; and for the deployment of the UAV, a DNN model is constructed under the optimal interference power, and the solution is carried out to ensure the best communication effect, thus ensuring communication. reliability; in line with the actual situation. UAVs are easy to deploy, maneuverable, and not restricted by complex terrain and obstacles; they have low cost and high reliability; they have strong practical communication applicability and high information transmission quality.

Description of drawings

In order to explain the technical solutions of the embodiments of the present application more clearly, the following will briefly introduce the drawings that need to be used in the embodiments of the present application. Obviously, the drawings described below are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can also be obtained from these drawings without creative effort.

FIG. 1 is a flowchart of a method for enhancing security transmission for a UAV relay network provided by an embodiment of the present invention.

2 is a schematic structural diagram of a UAV relay network-oriented security transmission enhancement system provided by an embodiment of the present invention;

In Figure 2: 1. UAV relay communication system building module; 2. Channel model and time slot building module; 3. Objective function building module; 4. Privacy rate maximization module; 5. Secure transmission module.

FIG. 3 is a schematic diagram of the relationship between the privacy rate of the UAV relay network and the position of the eavesdropper according to an embodiment of the present invention.

FIG. 4 is a schematic diagram of the relationship between the privacy rate and the relay power of the UAV relay network according to an embodiment of the present invention.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, but not to limit the present invention.

Aiming at the problems existing in the prior art, the present invention provides a method and system for enhancing safety transmission for a UAV relay network. The present invention will be described in detail below with reference to the accompanying drawings.

As shown in Figure 1, the security transmission enhancement method for the UAV relay network provided by the present invention includes the following steps:

S101: Establish a UAV relay communication system, the goal of which is to jointly optimize the interference power and the UAV relay position to maximize the privacy rate of the system;

S102: Establish a channel model of the ground source node-to-destination node communication system with the UAV as a relay, and in the two-time-slot-half-duplex mode, calculate the source according to the UAV received signal and the received signal of the destination node The private rate of the transmission link from the node to the destination node;

S103: Construct an optimization model with the objective function when the privacy rate reaches the maximum, design the interference power, and the position constraints of the UAV;

S104: Under the condition that the position of the drone is fixed, traverse the destination node D and the coordinate position of the eavesdropper E, and use the binary search algorithm to optimize the interference power allocation scheme to maximize the privacy rate;

S105: Under the condition of optimal interference power, use the exhaustive search method to obtain the optimal position of the UAV, generate a data set, build and train a DNN model, apply it to the test set, and use the high computational efficiency of DNN to find the UAV The best location to maximize the privacy rate and achieve secure transmission.

The security transmission enhancement method for the UAV relay network provided by the present invention can also be implemented by those skilled in the art by other steps. The security transmission enhancement method for the UAV relay network provided by the present invention in FIG. 1 is only Just a specific example.

As shown in Figure 2, the security transmission enhancement system for the UAV relay network provided by the present invention includes:

The UAV relay communication system establishment module 1 is used to establish a UAV relay communication system, the goal of which is to jointly optimize the interference power and the UAV relay position to maximize the privacy rate of the system;

The channel model and slot establishment module 2 are used to establish the channel model of the ground source node-to-destination node communication system with the UAV as the relay, and in the two time slot-half duplex mode, according to the UAV received signal and The received signal of the destination node is used to calculate the privacy rate of the transmission link from the source node to the destination node;

Objective function building module 3, which is used to construct an optimization model that takes when the privacy rate reaches the maximum as the objective function, design interference power, and UAV position constraints;

The privacy rate maximization module 4 is used to traverse the coordinate position of the destination node D and the eavesdropper E under the condition that the position of the drone is fixed, and use the binary search algorithm to optimize the interference power allocation scheme to maximize the privacy rate;

The safe transmission module 5 is used to obtain the optimal position of the UAV using the exhaustive search method under the condition of optimal interference power, generate a data set, build and train a DNN model, apply it to the test set, and utilize the high computational efficiency of DNN , to find the best position of the drone to maximize the privacy rate and achieve safe transmission.

The technical solutions of the present invention will be further described below with reference to the accompanying drawings.

The specific implementation steps of the security transmission enhancement method for the UAV relay network provided by the present invention are explained as follows:

The first step: establish a UAV relay communication system, the goal of which is to jointly optimize the interference power and the UAV relay position to maximize the privacy rate of the system; consider a source node (S), destination node (D), UAV relay communication system composed of UAV relay (R) and eavesdropper (E). S sends a signal to the UAV relay, then R amplifies and forwards the signal to D. For simplicity, the present invention takes into account that all network nodes are equipped with one antenna. In addition, considering that there is no direct link between S and D, the optimization goal of the UAV relay communication system is to maximize the privacy rate and achieve secure transmission by jointly optimizing the interference power distribution and UAV position deployment.

Step 2: Establish the channel model of the communication system from the ground source node to the destination node with the UAV as the relay and in two time slots - half-duplex mode, according to the UAV received signal and the received signal of the destination node , calculate the privacy rate of the transmission link from the source node to the destination node; the channel model of the communication system from the source node to the destination node on the ground:

where c is the speed of light, α _G is the path loss index of the ground communication link, f _c is the carrier frequency, d _eg is the distance between the eavesdropper and the ground user, and σ _G is the shadow fading variable of the channel;

The channels associated with UAVs include line-of-sight (LoS) groups and non-line-of-sight (NLoS) groups. Probability of LOS connection between ground user and UAV

where A and B are constants depending on the environment, (x _u , y _u ) is the position of the drone in the horizontal dimension, h is the height of the drone, and (x _g , y _g ) is the position of the ground user. Also, the probability of NLoS is P _NLoS =1-P _LoS . In addition, the path loss models for LoS and NLoS links are respectively

此外，α_L和α_N是LoS和NLoS信道的路径损耗指数。η_LoS和η_NLoS分别是LoS和NLoS的平均额外损失。本发明考虑概率平均路径损耗是在LoS和NLoS条件下平均得出的where d is the transmission distance,

Furthermore, _αL and _αN are the path loss indices for LoS and NLoS channels. _ηLoS and ηNLoS are the average additional losses of LoS and _NLoS , respectively. The present invention considers that the probability-averaged path loss is averaged under LoS and NLoS conditions

h _ij =P _LoS L _LoS +P _NLoS L _NLoS , i∈{R,D}, j∈{R,E,D};

In the first time slot, S transmits its signal to the drone relay (R), which is also tapped by the eavesdropper (E). At the same time, D emits artificial noise to confuse eavesdroppers. In the second time slot, S is in a silent state, and the drone relay amplifies the received signal and sends it to D, which is also eavesdropped by eavesdroppers. The present invention uses z _S and z _J to denote the confidential signal and cooperative jamming signal from S and D. Both z _S and z _J are normalized powers, ie |z _S | ² =1 and |z _J | ² =1, where |·| represents an absolute value. Therefore, in the first time slot, the signals received at UAV and E are

是R和D处的复杂加性白高斯噪声(AWGN)，遵循均值为零且方差为

的复高斯分布。where P _S and P _D are the transmit powers from S and D, respectively, n _R and

is complex additive white Gaussian noise (AWGN) at R and D, obeying zero mean and variance

complex Gaussian distribution.

In the second time slot, R amplifies and forwards the received signal to D with an amplification factor β. P _R is the transmit power of R. Therefore, β can be expressed as

Then, the signals received at D and E are

是D和E处的复杂加性高斯白噪声。由于干扰信号z_J源自D，而D对此有充分的了解。然后，D可以有效地去除该项，并且可以得到D处的接收信号为where n _D and

are complex additive white Gaussian noise at D and E. Since the interfering signal z _J originates from D, D has full knowledge of it. Then, D can effectively remove this term, and the received signal at D can be obtained as

Calculate the privacy rate of the transmission link from the source node to the destination node: For simplicity, the present invention redefines the channel-to-noise ratio:

In the first time slot, the signal-to-interference and noise ratio (SINR) of the eavesdropping link is:

The instantaneous SINRs of D and E are:

In order to obtain the best eavesdropping performance, the eavesdropper adopts the Maximum Ratio Combining (MRC) method. Then, the signal to interference and noise ratio (SINR) at the eavesdropping node E is

In the relay system based on physical layer security, the achievable privacy rate is as follows:

Step 3: Build an optimization model with the objective function when the privacy rate reaches the maximum, design the interference power, and the location constraints of the UAV:

where R _S is given in the second step, (x _u , y _u ) represents the horizontal position of the UAV in the considered area A. Although this problem seems simple, it is rather cumbersome to solve due to the non-convexity of the objective function and feasible region. In the following steps, the problem is decomposed into two levels of subproblems. The outer layer is the deployment optimization of the UAV position, and the inner layer is the optimization of the interference power distribution under the fixed UAV position.

Step 4: Under the condition that the position of the drone is fixed, traverse the destination node D and the coordinate position of the eavesdropper E, and use the binary search algorithm to optimize the interference power allocation scheme to maximize the privacy rate: place the drone on a fixed drone. In the case of , the optimization problem of interference power is as follows:

For this problem, there is a fundamental trade-off that needs to be handled carefully. If the jamming power is too small, the eavesdropper cannot be sufficiently disturbed. On the contrary, if the interference power is too large, it will not be able to effectively relay the source information due to the limited total power of the relay, thus also affecting the security of the communication system. Therefore, in order to meet the needs of both aspects, we consider optimizing the interference power.

To maximize the privacy rate of interference power, we first analyze the nature of the privacy rate. For the above optimization problem, the present invention uses the binary search algorithm to solve the obtained power, that is, the power distribution scheme of the current fixed UAV position, and the optimal interference power is

binary search algorithm

Step 1: Initialization: set the minimum and maximum values of _PD , given as P _min and P _max , where P _min =0; define a sufficiently small threshold ε;

R_S由第二步给出，如果

则P_max＝P^*，否则P_min＝P^*；Step 2: Order

R _S is given by the second step, if

Then P _max =P ^* , otherwise P _min =P ^* ;

Step 3: If |P _max -P _min |<ε, obtain the optimal interference power P ^* , and complete the power allocation strategy of the UAV relay communication system; otherwise, repeat step 2. In addition, by introducing a cost factor, the objective function in fractional form of the problem can be transformed into a subtraction at the cost of power, and the problem can be transformed into a convex optimization problem. The Dinkelbach algorithm is used to cyclically iterate the transformed problem to obtain The optimal solution of the original power optimization problem.

Step 5: Under the condition of optimal interference power, use the exhaustive search method to obtain the optimal position of the UAV, generate a data set, build and train the DNN model, apply it to the test set, and use the high computational efficiency of DNN to find the unmanned aerial vehicle. The best position of the man and the machine to maximize the privacy rate and achieve secure transmission.

On the basis of the fourth step, the optimal interference power is obtained, and the optimization problem of UAV position deployment is obtained as follows:

无人机最优位置的坐标(x_u,y_u)作为标签输出，可以表示为

其中i∈{1,2,···,N}。把Q＝[q₁,q₂,···,q_N]作为DNN的输入，q_i中的每一项对应一个输入神经元。取无人机的最优位置U＝[u₁,u₂,···,u_N]作为DNN模型的输出，u_i中的每一项对应一个输出神经元。最后，将经过良好训练的DNN模型用于测试集，快速有效的找到无人机的最优位置，求解无人机位置优化的问题，以实现最高的私密速率。此外，还可以在最优干扰功率的情况下，使用连续凸近似(SCA)的方法求解无人机位置部署问题，连续凸近似法是一种求解非凸优化问题的方法，它可以将原非凸表达式进行全局近似，通过将其非凸的目标函数和约束条件进行松弛和一阶泰勒展开，将非凸问题转化为凸优化问题，然后该过程在连续凸逼近算法框架下可求解出位置优化的最优解，找到无人机的最优位置。The interference power constraint in the optimization problem is obtained by solving the optimization problem in the fourth step, that is, under the condition of the optimal interference power, the optimization problem of the above-mentioned UAV position is solved. Since there are certain difficulties in solving the above problems, and there is currently no available solution with high computational efficiency, therefore, in this part, the present invention adopts a neural network model with multiple hidden layers and a large amount of training data to learn network features, and then solves the problem. The optimization problem of UAV position deployment, realizes a DNN-based UAV deployment scheme. Build a fully connected DNN model with one input layer, two hidden layers, and one output layer, learn its input/output relationship by training the DNN, apply it to the test set, and take advantage of the high computational efficiency of the DNN to find the best UAV. The best location to maximize the privacy rate and realize the secure transmission of the system. Consider a system located within a range of 200 meters by 200 meters. The fixed height h of the UAV is 150 meters, and the source node has a fixed position (0, 0, 0). First, the positions of S, D, and E are fixed, the horizontal position of the UAV is traversed, and the optimal jamming power P ^* is obtained by using the binary search algorithm. Under the condition of the optimal jamming power, the maximum privacy rate corresponds to the maximum privacy rate of the UAV. Excellent location. Then, by traversing the coordinate positions of D and E, the optimal position of the UAV corresponding to different D and E is found to construct the data set. Since the binary search algorithm in the fourth step can be implemented efficiently, data collection can be facilitated. In the input layer of the DNN model, the coordinates (x _d , y _d ) and (x _e , y _e ) of the target node and the eavesdropper are reshaped into a 4×1 data sample, which can be expressed as

The coordinates (x _u , y _u ) of the optimal position of the drone are output as labels, which can be expressed as

where i∈{1,2,...,N}. Take Q=[q ₁ , q ₂ , . . . , q _N ] as the input of DNN, and each item in q _i corresponds to an input neuron. Take the optimal position U=[u ₁ , u ₂ , . . . , u _N ] as the output of the DNN model, and each item in u _i corresponds to an output neuron. Finally, the well-trained DNN model is used for the test set to quickly and effectively find the optimal position of the drone, and solve the problem of the optimization of the drone position to achieve the highest privacy rate. In addition, in the case of optimal interference power, the continuous convex approximation (SCA) method can be used to solve the UAV position deployment problem. The continuous convex approximation method is a method for solving non-convex optimization problems. The convex expression is globally approximated, and the non-convex problem is transformed into a convex optimization problem by relaxing and first-order Taylor expansion of its non-convex objective function and constraints, and then the process can be solved under the framework of the continuous convex approximation algorithm. Optimize the optimal solution to find the optimal position of the UAV.

The technical effects of the present invention will be described in detail below in conjunction with experiments.

Figure 3 shows a schematic diagram of the relationship between the safety performance of the UAV relay network communication system and the location of the eavesdropper. As can be seen from Figure 3, the solution of the present invention is under different eavesdropper location conditions, because the interference conditions are met. , the privacy rate of the system changes with the change of location, first increases and then decreases with the change of location. In contrast, under the two schemes of direct transmission without relay and without interference, the privacy rate decreases when it is close to the eavesdropper and increases when it is far away from the eavesdropper. From the numerical point of view, the privacy rate of the proposed scheme is up to 1.7dB higher than the two schemes of direct transmission without relay and without interference, which strictly guarantees the secure transmission of the network.

Figure 4 shows a schematic diagram of the relationship between the safety performance of the UAV relay network communication system and the relay power. It can be seen from Figure 4 that, since the eavesdropper's position fixation condition is met, the privacy rate of the system under the scheme of the present invention is relatively different. This is an increase over the scheme without interference, and numerically, the privacy rate of the scheme without interference will be a maximum of 0.8 dB lower than the result of the scheme of the present invention. In the direct transmission scheme without relay, since there is no relay power, the privacy rate is 0.35dB, which remains unchanged. From the above results, it can be seen that the scheme proposed by the present invention has obvious advantages compared with the benchmark in terms of safety performance, and enhances the safe transmission of the existing UAV relay network.

It should be noted that the embodiments of the present invention may be implemented by hardware, software, or a combination of software and hardware. The hardware portion may be implemented using special purpose logic; the software portion may be stored in memory and executed by a suitable instruction execution system, such as a microprocessor or specially designed hardware. Those of ordinary skill in the art will appreciate that the apparatus and methods described above may be implemented using computer-executable instructions and/or embodied in processor control code, for example on a carrier medium such as a disk, CD or DVD-ROM, such as a read-only memory Such code is provided on a programmable memory (firmware) or a data carrier such as an optical or electronic signal carrier. The device and its modules of the present invention can be implemented by hardware circuits such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., It can also be implemented by software executed by various types of processors, or by a combination of the above-mentioned hardware circuits and software, such as firmware.

The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited to this. Any person skilled in the art is within the technical scope disclosed by the present invention, and all within the spirit and principle of the present invention Any modifications, equivalent replacements and improvements made within the scope of the present invention should be included within the protection scope of the present invention.

Claims (6)

1. a security transmission enhancement method oriented to the UAV relay network, is characterized in that, the described security transmission enhancement method oriented to the UAV relay network comprises: Establish a UAV relay communication system, the goal of which is to jointly optimize the jamming power and the UAV relay position to maximize the privacy rate of the system; Establish the channel model of the ground source node-to-destination node communication system with the UAV as the relay, and in the two-slot-half-duplex mode, according to the UAV received signal and the received signal of the destination node, calculate the source node to The privacy rate of the destination node transmission link; Build an optimization model with the objective function when the privacy rate reaches the maximum, and design the interference power and UAV position constraints; Under the condition that the position of the UAV is fixed, traverse the destination node D and the coordinate position of the eavesdropper E, and use the binary search algorithm to optimize the interference power allocation scheme to maximize the privacy rate; Under the condition of optimal interference power, use the exhaustive search method to obtain the optimal position of the UAV, generate a data set, build and train the DNN model, apply it to the test set, and use the high computational efficiency of DNN to find the optimal position of the UAV. The best location to maximize the privacy rate and achieve secure transmission; The security transmission enhancement method for the UAV relay network establishes the channel model of the ground source node to destination node communication system with the UAV as the relay, and in the two-time-slot-half-duplex mode, according to the unmanned aerial vehicle. Calculate the privacy rate of the transmission link from the source node to the destination node; the channel model of the communication system from the source node to the destination node on the ground:

where c is the speed of light, α _G is the path loss index of the ground communication link, f _c is the carrier frequency, d _eg is the distance between the eavesdropper and the ground user, and σ _G is the shadow fading variable of the channel; The channels related to the UAV include the line-of-sight LoS group and the non-line-of-sight NLoS group, and the probability of having a LOS connection between the ground user and the UAV is:

where A and B are constants depending on the environment, (x _u , y _u ) is the position of the drone in the horizontal dimension, h is the height of the drone, (x _g , y _g ) is the position of the ground user, The probability of NLoS is P _NLoS =1-P _LoS , and the path loss models of LoS and NLoS links are:

where d is the transmission distance,

α _L and α _N are the path loss exponents for LoS and NLoS channels, η _LoS and η _NLoS are the average additional losses for LoS and NLoS, respectively; the probabilistic average path loss is averaged under LoS and NLoS conditions: h _ij =P _LoS L _LoS +P _NLoS L _NLoS , i∈{R,D}, j∈{R,E,D}; In the first time slot, S transmits its signal to the drone relay R, which is also eavesdropped by the eavesdropper E; at the same time, D emits artificial noise, which confuses the eavesdropper; in the second time slot, S is silent state, the drone relay amplifies the received signal and sends it to D, D is also eavesdropped by eavesdroppers, denoted by z _S and z _J for confidential signals and cooperative jamming signals from S and D, both z _S and z _J are normalized powers, ie |z _S | ² =1 and |z _J | ² =1, where |·| represents the absolute value, the signals received at the first time slot, UAV and E are:

where P _S and P _D are the transmit powers from S and D, respectively, n _R and

is the complex additive white Gaussian noise AWGN at R and D, following zero mean and variance

The complex Gaussian distribution of ; In the second time slot, R amplifies and forwards the received signal to D with an amplification factor β, where P _R is the transmit power of R, and β is expressed as:

Then, the signals received at D and E are:

where n _D and

is the complex additive white Gaussian noise at D and E, D effectively removes the interfering signal term, and the received signal at D is obtained as:

Calculate the privacy rate of the transmission link from the source node to the destination node, and redefine the channel-to-noise ratio:

In the first time slot, the SINR of the eavesdropping link is:

The instantaneous SINRs of D and E are:

The eavesdropper adopts the maximum proportion combined MRC method, and the signal-to-interference-noise ratio SINR at the eavesdropping node E is:

In the relay system based on physical layer security, the private rate is obtained as follows:

The security transmission enhancement method for the UAV relay network constructs an optimization model that takes the privacy rate reaching the maximum as the objective function, and designs the interference power and UAV position constraints:

(x _u , y _u )∈Α; where (x _u , y _u ) represents the horizontal position of the UAV in the considered area A; The security transmission enhancement method for the UAV relay network determines that under the condition that the UAV position is fixed, traverse the coordinate position of the destination node D and the eavesdropper E, and use the binary search algorithm to optimize the interference power distribution scheme to maximize the interference power distribution scheme. Optimizing the privacy rate: In the case of fixed UAV placement, the optimization problem of jamming power is as follows:

The security transmission enhancement method for the UAV relay network uses the binary search algorithm to obtain the power obtained by solving the power distribution scheme of the current fixed UAV position, and the optimal interference power is

Step 1: Initialization: set the minimum and maximum values of _PD , given as P _min and P _max , where P _min =0; define a sufficiently small threshold ε; Step 2: Order

R _S is given by the second step, if

Then P _max =P ^* , otherwise P _min =P ^* ; Step 3: If |P _max -P _min |<ε, obtain the optimal interference power P ^* , and complete the power allocation strategy of the UAV relay communication system; otherwise, repeat step 2. 2. The safety transmission enhancement method for UAV relay network as claimed in claim 1, is characterized in that, the described safety transmission enhancement method for UAV relay network establishes a UAV relay communication system, which The goal is to jointly optimize the interference power and UAV relay position to maximize the privacy rate of the system; consider a UAV relay communication system composed of source node S, destination node D, UAV relay R and eavesdropper E, S sends a signal to the UAV relay, R amplifies and forwards the signal to D. 3. the safe transmission enhancement method for UAV relay network as claimed in claim 1 is characterized in that, under the condition of optimal interference power, the described safe transmission enhancement method for UAV relay network uses The exhaustive search method obtains the optimal position of the drone, generates a data set, constructs and trains the DNN model, applies it to the test set, and uses the high computational efficiency of DNN to find the best position of the drone to maximize the privacy rate and achieve secure transmission; To obtain the optimal jamming power, the optimization problem of UAV position deployment is obtained as follows:

st(x _u , y _u )∈Α

4. The safety transmission enhancement method for UAV relay network as claimed in claim 3 is characterized in that, the safety transmission enhancement method for UAV relay network adopts a plurality of hidden layers and a large amount of training data The neural network model learns network features to solve the optimization problem of UAV position deployment, and realizes the UAV deployment scheme based on DNN; constructs a fully connected DNN model with one input layer, two hidden layers and one output layer , learn its input/output relationship by training DNN, apply it to the test set, use the high computational efficiency of DNN to find the best position of the drone, maximize the privacy rate, and realize the safe transmission of the system, consider a location at 200 meters × For a system within a range of 200 meters, the fixed height h of the UAV is 150 meters, and the source node has a fixed position (0, 0, 0). First, the positions of S, D, and E are fixed, and the horizontal position of the UAV is traversed. The optimal interference power P ^* is obtained by using the binary search algorithm. Under the condition of the optimal interference power, the maximum privacy rate corresponds to the optimal position of the UAV; The optimal position of the UAV to construct a data set; In the input layer of the DNN model, the coordinates (x _d , y _d ) and (x _e , y _e ) of the target node and the eavesdropper are reshaped into a 4×1 data sample, denoted as

The coordinates (x _u , y _u ) of the optimal position of the UAV are output as labels, which are expressed as

where i∈{1,2,...,N}, take Q=[q ₁ ,q ₂ ,...,q _N ] as the input of DNN, each item in q _i corresponds to an input neuron, take the UAV The optimal position of U=[u ₁ , u ₂ ,...,u _N ] is used as the output of the DNN model, each item in u _i corresponds to an output neuron, and finally, the well-trained DNN model is used for the test set , quickly and effectively find the optimal position of the UAV, and solve the problem of UAV position optimization. 5. A computer device, characterized in that the computer device comprises a memory and a processor, the memory stores a computer program, and when the computer program is executed by the processor, the processor is made to execute claim 1 The security transmission enhancement method for a UAV relay network according to any one of ~4. 6 . A safety transmission enhancement system for a UAV relay network implementing the method for enhancing safety transmission for a UAV relay network according to any one of claims 1 to 4 , wherein the unmanned aircraft-oriented The security transmission enhancement system of the machine relay network includes: The UAV relay communication system establishment module is used to establish the UAV relay communication system, and its goal is to jointly optimize the interference power and the UAV relay position to maximize the privacy rate of the system; The channel model and slot establishment module are used to establish the channel model of the ground source node-to-destination node communication system with the UAV as the relay, and in the two-slot-half-duplex mode, according to the UAV received signal and the purpose The received signal of the node calculates the privacy rate of the transmission link from the source node to the destination node; The objective function building module is used to construct an optimization model that takes the privacy rate reaching the maximum as the objective function, design interference power, and UAV position constraints; The privacy rate maximization module is used to traverse the coordinate position of the destination node D and the eavesdropper E under the condition that the position of the drone is fixed, and use the binary search algorithm to optimize the interference power allocation scheme to maximize the privacy rate; The safe transmission module is used to obtain the optimal position of the UAV using the exhaustive search method under the condition of optimal interference power, generate a data set, build and train a DNN model, apply it to the test set, and utilize the high computational efficiency of DNN, Find the best position of the drone to maximize the privacy rate and achieve safe transmission; Establish the channel model of the ground source node-to-destination node communication system with the UAV as the relay, and in the two-slot-half-duplex mode, according to the UAV received signal and the received signal of the destination node, calculate the source node to The privacy rate of the destination node transmission link; the channel model of the ground source node to destination node communication system:

where c is the speed of light, α _G is the path loss index of the ground communication link, f _c is the carrier frequency, d _eg is the distance between the eavesdropper and the ground user, and σ _G is the shadow fading variable of the channel; The channels related to the UAV include the line-of-sight LoS group and the non-line-of-sight NLoS group, and the probability of having a LOS connection between the ground user and the UAV is:

where A and B are constants depending on the environment, (x _u , y _u ) is the position of the drone in the horizontal dimension, h is the height of the drone, (x _g , y _g ) is the position of the ground user, The probability of NLoS is P _NLoS =1-P _LoS , and the path loss models of LoS and NLoS links are:

where d is the transmission distance,

α _L and α _N are the path loss exponents for LoS and NLoS channels, η _LoS and η _NLoS are the average additional losses for LoS and NLoS, respectively; the probabilistic average path loss is averaged under LoS and NLoS conditions: h _ij =P _LoS L _LoS +P _NLoS L _NLoS , i∈{R,D}, j∈{R,E,D}; In the first time slot, S transmits its signal to the drone relay R, which is also eavesdropped by the eavesdropper E; at the same time, D emits artificial noise, which confuses the eavesdropper; in the second time slot, S is silent state, the drone relay amplifies the received signal and sends it to D, D is also eavesdropped by eavesdroppers, denoted by z _S and z _J for confidential signals and cooperative jamming signals from S and D, both z _S and z _J are normalized powers, ie |z _S | ² =1 and |z _J | ² =1, where |·| represents the absolute value, the signals received at the first time slot, UAV and E are:

where P _S and P _D are the transmit powers from S and D, respectively, n _R and

is the complex additive white Gaussian noise AWGN at R and D, following zero mean and variance

The complex Gaussian distribution of ; In the second time slot, R amplifies and forwards the received signal to D with an amplification factor β, where P _R is the transmit power of R, and β is expressed as:

Then, the signals received at D and E are:

where n _D and

is the complex additive white Gaussian noise at D and E, D effectively removes the term, and the received signal at D is obtained as:

Calculate the privacy rate of the transmission link from the source node to the destination node, and redefine the channel-to-noise ratio:

In the first time slot, the SINR of the eavesdropping link is:

The instantaneous SINRs of D and E are:

The eavesdropper adopts the maximum proportion combined MRC method, and the signal-to-interference-noise ratio SINR at the eavesdropping node E is:

In the relay system based on physical layer security, the private rate is obtained as follows:

Construct an optimization model with the objective function when the privacy rate reaches the maximum, and design the interference power and UAV position constraints:

(x _u , y _u )∈Α; where (x _u , y _u ) represents the horizontal position of the UAV in the considered area A; Determine the location of the drone, traverse the destination node D, the coordinate position of the eavesdropper E, and use the binary search algorithm to optimize the interference power allocation scheme to maximize the privacy rate: in the case of fixed drone placement, The optimization problem of interference power is as follows:

The power obtained by solving the binary search algorithm is the power allocation scheme of the current fixed UAV position, and the optimal interference power is

Step 1: Initialization: set the minimum and maximum values of _PD , given as P _min and P _max , where P _min =0; define a sufficiently small threshold ε; Step 2: Order

R _S is given by the second step, if

Then P _max =P ^* , otherwise P _min =P ^* ; Step 3: If |P _max -P _min |<ε, obtain the optimal interference power P ^* , and complete the power allocation strategy of the UAV relay communication system; otherwise, repeat step 2.

Images