A cooperative jamming method based on UAV relay system

Abstract

The present application relates to a cooperative interference method based on a drone relay system, which includes establishing a cooperative interference model based on a drone relay system, wherein the cooperative interference model includes a source node S, a drone relay node R, a drone jammer node J, a legitimate destination node D, and an eavesdropper E, wherein the drone jammer node J is used to transmit an interference signal to counter the eavesdropper; based on the cooperative interference model, constructing an objective function for maximizing the average confidentiality rate; optimizing the objective function for maximizing the average confidentiality rate, thereby enhancing the confidentiality performance of a drone cooperative interference-assisted relay network system. The present application has the effect of reducing the excessive dependence of the confidentiality rate on the channel quality and improving the physical layer security of the drone relay system.

Description

Technical Field

[0001] The present application relates to the field of wireless communication technology, and in particular to a collaborative interference method based on an unmanned aerial vehicle relay system. Background Art

[0002] The broadcast characteristics of wireless communications pose a great threat to secure communications, so how to avoid information leakage is an important issue in wireless network design.

[0003] In practical applications, drones are flexibly deployed to assist transmission between source nodes and target nodes in the signal transmission channel. Drones as relay nodes can increase data rates. However, as a result, transceivers in drone-assisted communication networks are susceptible to eavesdropping. The existing channel confidentiality rate is overly dependent on channel quality, and the physical layer security of drone relay systems needs to be improved. Summary of the invention

[0004] The purpose of this application is to provide a collaborative interference method based on a drone relay system, which has the characteristics of reducing the excessive dependence of the confidentiality rate on the channel quality and improving the physical layer security of the drone relay system.

[0005] The present application provides a cooperative jamming method based on a drone relay system, which adopts the following technical solution:

[0006] A collaborative interference model based on a drone relay system is established, wherein the collaborative interference model includes a source node S, a drone relay node R, a drone jammer node J, a legitimate destination node D, and an eavesdropper E. The drone jammer node J is used to transmit an interference signal to counter the eavesdropper.

[0007] Based on the collaborative interference model, construct an objective function for maximizing the average confidentiality rate;

[0008] The objective function of maximizing the average confidentiality rate is optimized to enhance the confidentiality performance of the UAV collaborative interference assisted relay network system.

[0009] By adopting the above technical solution, a collaborative interference strategy in which drones act as jammers is used in the collaborative interference model to transmit interference signals to counter eavesdroppers, thereby improving the confidentiality performance of the system. At the same time, based on the use of drone jammers, an objective function for maximizing the average confidentiality rate is constructed and optimized, thereby achieving the enhancement of the confidentiality performance of the drone collaborative interference-assisted relay network system, reducing the excessive dependence of the channel positive confidentiality rate on the channel quality, and improving the physical layer security of the drone relay system.

[0010] Optionally, when there is hardware damage to the drone relay, the drone relay is adjusted to move away from the eavesdropper and hover near the source node and the destination node, and at the same time, the drone jammer is adjusted to hover near the eavesdropper.

[0011] By adopting the above technical solution, the UAV relay is adjusted to move away from the eavesdropper and hover close to the source node and the destination node, which can improve the signal-to-noise ratio of the destination node, and the UAV jammer is adjusted to hover near the eavesdropper to further improve the average confidentiality rate.

[0012] Optionally, when the drone relay performs a communication task within T time, when the hardware damage degree of the drone relay increases, the drone relay is adjusted to move in a direction farther away from the eavesdropper to reduce the time the drone relay hovers around the source node and the destination node.

[0013] By adopting the above technical solution, when the hardware damage level is higher, the drone relay moves in a direction farther away from the eavesdropper. The drone relay can be adjusted to reduce the hovering time at the source node and the destination node, thereby further ensuring the completion of the task while maximizing the system confidentiality performance.

[0014] Optionally, the objective function of maximizing the average confidentiality rate is constructed in the following manner:

[0015]

[0016] in, To maximize the average confidentiality rate, R D [n] is the achievable data rate of the destination node in each time slot, R E [n] is the achievable data rate of the eavesdropper in each time slot; N is the number of time slots, i2∈{I1+1,I1+2,…,I}, n∈{1,2,…,N}, i1∈{1,2,…,I1}; I1 is the number of no-fly zones relayed by the drone, and I represents the maximum number of no-fly zones; (x R [n],y R [n],H) and (x J [n],y J [n],H) are the coordinates of the drone relay and drone jammer in each time slot; (x R0 ,y R0 ,H) and (x RF ,y RF ,H) are the flight start and end coordinates of the drone relay; (x J0 ,y J0 ,H) and (x JF ,y JF ,H) are the starting and ending coordinates of the UAV jammer; (x NOi ,yNOi ,0),i∈{i1,i2} is the center of the no-fly zone, r NOi is the radius; V is the moving distance of the UAV relay and UAV jammer, which cannot exceed the maximum movable distance; P S [n], P R [n] and P J [n] represents the transmission power of the source node, UAV relay and UAV jammer, respectively, P Smax , P Rmax and P Jmax They represent the corresponding peak power, and They represent the corresponding mean power respectively; d is the minimum distance between two UAVs to avoid collision.

[0017] By adopting the above technical solution, the transmission power and drone trajectory are jointly optimized to maximize the average confidentiality rate of the system; and it can cope with more practical flight environments. In some private or public places where drone flights are prohibited, the constraints of the no-fly zone are introduced into the construction of the objective function of maximizing the average confidentiality rate, further enhancing the applicability of the drone collaborative interference assisted relay network system. This application also introduces a minimum spacing constraint between drones to avoid collisions between drones.

[0018] Optionally, the signal y received by the drone relay node R [n] is:

[0019]

[0020] Among them, x S [n]~CN(0,1) indicates the source node transmission signal, is the distortion noise caused by hardware damage, n R ~CN(0,σ 2 ) represents the additive noise of the drone relay.

[0021] By adopting the above technical solution, we can deal with situations where nodes are subject to varying degrees of hardware damage and transmission signal distortion. The signal distortion caused by hardware damage can be introduced into the construction of the objective function of maximizing the average confidentiality rate, further enhancing the applicability of the UAV collaborative interference assisted relay network system.

[0022] Optionally, the objective function of maximizing the average confidentiality rate is optimized in the following manner:

[0023] The optimization problem of the objective function is decomposed into four sub-problems, namely, the source node transmission power and the UAV relay node transmission power optimization sub-problem, the UAV jammer node transmission power optimization sub-problem, the UAV relay trajectory optimization sub-problem and the UAV jammer trajectory optimization sub-problem;

[0024] For the four sub-problems mentioned above, the corresponding non-convex objective functions in the corresponding sub-problems are converted into convex objective functions, and the convex optimization method is used to solve them.

[0025] By adopting the above technical solution, the original optimization problem is decomposed into four sub-problems, which effectively solves the coupling problem of trajectory optimization variables and power optimization variables. At the same time, the non-convex objective function is converted into a convex objective function and solved using the convex optimization method, which can reduce the complexity.

[0026] Optionally, the sub-problems of optimizing the transmission power of the source node and the transmission power of the drone relay node are optimized in the following ways:

[0027] Fix the transmission power of the UAV jammer node, the UAV relay trajectory, and the UAV jammer trajectory, and obtain a new objective function based on the transmission power of the source node and the transmission power of the UAV relay node:

[0028]

[0029] Introducing auxiliary variable R b1 = {R b1 [1],R b1 [2],...,R b1 [N]}, the non-convex constraints in the constraints are converted into convex constraints, and the non-convex objective function is converted into a convex function. The approximate expression of the transmission power optimization sub-problem of the source node and the UAV relay node is:

[0030]

[0031] Then the convex optimization method is used to solve it.

[0032] By adopting the above technical solution, the introduction of auxiliary variables is more conducive to analysis. At the same time, the non-convex constraints in the constraints are converted into convex constraints, and the non-convex objective function is converted into a convex function, which can reduce complexity and improve operating efficiency.

[0033] Optionally, the sub-problem of optimizing the transmission power of the drone jammer node is optimized in the following ways:

[0034] By fixing the transmission power of the source node and the UAV relay node, the UAV relay trajectory, and the UAV jammer trajectory, a new objective function based on the transmission power of the UAV jammer node is obtained:

[0035]

[0036] Then the convex optimization method is used to solve it.

[0037] By adopting the above technical solution, the convex optimization algorithm can be directly used for solving, which reduces complexity and improves operating efficiency.

[0038] Optionally, the drone relay trajectory optimization sub-problem can be optimized in the following ways:

[0039] By fixing the transmission power of the source node and the UAV relay node, the transmission power of the UAV jammer node, and the trajectory of the UAV jammer, a new objective function based on the trajectory of the UAV relay node is obtained:

[0040]

[0041] Introducing auxiliary variable R b2 = {R b2 [1],R b2 [2],...,R b2 [N]} and D e ={D e [1],D e [2],...,D e [N]}, the non-convex constraints in the constraints are converted into convex constraints, and the non-convex objective function is converted into a convex function. The approximate expression of the drone relay trajectory optimization subproblem is:

[0042]

[0043] Then the convex optimization method is used to solve it.

[0044] By adopting the above technical solution, the introduction of auxiliary variables is more conducive to analysis. At the same time, the non-convex constraints in the constraints are converted into convex constraints, and the non-convex objective function is converted into a convex function, which can reduce complexity and improve operating efficiency.

[0045] Optionally, the trajectory optimization sub-problem of the UAV jammer can be optimized in the following ways:

[0046] By fixing the transmission power of the source node and the UAV relay node, the transmission power of the UAV jammer node, and the UAV relay trajectory, a new objective function based on the trajectory of the UAV jammer is obtained:

[0047]

[0048] Introducing auxiliary conditions The non-convex constraints in the constraints are converted into convex constraints, and the non-convex objective function is converted into a convex function. The approximate expression of the UAV jammer trajectory optimization subproblem is:

[0049]

[0050] Then the convex optimization method is used to solve it.

[0051] By adopting the above technical solution and introducing auxiliary conditions, non-convex constraints in the constraints can be converted into convex constraints, and non-convex objective functions can be converted into convex functions, which can reduce complexity and improve operating efficiency.

[0052] In summary, the present application includes at least one of the following beneficial technical effects:

[0053] (1) This application is applicable to the UAV relay system under hardware damage conditions. Other UAVs are introduced to act as jammers to solve the problem that the system's positive confidentiality rate is overly dependent on the channel quality. At the same time, considering the transmission power, motion trajectory and information causality constraints, a resource allocation algorithm is designed to maximize the system's average confidentiality rate.

[0054] (2) This application proposes a two-layer solution framework, which solves the variable coupling problem in the outer layer and introduces slack variables in the inner layer and uses continuous convex approximation to obtain approximate solutions to sub-problems, which has low complexity;

[0055] (3) This application takes into account a more practical flight environment and prohibits drone flights in some private or public places. At the same time, it introduces a minimum spacing constraint between drones to avoid collisions between drones. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 This is a flow chart of a collaborative interference method based on a drone relay system proposed in this application.

[0057] Figure 2 It is a flowchart of one specific implementation method of step S3 in this application.

[0058] Figure 3 This is a schematic diagram of a collaborative interference model based on a drone relay system proposed in this application.

[0059] Figure 4 This is a comparison chart of the average confidentiality rate of the scheme proposed in this application and three benchmark schemes versus the number of iterations.

[0060] Figure 5 It is a comparison chart of the trajectory changes of the drone relay and the drone jammer in the solution proposed in this application and the three benchmark solutions.

[0061] Figure 6 This is a comparison chart of the drone trajectories under different no-fly zone constraints of the solution proposed in this application.

[0062] Figure 7 This is a comparison chart of the average confidentiality rate versus average transmission power curve under different hardware damage levels of the solution proposed in this application.

[0063] Figure 8This is a comparison chart of the transmission power change curve in each time slot under different hardware damage levels of the solution proposed in this application.

[0064] Fig. 9 This is a comparison chart of the trajectory change curves of the drone in each time slot under different hardware damage levels of the solution proposed in this application.

[0065] Fig.10 This is a comparison chart of the average confidentiality rate of the solution proposed in this application and three benchmark solutions versus average transmission power. DETAILED DESCRIPTION

[0066] The following is combined with Figure 1-10 This application is described in further detail.

[0067] In order to make the purpose, technical solutions and advantages of this application more clear, the following Figure 1-10 It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0068] An embodiment of the present application provides a collaborative interference method based on a drone relay system.

[0069] Reference Figure 1 , an embodiment of the present application discloses a cooperative interference method based on a drone relay system, comprising:

[0070] S1. Establish a collaborative interference model based on the drone relay system. The collaborative interference model includes the source node S, the drone relay node R, the drone jammer node J, the legitimate destination node D and the eavesdropper E. The drone jammer node J is used to transmit interference signals to counter the eavesdropper.

[0071] Specifically, refer to Figure 3 ,The cooperative jamming model of the set UAV relay system includes the source node S, UAV relay node R, UAV jammer node J, the legitimate destination node D, and the eavesdropper E. All nodes are equipped with an antenna for signal transmission. The source node transmits the confidential signal to the destination node through the UAV relay node. At the same time, UAVs are introduced to transmit jamming signals to counter the eavesdropper. The UAV jammer refers to the UAV that acts as a jammer, which improves the confidentiality performance of the system.

[0072] Based on the above nodes, the steps to establish a cooperative interference model based on the drone relay system are as follows:

[0073] First, establish a three-dimensional Cartesian coordinate system to facilitate analysis. The source node coordinates are (x S ,y S ,0), the coordinates of the destination node are (x D ,yD ,0), the eavesdropper’s coordinates are (x E ,y E ,0), the coordinates of the drone relay node are (x R ,y R ,H), the coordinates of the UAV jammer node are (x J ,y J ,H); the flight period of the drone relay and the drone jammer is T, and the flight altitude is H. The flight period of the drone is discretized into N time slots. The position of the drone in each time slot can be regarded as fixed. The coordinates of the drone relay and the drone jammer in each time slot are expressed as (x R [n],y R [n],H) and (x J [n],y J [n],H). During the flight, the starting point and end point of the drone relay are represented as (x R0 ,y R0 ,H) and (x RF ,y RF ,H), the starting point and end point of the UAV jammer are represented as (x J0 ,y J0 ,H) and (x JF ,y JF ,H).

[0074] S2. Based on the collaborative interference model, construct an objective function to maximize the average confidentiality rate.

[0075] Specifically, the steps to construct the objective function for maximizing the average confidentiality rate are as follows:

[0076] First, in view of the fact that the drone relay and the drone jammer pass through some no-fly zones during the flight, in this embodiment, a cylindrical no-fly zone is set, wherein the center of the no-fly zone is (x NOi ,y NOi ,0),i∈{i1,i2}, the radius is r NOi , height H NO , and H NO >>H.

[0077] Secondly, the UAV relay and the UAV jammer fly from the starting point to the end point, and always follow the maximum flight speed constraint during the flight. Assume that the maximum flight speed of the UAV relay and the UAV jammer is v max , within the UAV flight time slot T / N, the moving distance of the UAV relay and the UAV jammer cannot exceed the maximum movable distance V = (v max T) / N, so the following constraints are added, namely:

[0078] (x R[1],y R [1],H)=(x R0 ,y R0 ,H) (1)

[0079] (x R [N],y R [N],H)=(x RF ,y RF ,H) (2)

[0080] (x R [n+1]-x R [n]) 2 +(y R [n+1]-y R [n]) 2 ≤V 2 ,n∈{1,2,...,N} (3)

[0081] (x J [1],y J [1],H)=(x J0 ,y J0 ,H) (4)

[0082] (x J [N],y J [N],H)=(x JF ,y JF ,H) (5)

[0083] (x J [n+1]-x J [n]) 2 +(y J [n+1]-y J [n]) 2 ≤V 2 ,n∈{1,2,...,N}(6)

[0084] In addition, drones will be affected by no-fly zones during missions, and drones can only avoid no-fly zones and cannot fly over them. Therefore, the trajectories of drone relays and drone jammers are subject to no-fly zone constraints, namely:

[0085]

[0086] Where I1 is the number of no-fly zones of the drone relay, and I represents the maximum number of no-fly zones. Formula (8) allows the drone jammer to be in a non-ideal state, that is, the eavesdropper is located in the no-fly zone. Even if the drone jammer cannot reach the ideal position to fight the eavesdropper, it can still fly along the edge of the no-fly zone to maximize the fight against the eavesdropper, further improving the security of system transmission.

[0087] Furthermore, during the flight, the drone relay and the drone jammer fly at the same altitude, and different drones have different no-fly zones. Then, in order to avoid collision between the two drones during flight, the flight trajectories of the drone relay and the drone jammer also add the following constraints:

[0088] (x J [n]-x R [n]) 2 +(y J [n]-y R [n]) 2 ≥d 2 ,n∈{1,2,...,N} (8)

[0089] d is the minimum distance between two drones to avoid collision.

[0090] In actual applications, there is no possibility of a direct link between the source node and the destination node. The source node transmits confidential signals to the destination node with the help of the drone relay node, and the drone jammer transmits jamming signals to counter eavesdroppers. Due to the node energy limitation in actual applications, the node transmission power is constrained by the average power and peak power. Then, the following constraints are added to the source node transmission power, the drone relay node transmission power, and the drone jammer node transmission power:

[0091] 0≤P S [n]≤P Smax (9)

[0092] 0≤P R [n]≤P Rmax (10)

[0093] 0≤P J [n]≤P Jmax (11)

[0094]

[0095]

[0096] Among them, P S [n], P R [n] and P J [n] represents the transmission power of the source node, UAV relay and UAV jammer, respectively, P Smax , P Rmax and P Jmax They represent the corresponding peak power, and Represent the corresponding mean power respectively.

[0097] If the system is a delay-tolerant system, the data decoded by the drone relay will be forwarded in the remaining time slot, so the following information causality constraints are added to the transmission process, namely:

[0098]

[0099] Among them, R D [i] and R R [i] represents the achievable data rates of the destination node and relay node respectively.

[0100] The channel gains between the drone relay and the source node, destination node and eavesdropper are:

[0101]

[0102] From formula (17), it can be seen that the channel gain between the set UAV relay and the source node is inversely proportional to the distance. When the UAV relay is close to the source node, there is a higher channel gain between the source node and the UAV relay;

[0103] It can be seen from formula (18) that the channel gain between the set UAV relay and the destination node is inversely proportional to the distance. When the UAV relay is close to the destination node, there is a higher channel gain between the UAV relay and the destination node. When the UAV relay is close to the source node, the distance between the UAV relay and the destination node increases. When other conditions remain unchanged, the destination node can achieve a lower throughput.

[0104] The channel gain between the UAV jammer and the eavesdropper is:

[0105]

[0106] Among them, g j [n]=|h j [n]| 2 ,j∈{SR,RD,RE,JE}, β0 represents the reference channel power when the reference distance is 1 meter, d j [n] is the distance from the drone relay to the source node, the drone relay to the destination node, the drone relay to the eavesdropper, and the drone jammer to the eavesdropper. From formula (20), it can be seen that the set channel gain between the drone jammer and the eavesdropper is inversely proportional to the distance. When the drone jammer is close to the eavesdropper, the drone jammer and the eavesdropper have a higher channel gain; and when the drone jammer is far away from the eavesdropper, the drone jammer can increase the transmission power to resist the eavesdropper to ensure the confidentiality of the system.

[0107] Furthermore, when there is hardware damage, the signal received by the drone relay node is:

[0108]

[0109] Among them, x S [n]~CN(0,1) indicates the source node transmission signal, is the distortion noise caused by hardware damage, n R ~CN(0,σ 2 ) represents the additive noise of the drone relay.

[0110] According to formula (21), the signal-to-noise ratio of the drone relay node in each time slot can be obtained as:

[0111]

[0112] Where β = β0 / σ 2 , Then the throughput from the source node to the drone relay in each time slot is obtained, that is:

[0113] R R [n] = log2(1 + γ R [n]) (22)

[0114] The drone relay node forwards the signal to the legitimate destination node using the DF protocol, while the drone jammer node sends the jamming signal to the eavesdropper. When there is hardware damage, the signals received by the destination node and the eavesdropper are:

[0115]

[0116] Among them, x R [n]~CN(0,1) represents the transmission signal of the drone relay node, x J [n]~CN(0,1) represents the transmission signal of the UAV jammer node, represents the distortion noise caused by hardware damage, n E ~CN(0,σ 2 ) represents the additive noise of the eavesdropper.

[0117] From equations (24) and (25), we can see that the signal-to-noise ratios of the destination node and the eavesdropper in each time slot are:

[0118]

[0119] in, It can be seen from formula (26) and formula (27) that the transmission power of the set drone relay is proportional to the signal-to-noise ratio of the destination node, and the transmission power of the set drone jammer is inversely proportional to the signal-to-noise ratio of the eavesdropper. The increase in transmission power makes the legitimate destination node receive a higher signal-to-noise ratio, and the increase in the transmission power of the drone jammer further suppresses the eavesdropper's eavesdropping rate.

[0120] It can be seen from formula (19) and formula (27) that when the drone relay is far away from the eavesdropper, the distance between the drone relay and the eavesdropper increases, the channel gain between the two decreases, the signal-to-noise ratio received by the eavesdropper is proportional to the channel gain, and the signal-to-noise ratio received by the eavesdropper decreases; and for the drone jammer, it can be seen from formula (20) and formula (27) that when the drone jammer is close to the eavesdropper, the channel gain between the two increases, and the signal-to-noise ratio received by the eavesdropper is inversely proportional to the channel gain, which will further reduce the signal-to-noise ratio received by the eavesdropper; it can be seen from formula (18) and formula (26) that when the drone relay is close to the destination node, the channel gain between the two increases, the signal-to-noise ratio of the destination node is proportional to the channel gain, and the signal-to-noise ratio received by the destination node increases; further, when the drone relay is close to the source node, the distance between the drone relay and the source node is small, the channel gain between the two increases, and the signal-to-noise ratio of the signal received by the relay node increases.

[0121] In summary, it can be seen from formula (26) and formula (27) that the presence of hardware damage will lead to a decrease in the received signal-to-noise ratio. The present application can improve the signal-to-noise ratio of the destination node by adjusting the drone relay to move away from the eavesdropper and hover close to the source node and the destination node, and adjust the drone jammer to hover near the eavesdropper to further improve the average confidentiality rate.

[0122] According to formulas (26) and (27), the achievable data rate of the destination node and the eavesdropper in each time slot is obtained, that is:

[0123] R D [n] = log2(1 + γ D [n]) (27)

[0124] R E [n] = log2(1 + γ E [n]) (28)

[0125] In this embodiment, the confidentiality rate of the system is taken as the optimization target. According to the confidentiality rate calculation formula, the confidentiality rate that can be achieved in each time slot is:

[0126] R[n]=[R D [n]-R E [n]] + (29)

[0127] To achieve a positive confidentiality rate, the signal-to-noise ratio of the legitimate destination node must be greater than the signal-to-noise ratio of the eavesdropper. The confidentiality rate that can be achieved in each time slot is equivalent to:

[0128] R[n]=R D [n]-R E [n] (30)

[0129] Then we can get the average confidentiality rate that the system can achieve in the entire transmission time slot:

[0130]

[0131] It can be seen from formula (32) that in order to ensure the confidentiality rate of the system, when the degree of hardware damage is higher, the drone relay can be adjusted to move in a direction farther away from the eavesdropper to ensure the confidentiality rate of the system. Furthermore, in this embodiment, the drone is set to perform the communication task within T time due to factors such as the drone battery capacity. When the degree of hardware damage is higher, the drone relay needs to move in a direction farther away from the eavesdropper. The drone relay can be adjusted to reduce the hovering time at the source node and the destination node, thereby further ensuring the completion of the task while maximizing the confidentiality performance of the system.

[0132] In this embodiment, the source node transmission power, the drone relay transmission power, the drone jammer transmission power, the drone relay trajectory, and the drone jammer trajectory are jointly optimized to maximize the system average confidentiality rate, which can maximize the system confidentiality performance within limited resources. Specifically, the auxiliary variable P is introduced S , P R , P J 、(x R ,y R ) and (x J ,y J ),Right now:

[0133] P S = {P S [1],P S [2],...,P S [N]} (32)

[0134] P R = {P R [1],P R [2],...,P R [N]} (33)

[0135] P J = {P J [1],P J [2],...,P J [N]} (34)

[0136] (x R ,y R )={(x R [1],y R [1]),(x R [2],y R [2]),...,(x R[N],y R [N])} (35)

[0137] (x J ,y J )={(x J [1],y J [1]),(x J [2],y J [2]),...,(x J [N],y J [N])} (36)

[0138] Then the objective function to maximize the average confidentiality rate is:

[0139]

[0140] S3. Optimize the objective function of maximizing the average confidentiality rate to enhance the confidentiality performance of the UAV collaborative interference assisted relay network system.

[0141] Specifically, from the constructed maximization average confidentiality rate objective function formula (38), it can be seen that the objective function, constraint C7, constraint C8, constraint C15 and constraint C16 are non-convex, and multiple optimization variables are coupled together and cannot be solved directly.

[0142] Therefore, refer to Figure 2 This is a specific implementation method of step S3. Specifically, the objective function of maximizing the average confidentiality rate can be optimized in the following manner:

[0143] S10, decomposing the optimization problem of the objective function into four sub-problems, namely, the source node transmission power and the drone relay node transmission power optimization sub-problem, the drone jammer node transmission power optimization sub-problem, the drone relay trajectory optimization sub-problem and the drone jammer trajectory optimization sub-problem;

[0144] S11. For each of the four sub-problems, convert the corresponding non-convex objective functions in the corresponding sub-problems into convex objective functions, and solve them using a convex optimization method.

[0145] In one embodiment, the optimization sub-problems of the source node transmission power and the drone relay node transmission power are optimized in the following ways:

[0146] Fixed UAV jammer node transmission power, UAV relay trajectory and UAV jammer trajectory, that is, at a fixed P J 、(x R ,y R ) and (x J ,y J) case, a new objective function based on the source node transmission power and the drone relay node transmission power is obtained:

[0147]

[0148] in, According to formula (39), constraints C9, C10, C12 and C13 are linear constraints, the objective function and constraint C15 are non-convex, and the objective function conforms to the form of subtracting two concave functions.

[0149] Introducing auxiliary variable R b1 = {R b1 [1],R b1 [2],...,R b1 [N]}, the objective function is:

[0150]

[0151] At this time, all constraints are convex constraints. We only need to convert the non-convex objective function into a convex function to convert the optimization problem into a convex optimization problem, and then use the convex optimization method to solve it.

[0152] Although log2(1+γ E [n]) is about P R [n] is a concave function, but its upper bound can be found by a first-order Taylor series. Then, given In the case of:

[0153]

[0154] in,

[0155] Therefore, the approximate expression of the transmission power optimization subproblem of the source node and the UAV relay node is:

[0156]

[0157] The optimization problem P2.2 is an approximation of the original optimization subproblem P2. For the optimization problem P2.2, both the objective function and the constraints are convex, so P2.2 is a convex optimization problem, and the CVX toolbox is used to solve it.

[0158] In one embodiment, the transmission power optimization sub-problem of the drone jammer node is optimized in the following manner:

[0159] The transmission power of the fixed source node and the UAV relay node, the UAV relay trajectory, and the UAV jammer trajectory are fixed. S , P R 、(x R ,yR ) and (x J ,y J ) case, a new objective function based on the transmission power of the UAV jammer node is obtained:

[0160]

[0161] At this time, the objective function is convex and the constraints are also convex, so the UAV jammer transmission power optimization subproblem is a convex optimization problem, and then the CVX toolbox is used to solve it.

[0162] In one embodiment, the drone relay trajectory optimization sub-problem is optimized in the following ways:

[0163] The transmission power of the fixed source node and the UAV relay node, the transmission power of the UAV jammer node, and the trajectory of the UAV jammer, that is, the fixed P S , P R , P J and (x J ,y J ) case, a new objective function based on the trajectory of the drone relay node is obtained:

[0164]

[0165] For the objective function P4, constraints C1, C2 and C5 are convex constraints, the objective function, constraint C7, constraint C15 and constraint C16 are non-convex, and the objective function conforms to the form of subtracting two convex functions.

[0166] Introducing auxiliary variable R b2 = {R b2 [1],R b2 [2],...,R b2 [N]} and D e ={D e [1],D e [2],...,D e [N]}, which is more conducive to analysis, then the optimization target P4 is equivalent to:

[0167]

[0168] At this time, the objective function is convex. We only need to convert the non-convex constraints into convex constraints to convert the optimization problem into a convex optimization problem, and then apply the convex optimization algorithm to solve the sub-problem.

[0169] Specifically, for constraint C7, the left side is convex, and it can be seen that at any feasible point, the first-order Taylor series is its lower bound. Then, at a given point Under the condition of , we can get the following approximation:

[0170]

[0171] in,

[0172] For constraint C15, the right side is about A convex function, at any feasible point At , we can get the following inequality:

[0173]

[0174] in,

[0175] For constraint C16, the left side is convex, and it can be seen that at any feasible point, the first-order Taylor series is its lower bound. Then, at a given point Under the condition of , we can get the following approximation:

[0176]

[0177] in,

[0178] For constraint C18, the right side is about A convex function, we know that at any feasible point At , we can get the following inequality:

[0179]

[0180] in,

[0181] For constraint C19, the right side is convex, and it can be seen that at any feasible point, the first-order Taylor series is its lower bound. Then, at a given point Under the condition of , we can get the following approximation:

[0182]

[0183] in,

[0184] Based on these approximations, the optimization problem can be transformed into:

[0185]

[0186] The optimization problem P4.2 is an approximation of the original optimization subproblem P4. For the optimization problem P4.2, both the objective function and the constraints are convex, so P4.2 is a convex optimization problem, and the CVX toolbox is used to solve it.

[0187] In one embodiment, the trajectory optimization sub-problem of the UAV jammer is optimized in the following manner:

[0188] The transmission power of the fixed source node and the UAV relay node, the transmission power of the UAV interference node, and the UAV relay trajectory, that is, the fixed P S , P R , P J and (x R ,y R ), a new objective function based on the trajectory of the UAV jammer is obtained:

[0189]

[0190] At this time, the objective function, constraint C8 and constraint C16 are non-convex, and the optimization subproblem is still a non-convex optimization problem.

[0191] Although it is difficult to calculate R E [n]About (x J ,y J ) is uneven, but it is easy to get R E [n] is about Convex function, specifically, at any feasible point Introducing this auxiliary condition into the objective function, we get the following approximation:

[0192]

[0193] in,

[0194] For constraint C8, the left side is convex, and it can be seen that at any feasible point, the first-order Taylor series is its lower bound. Then, at a given point Under the condition of , we can get the following approximation:

[0195]

[0196] in,

[0197] For constraint C16, the left side is convex, and it can be seen that at any feasible point, the first-order Taylor series is its lower bound. Then, at a given point Under the condition of , we can get the following approximation:

[0198]

[0199] in,

[0200] According to the above approximation, the trajectory optimization sub-problem of the UAV jammer is approximated as:

[0201]

[0202] For the optimization problem P5.1, both the objective function and the constraints are convex and can be solved using the CVX toolbox.

[0203] Since the trajectory optimization variables and power optimization variables are coupled, the original optimization problem is decomposed into four sub-problems for solution. Then, the convex optimization algorithm is applied to obtain the optimal solution of the UAV jammer transmission power optimization sub-problem and the continuous convex approximation is used to obtain the approximate solutions of the remaining sub-problems. Finally, the block coordinate descent method is applied to obtain the global suboptimal solution, thereby enhancing the confidentiality performance of the UAV cooperative jamming assisted relay network system.

[0204] In summary, the UAV relay system assisted jamming method has the following effects:

[0205] (1) This method is suitable for UAV relay systems under hardware damage conditions. Other UAVs are introduced to act as jammers to solve the problem that the system's positive confidentiality rate is overly dependent on channel quality. At the same time, considering the transmission power, motion trajectory, and information causality constraints, a resource allocation algorithm is designed to maximize the system's average confidentiality rate.

[0206] (2) This method proposes a two-layer solution framework, solving the variable coupling problem in the outer layer and introducing slack variables in the inner layer and using continuous convex approximation to obtain approximate solutions to subproblems, which has low complexity;

[0207] (3) This method takes into account a more realistic flight environment, prohibiting drone flights in some private or public places, and introducing a minimum spacing constraint between drones to avoid collisions between drones.

[0208] In order to verify the effect of this application, the following simulation experimental data is provided:

[0209] Simulation conditions: The coordinates of the source node, destination node and eavesdropper are (0,0,0), (2000,0,0) and (1000,-500,0) respectively. The center coordinates of the no-fly zone are (1000,0,0) and (1000,-500,0) respectively. The flight altitude of the drone is 100m. The minimum distance between drones is 1m. The communication bandwidth is 20MHz. The noise power spectral density is -169dBm / Hz.

[0210] Figure 4It is a comparison chart of the average confidentiality rate of the scheme proposed in this application and the three benchmark schemes with the number of iterations. Among them, the three benchmark schemes are the jammer-free scheme, the power optimization scheme, and the trajectory optimization scheme. The jammer-free scheme refers to a scheme without jammer nodes, the power optimization scheme refers to a scheme that only optimizes the node transmission power but does not optimize the node trajectory based on the presence of jammer nodes, and the trajectory optimization scheme refers to a scheme that only optimizes the trajectory but does not optimize the node transmission power based on the presence of jammer nodes. The average confidentiality rate of the four schemes increases with the increase in the number of iterations, and all can reach convergence within 5 iterations. In addition, when the number of iterations is 5, the average confidentiality rates of the proposed scheme and the three benchmark schemes are 4.51 bits / s / Hz, 3.65 bits / s / Hz, 4.25 bits / s / Hz and 4.13 bits / s / Hz, respectively. The average confidentiality rate of the proposed scheme is significantly higher than that of the other three benchmark schemes, and the trajectory optimization scheme can achieve a higher average confidentiality rate than the power optimization scheme. Specifically, this is because the drone relay can adjust its trajectory away from the eavesdropper and hover close to the source and destination nodes, and the drone jammer can adjust its trajectory to hover near the eavesdropper;

[0211] Figure 5 It is a comparison chart of the trajectory changes of the drone relay and the drone jammer in the proposed scheme and the three benchmark schemes. The drone relay and the drone jammer in the proposed scheme and the three benchmark schemes all avoid the drone no-fly zone to complete the task. For the power optimization scheme, the two drones fly from the starting point to the end point along a pre-fixed trajectory during the mission; for the jammer-free scheme, the drone relay first flies from the starting point to the source node, then flies away from the eavesdropper to the destination node, and finally flies back to the drone end point; for the proposed scheme and the trajectory optimization scheme, the drone relay first flies from the starting point to the source node, and hovers over the source node as much as possible, then flies away from the eavesdropper to the destination node, and hovers over the destination node as much as possible, and finally flies back to the end point, while in these two schemes, the drone jammer flies from the starting point to the eavesdropper and hovers near the eavesdropper. Because there is a no-fly zone near the eavesdropper, the drone jammer cannot fly over the eavesdropper, so the drone jammer flies along the edge of the no-fly zone. Compared with the trajectory optimization scheme, the coordinate points of the drone relay near the source node and the destination node in the proposed scheme are denser, from which it can be concluded that the number of hovering time slots at the source node and the destination node is greater, so the drone relay of the proposed scheme hovers near the source node and the destination node for a longer time.

[0212] Figure 6This is a comparison chart of the drone trajectories under different no-fly zone constraints of the solution proposed in this application. For the drone relay, when there is no no-fly zone constraint, the drone relay first flies to the source node, then flies to the destination node away from the eavesdropper, and finally flies to the end point. When there is a no-fly zone constraint, the drone relay flies to the destination node along the edge of the no-fly zone away from the eavesdropper, and as the radius of the no-fly zone increases, the trajectory of the drone relay will also move in a direction farther away from the eavesdropper. For the drone jammer, when there is no no-fly zone constraint, the drone jammer hovers at the position closest to the eavesdropper and then flies to the end point; but when there is a no-fly zone constraint, the drone jammer cannot reach the ideal position to fight the eavesdropper and can only fly along the edge of the no-fly zone;

[0213] Figure 7 This is a comparison chart of the average confidentiality rate versus average transmission power curve under different hardware damage levels of the solution proposed in this application. Under different hardware damage levels, when the average transmission power increases, the proposed solution can achieve higher confidentiality performance. When the transmission power is 1W, the average confidentiality rates that can be achieved are 5.045 bits / s / Hz and 4.548 bits / s / Hz respectively. This is because when the transmission power increases, the signal-to-noise ratio of the signal received by the destination node is greater, while the signal-to-noise ratio of the confidential signal eavesdropped by the eavesdropper is smaller, so higher confidentiality performance can be achieved. As the degree of hardware damage increases, the confidentiality performance of the system decreases. This is because the increased degree of hardware damage causes the received signal to deviate more from the ideal signal.

[0214] Figure 8 This is a comparison chart of the transmission power change curve in each time slot under different hardware damage levels of the solution proposed in this application. When the drone relay is close to the source node, the transmission power of the source node is higher; when the drone relay is close to the destination node, the transmission power of the drone relay node is higher. This is because when the drone relay is close to the source node, there is a higher channel gain between the source node and the drone relay, and when the drone relay is close to the destination node, there is a higher channel gain between the drone relay and the destination node. At the same time, when the degree of hardware damage increases, the transmission power of the source node decreases when the drone relay is close to the source node; the transmission power of the drone relay node decreases when the drone relay is close to the destination node. This is because it is affected by hardware damage. In addition, for the drone jammer, when the drone jammer is far away from the eavesdropper, the drone jammer will increase the transmission power to ensure the confidentiality performance of the system; when the drone relay is close to the source node, the destination node can achieve a lower throughput, so the drone jammer needs a higher transmission power;

[0215] Fig. 9This is a comparison chart of the drone trajectory change curve in each time slot under different hardware damage levels of the solution proposed in this application. As the hardware damage level increases, the signal-to-noise ratio of the legitimate destination node will decrease, so the drone relay trajectory moves in a direction farther away from the eavesdropper, and the drone relay hovers at the source node and the destination node for a shorter time.

[0216] Fig.10 This is a comparison of the average confidentiality rate of the scheme proposed in this application and the three benchmark schemes as the average transmission power changes. When the transmission power is 1W, the average confidentiality rates achievable by the four schemes are 5.0456bits / s / H, 4.7545bits / s / Hz, 4.6767bits / s / Hz and 4.3091bits / s / Hz, respectively, which shows that the average confidentiality rate achievable by the proposed scheme is the highest, while the average confidentiality rate achievable by the power optimization scheme is the lowest. As the average transmission power increases, the average confidentiality rates achievable by the four schemes all increase. This is because the increase in transmission power makes the legitimate destination node receive a higher signal-to-noise ratio, and the increase in the transmission power of the drone jammer further suppresses the eavesdropper's eavesdropping rate.

[0217] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0218] The above are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Any feature disclosed in this specification (including the abstract and drawings), unless otherwise stated, can be replaced by other equivalent or alternative features with similar purposes. That is, unless otherwise stated, each feature is only an example of a series of equivalent or similar features.

Claims

1. A cooperative jamming method based on a drone relay system, characterized in that: A collaborative interference model based on a drone relay system is established, wherein the collaborative interference model includes a source node S, a drone relay node R, a drone jammer node J, a legitimate destination node D, and an eavesdropper E. The drone jammer node J is used to transmit an interference signal to counter the eavesdropper. Based on the collaborative interference model, construct an objective function for maximizing the average confidentiality rate; Optimizing the objective function of maximizing the average confidentiality rate, thereby achieving the confidentiality performance of the UAV cooperative interference assisted relay network system; The objective function of maximizing the average confidentiality rate is constructed in the following way: in, To maximize the average confidentiality rate, R D [n] is the achievable data rate of the destination node in each time slot, R E [n] is the achievable data rate of the eavesdropper in each time slot; N is the number of time slots, i2∈{I1+1, I1+2, ..., I}, n∈{1,2, ..., N}, i1∈{1,2, ..., I1}; is the number of no-fly zones relayed by the drone, and I represents the maximum number of no-fly zones; (x R [n],y R [n], H) and (x J [n],y J [n], H) are the coordinates of the drone relay and drone jammer in each time slot; (x R0 ,y R0 , H) and (x RF ,y RF , H) are the flight start and end coordinates of the drone relay; (x J0 ,y J0 , H) and (x JF ,y JF , H) are the starting and ending coordinates of the UAV jammer; (x NOi ,y NOi , 0), i∈{i1,i2} is the center of the no-fly zone, r NOi is the radius; V is the moving distance of the UAV relay and UAV jammer, which cannot exceed the maximum movable distance; P S [n], P R [n] and P J [n] represents the transmission power of the source node, UAV relay and UAV jammer, respectively, P Smax , P Rmax and P Jmax They represent the corresponding peak power, and They represent the corresponding mean power respectively; d is the minimum distance between two UAVs to avoid collision; The objective function of maximizing the average confidentiality rate is optimized in the following way: The optimization problem of the objective function is decomposed into four sub-problems, namely, the source node transmission power and the UAV relay node transmission power optimization sub-problem, the UAV jammer node transmission power optimization sub-problem, the UAV relay trajectory optimization sub-problem and the UAV jammer trajectory optimization sub-problem; For each of the four sub-problems, the corresponding non-convex objective functions in the corresponding sub-problems are converted into convex objective functions, and the convex optimization method is used to solve them; The optimization sub-problems of the source node transmission power and the drone relay node transmission power are optimized in the following ways: Fix the transmission power of the UAV jammer node, the UAV relay trajectory, and the UAV jammer trajectory, and obtain a new objective function based on the transmission power of the source node and the transmission power of the UAV relay node: Introducing auxiliary variable R b1 = {R b1 [1],R b1 [2],...,R b1 [N]}, and auxiliary variables and Among them, d j [n],j∈{SR,RD,RE,JE} is the distance from the drone relay to the source node, the drone relay to the destination node, the drone relay to the eavesdropper, and the drone jammer to the eavesdropper. The non-convex constraints in the constraints are converted into convex constraints, and the non-convex objective function is converted into a convex function. The approximate expression of the transmission power optimization sub-problem of the source node and the UAV relay node is: in, is an auxiliary variable introduced in the derivation process and is defined as θ1 is an auxiliary variable introduced in the derivation process and is defined as: in, Then use convex optimization method to solve; For the sub-problem of optimizing the transmission power of the UAV jammer node, the optimization is performed in the following ways: the transmission power of the source node and the UAV relay node, the UAV relay trajectory and the UAV jammer trajectory are fixed, and a new objective function based on the transmission power of the UAV jammer node is obtained: Then the convex optimization method is used to solve it.

2. The method according to claim 1, characterized in that: When the drone relay has hardware damage, the drone relay is adjusted to move away from the eavesdropper and hover close to the source node and the destination node. At the same time, the drone jammer is adjusted to hover near the eavesdropper.

3. The method according to claim 2, characterized in that: When the drone relay performs communication tasks within T time, when the hardware damage of the drone relay increases, the drone relay is adjusted to move in a direction farther away from the eavesdropper to reduce the time the drone relay hovers around the source node and the destination node.

4. The method according to claim 1, characterized in that: The signal y received by the drone relay node R [n] is: Among them, x S [n]~CN(0,1) indicates the source node transmission signal, is the distortion noise caused by hardware damage, n R ~CN(0,σ2) represents the additive noise of the drone relay, h SR [n] represents the channel coefficient between the source node and the drone relay node, and the channel gain between the drone jammer and the eavesdropper is: Among them, g j [n]=|h j [n]| 2 , j∈{SR, RD, RE, JE}, β0 represents the reference channel power when the reference distance is 1 meter, d j [n] is the distance from the drone relay to the source node, the drone relay to the destination node, the drone relay to the eavesdropper, and the drone jammer to the eavesdropper.

5. The method according to claim 1, characterized in that: For the drone relay trajectory optimization sub-problem, optimization is performed in the following ways: By fixing the transmission power of the source node and the UAV relay node, the transmission power of the UAV jammer node, and the trajectory of the UAV jammer, a new objective function based on the trajectory of the UAV relay node is obtained: Introducing auxiliary variable R b2 = {R b2 [1],R b2 [2],...,R b2 [N]} and D e ={D e [1],D e [2],...,D e [N]}, the non-convex constraints in the constraints are converted into convex constraints, and the non-convex objective function is converted into a convex function. The approximate expression of the drone relay trajectory optimization subproblem is: ψ1 is defined as: in, ψ2 is defined as: in, ψ3 is defined as: in, ψ4 is defined as: in, ψ5 is defined as: in, 6. The method according to claim 1, characterized in that: For the UAV jammer trajectory optimization sub-problem, optimization is performed in the following ways: By fixing the transmission power of the source node and the UAV relay node, the transmission power of the UAV jammer node, and the UAV relay trajectory, a new objective function based on the trajectory of the UAV jammer is obtained: Introducing auxiliary conditions Convert the non-constraints in the constraints into convex constraints and convert the non-convex objective function into a convex function. The approximate expression of the UAV jammer trajectory optimization subproblem is: θ2 is defined as: in, ψ6 is defined as: in, ψ7 is defined as: in, Then the convex optimization method is used to solve it.

Images