A concealed communication method and system combining unmanned aerial vehicle and intelligent reflective surface

Abstract

The present invention provides a covert communication method and system combining a drone and an intelligent reflective surface, the method comprising: extracting electromagnetic environment parameters; constructing and training an evolutionary neural network, designing a multi-objective fitness function, using an evolutionary algorithm to train to obtain an optimal network that best adapts to the current electromagnetic environment, and outputting an optimal covert communication scheme based on the optimal covert communication scheme; and controlling the actions of the drone and the intelligent reflective surface in the communication area according to the optimal covert communication scheme. The present invention is aimed at the actual covert communication scenario of a drone equipped with an intelligent reflective surface, and effectively implements covert communication by jointly optimizing the communication resources of legitimate users and intelligent reflective surfaces, and utilizes the advantages of evolutionary neural networks to digitize complex networks through chromosome encoding, thereby optimizing the network by simulating the evolutionary process in nature using an evolutionary algorithm, which can effectively solve the neural network construction problem and covert communication problem under complex problems.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a covert communication method and system combining an unmanned aerial vehicle and an intelligent reflective surface. Background Art

[0002] As the requirements for communication security continue to increase, information encryption technology and physical layer security technology are difficult to meet the current wireless communication security needs. As a communication method that can hide the communication behavior itself, covert communication technology effectively improves the security of information transmission by using noise and channel uncertainty to counter the detection of eavesdroppers. In particular, the use of artificial intelligence methods to achieve adaptive covert communication will overcome the shortcomings of traditional algorithms and further promote the promotion and application of covert communication.

[0003] Khurram et al. proposed using a full-duplex receiver to achieve wireless covert communication. The receiver generates artificial noise with random power changes, which increases the uncertainty of the signal received by the eavesdropper and effectively achieves covert communication. This method introduces artificial noise, which poses the risk of self-interference and information leakage. At the same time, the receiver radiation power easily exposes the receiver's location.

[0004] M. Wang et al. proposed a scheme using multi-antenna relay transmission, which jointly designs the multi-antenna beamforming vectors of the source node and the relay node to meet the concealment requirements under low-power transmission conditions. This method uses fixed relay nodes to achieve covert communication. The fixed deployment of relay nodes has poor mobility and is easy to be detected and destroyed, resulting in interruption of information transmission.

[0005] S.Bi et al. proposed an intelligent reflective surface-assisted UAV covert communication system based on deep reinforcement learning algorithm, which enables the dual-depth Q network optimization algorithm to effectively provide high-dimensional data processing and decision-making capabilities. This method uses the deep reinforcement learning method, and the trained network has poor generalization ability, and cannot be quickly adjusted and deployed to handle covert communication tasks in practical applications. Summary of the invention

[0006] The present invention provides a covert communication method and system combining an unmanned aerial vehicle and an intelligent reflective surface, so as to solve at least one of the above-mentioned defects in the prior art. For the scenario where the unmanned aerial vehicle is equipped with an intelligent reflective surface to assist in covert communication, an evolutionary neural network algorithm is used to effectively implement covert communication.

[0007] In a first aspect, the present invention provides a covert communication method combining an unmanned aerial vehicle and an intelligent reflective surface, comprising: extracting electromagnetic environment parameters as input of a neural network; the electromagnetic environment parameters include: three-dimensional position information of a legitimate user's transmitting end, three-dimensional position information of a legitimate user's receiving end, an estimated activity range of an eavesdropper, channel distribution information of a link between a legitimate user's transmitting end and an intelligent reflective surface, channel distribution information of a link between an intelligent reflective surface and a legitimate user's receiving end, and a detection threshold of an eavesdropper; wherein the intelligent reflective surface is mounted on the unmanned aerial vehicle; taking maximization of the covert communication rate of the legitimate user and maximization of the detection error probability of the eavesdropper as optimization goals, taking transmission power, lossless reflection of the intelligent reflective surface, and flight altitude of the unmanned aerial vehicle as constraints, designing a multi-objective fitness function, and optimizing the neural network using an evolutionary algorithm to obtain an optimal neural network and an optimal covert communication scheme output by the optimal neural network; based on the optimal covert communication scheme, controlling the actions of the unmanned aerial vehicle and the intelligent reflective surface in the communication area; wherein the optimal covert communication scheme includes the position of the unmanned aerial vehicle and the reflection coefficient matrix of the intelligent reflective surface.

[0008] According to the covert communication method combining a drone and an intelligent reflective surface provided by the present invention, the estimated activity range of the eavesdropper is estimated according to the position distribution law of the eavesdropper; the position where the eavesdropper appears obeys a Gaussian distribution.

[0009] According to the covert communication method combining a drone and an intelligent reflective surface provided by the present invention, the neural network is a three-layer neural network. When optimizing the neural network, the number of hidden layer nodes and corresponding weight parameters of the neural network are optimized.

[0010] According to the covert communication method combining a drone and an intelligent reflective surface provided by the present invention, the multi-objective fitness function is specifically:

[0011]

[0012]

[0013] Among them, F1 represents the covert communication rate of the legitimate user, F2 represents the detection error probability of the eavesdropper, R is the covert communication rate, λ is the detection error probability of the eavesdropper, and R max is the maximum value of the covert communication rate, λ max is the maximum error probability, and a and b are weighting coefficients.

[0014] According to the covert communication method combining a drone and an intelligent reflective surface provided by the present invention, an evolutionary algorithm is used to optimize a neural network with fixed input nodes and output nodes, including performing a chromosome encoding operation on the neural network: a group of chromosomes represents a topological structure, each chromosome represents a connection in a network, and each chromosome has four genes: a gene sequence number, a connection vector, an enable mark, and a weight parameter; wherein the gene sequence number is used to mark the connection vector; the connection vector is used to mark the input node and the output node; the enable mark is used to mark the enable state of the connection; and the weight parameter is the weight parameter of the connection.

[0015] According to the covert communication method combining a drone and an intelligent reflective surface provided by the present invention, after the chromosome encoding of the neural network, the evolutionary algorithm also includes: performing mutation operations, crossover operations, population division, execution of elite strategies and offspring inheritance on the chromosomes generated by the encoding to complete the optimization of the neural network.

[0016] In a second aspect, the present invention also provides a covert communication system combining a drone and an intelligent reflective surface, comprising: an environmental parameter acquisition module, a network construction and training module, and a program execution module;

[0017] The environmental parameter acquisition module is used to extract electromagnetic environment parameters as input of the neural network; the electromagnetic environment parameters include: three-dimensional position information of the legitimate user's transmitting end, three-dimensional position information of the legitimate user's receiving end, estimated activity range of the eavesdropper, channel distribution information of the link between the legitimate user's transmitting end and the intelligent reflecting surface, channel distribution information of the link between the intelligent reflecting surface and the legitimate user's receiving end, and detection threshold of the eavesdropper; wherein the intelligent reflecting surface is mounted on the drone;

[0018] The network construction and training module is used to maximize the covert communication rate of legitimate users and the detection error probability of eavesdroppers as optimization goals, and to design a multi-objective fitness function with the transmission power, lossless reflection of the intelligent reflective surface, and the flight altitude of the UAV as constraints, and to optimize the neural network using an evolutionary algorithm to obtain the optimal neural network and the optimal covert communication solution output by the optimal neural network;

[0019] The scheme execution module is used to control the actions of the UAV and the intelligent reflective surface in the communication area based on the optimal covert communication scheme; wherein the optimal covert communication scheme includes the position of the UAV and the reflection coefficient matrix of the intelligent reflective surface.

[0020] In a third aspect, the present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the covert communication method combining a drone and an intelligent reflective surface as described above are implemented.

[0021] In a fourth aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of a covert communication method combining a drone and an intelligent reflective surface as described in any of the above.

[0022] In a fifth aspect, the present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the covert communication method combining a drone and an intelligent reflective surface as described above.

[0023] The covert communication method and system combining a UAV and an intelligent reflective surface provided by the present invention fully utilize the maneuverability of the UAV and the high efficiency and low cost of the intelligent reflective surface in terms of hardware, and utilize the complementary advantages of evolutionary algorithms and neural networks in terms of algorithms, effectively solving the communication problems under the conditions of covert communication assisted by an intelligent reflective surface mounted on a UAV.

[0024] The present invention aims at the actual covert communication scenario of unmanned aerial vehicles equipped with intelligent reflective surfaces. It effectively implements covert communication by jointly optimizing the communication resources of legitimate users and intelligent reflective surfaces. It takes advantage of evolutionary neural networks and digitizes complex networks through chromosome encoding, so as to optimize the network by simulating the evolutionary process in nature using evolutionary algorithms. It can effectively solve the neural network construction problems and covert communication problems under complex problems. It has the advantages of strong generalization ability and high processing efficiency, and can quickly adjust and deploy the processing of covert communication tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0026] Figure 1 This is one of the flow diagrams of the covert communication method combining a UAV and an intelligent reflective surface provided by the present invention;

[0027] Figure 2 This is the second flow chart of the covert communication method combining the UAV and the intelligent reflective surface provided by the present invention;

[0028] Figure 3 It is a main operation schematic diagram of the evolutionary neural network algorithm provided by the present invention;

[0029] Figure 4 It is a schematic diagram of fitness change of the method for assisting covert communication by using an intelligent reflective surface mounted on a UAV based on an evolutionary neural network provided by the present invention;

[0030] Figure 5 It is a schematic diagram of population changes during the evolution of the method for assisting covert communication using an intelligent reflective surface mounted on a UAV based on an evolutionary neural network provided by the present invention;

[0031] Figure 6 It is a schematic diagram of the optimal network structure of the method for assisting covert communication using an intelligent reflective surface mounted on a UAV based on an evolutionary neural network provided by the present invention;

[0032] Figure 7 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] It should be noted that in the description of the embodiments of the present invention, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "include one..." do not exclude the presence of other identical elements in the process, method, article or device including the elements. The orientation or positional relationship indicated by the terms "upper", "lower", etc. is based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances.

[0035] Combine the following Figure 1-Figure 7 The covert communication method and device combining a UAV and an intelligent reflective surface provided in an embodiment of the present invention are described.

[0036] Figure 1 This is one of the flow charts of the covert communication method combining the UAV and the intelligent reflective surface provided by the present invention, such as Figure 1 As shown, including but not limited to the following steps:

[0037] Step 101: Extract electromagnetic environment parameters as input of the neural network.

[0038] Among them, the electromagnetic environment parameters include: three-dimensional position information of the legitimate user's transmitting end, three-dimensional position information of the legitimate user's receiving end, estimated activity range of the eavesdropper, channel distribution information of the link between the legitimate user's transmitting end and the intelligent reflecting surface, channel distribution information of the link between the intelligent reflecting surface and the legitimate user's receiving end, and the detection threshold of the eavesdropper; the intelligent reflecting surface is mounted on the drone.

[0039] Step 102: Taking maximizing the covert communication rate of legitimate users and maximizing the detection error probability of eavesdroppers as optimization goals, and taking the transmission power, lossless reflection of the intelligent reflective surface, and the flight altitude of the UAV as constraints, a multi-objective fitness function is designed, and an evolutionary algorithm is used to optimize the neural network to obtain the optimal neural network and the optimal covert communication solution output by the optimal neural network.

[0040] Step 103: Based on the optimal covert communication solution, control the movements of the UAV and the intelligent reflective surface in the communication area.

[0041] Among them, the optimal covert communication scheme includes the position of the UAV and the reflection coefficient matrix of the smart reflective surface.

[0042] Figure 2 This is the second flow chart of the covert communication method combining the UAV and the intelligent reflective surface provided by the present invention. Figure 2 The technical solution of the present invention is further described.

[0043] Based on the contents of the above embodiments, as an optional embodiment, the present invention provides a covert communication method combining a drone and an intelligent reflective surface, extracting electromagnetic environment parameters as input of a neural network to construct and train an evolutionary neural network.

[0044] Estimate the activity range of the eavesdropper. Estimate the activity range of the eavesdropper. The eavesdropper is active in a certain range. Its location (x Eve ,y Eve ,z Eve ) obeys Gaussian distribution

[0045] It is expressed as:

[0046]

[0047]

[0048]

[0049] in, is the center of the estimated eavesdropper’s location, Δx Eve , Δy Eve , Δz Eve is the standard deviation of the eavesdropper’s position, f(x Eve ), f(y Eve ), f(z Eve ) represent the Gaussian distribution functions of the three-dimensional positions of the eavesdroppers.

[0050] The channel distribution information of the link between the legitimate user transmitter and the smart reflector, and the channel distribution information of the link between the smart reflector and the legitimate user receiver, the signal goes through the Rayleigh channel, which is expressed as:

[0051]

[0052] Among them, x, y, z represent the three-dimensional position coordinate information of the node, the subscripts are used to identify different types of nodes, A and B represent the legitimate user transmitter and receiver, Eve is the eavesdropper, UAV is the unmanned aerial vehicle equipped with an intelligent reflective surface device, and h ij represents the channel distribution information of node i and node j, α0 is the path loss index, and ρ is the carrier frequency f c The related constant, d ij is the distance between two points, g ij represents Rayleigh fading with a mean of 0 and a variance of 1, and the speed of light c = 3×10 8 m / s.

[0053] By solving the optimization function of the best detection threshold as the detection threshold of the eavesdropper, the best detection threshold is expressed as:

[0054]

[0055] in, In the nth (n∈{1,…,N}) time slot, p UAV is the reflected power of the UAV, p A The transmission power of the legitimate user transmitter, σ Eve represents the noise at the eavesdropper, P(H0) and P(H1) represent the prior probabilities that the legitimate user does not perform covert communication and performs covert communication, respectively, τ op represents the optimal detection threshold, and τ represents the detection threshold.

[0056] Based on the contents of the above embodiments, as an optional embodiment, the present invention provides a covert communication method combining a drone and an intelligent reflective surface, wherein the neural network is a three-layer neural network, and when optimizing the neural network, the number of hidden layer nodes and the corresponding weight parameters of the neural network are optimized.

[0057] The present invention adopts evolutionary algorithm to construct and train the network, which has fixed input and output node numbers and evolves the structure and weight of hidden layer nodes simultaneously. The network is trained unsupervised by designing multi-objective fitness function to solve the problem of objective optimization with constraints, and the ultimate goal is to enable legitimate users to obtain covert communication schemes that eavesdroppers cannot detect.

[0058] The multi-objective fitness function is an indicator for evaluating the current network. The training process is the process of maximizing the multi-objective fitness, which can be expressed as:

[0059]

[0060]

[0061] Among them, F1 represents the covert communication rate of the legitimate user, F2 represents the detection error probability of the eavesdropper, R is the covert communication rate, λ is the detection error probability of the eavesdropper, and R max is the maximum value of the covert communication rate, λ max is the maximum error probability, a and b are weighted coefficients. To maximize the covert communication rate, when the covert communication rate is higher, F1 is larger, that is, the fitness is higher; at the same time, to maximize the probability of eavesdropper detection error, when the detection error probability is higher, F2 is larger, that is, the fitness is higher.

[0062] The covert communication rate R and detection error probability λ in the multi-objective fitness function are expressed as:

[0063]

[0064] λ=P(H0)P(D1|H0)+P(H1)P(D0|H1),

[0065] Among them, σ B is the noise at the receiving end of the legitimate user, H0 and H1 represent the events that A does not send a covert communication signal and sends a covert communication signal, respectively, P(H0) and P(H1)=1-P(H0) represent the prior probabilities that A does not send a covert signal and A sends a covert signal, respectively, P(D1|H0) and P(D0|H1) represent the false alarm probability and missed alarm probability, respectively, and the judgment criteria are expressed as follows:

[0066] Y>τ op ,D1,

[0067] Y<τ op ,D0,

[0068] Among them, τ op represents the optimal detection threshold of Eve, D0 and D1 represent the judgment conclusions of A not performing covert communication and performing covert communication respectively, and Y represents the receiving power of Eve, which is expressed as follows:

[0069]

[0070] The objective optimization problem with constraints is to maximize the covert communication rate of legitimate users and the detection error probability of eavesdroppers, with the transmission power, lossless reflection of the intelligent reflective surface and the flight altitude of the UAV as constraints, which can be expressed as:

[0071]

[0072] in, is the maximum transmission power of the legitimate user transmitter, z max UAV maximum flight altitude,θ l represents the phase shift angle of the lth reflection unit, that is, Represents the reflection parameter of the lth row and lth column in the reflection coefficient matrix. Note that j here specifically refers to the complex number representation.

[0073] Figure 3 This is a schematic diagram of the main operation of the evolutionary neural network algorithm provided by the present invention. The evolutionary algorithm includes six steps: chromosome encoding, mutation operation, population division, elite strategy, and offspring inheritance. Specifically:

[0074] Step 1. Chromosome encoding is to represent the chromosome group as a topological structure. Each chromosome represents a connection in a network. A single chromosome consists of four genes, namely gene sequence number, connection vector, enable flag, and weight parameter. The encoding method is expressed as follows:

[0075] genome={Innonumber,Connection,Status,Weight},

[0076] Among them, the Innonumber is the sequence number of the gene, which is used to mark the connection vector, written in sequence starting from 1, and can be abbreviated as I; the Connection is a connection vector, used to mark the input node and the output node, denoted by C; the Status is the enable mark of the gene, used to mark the enable state of the connection, 0 represents prohibition, 1 represents enablement, denoted by S; the Weight is the weight parameter of the connection, denoted by W.

[0077] Step 2. Mutation operation, including adding connection and adding node. Adding connection is to add a chromosome, and adding node is to prohibit a chromosome and then add two chromosomes pointing to the two nodes of the original connection respectively.

[0078] Step 3. The crossover operation is to cross the two topological structures as parents, that is, to cross the two chromosome groups, and align the matching chromosomes in the two chromosome groups according to the corresponding sequence numbers, and inherit from the parent with higher fitness. If the fitness of the two parents is the same, they will be inherited randomly; unmatched chromosomes will be directly inherited. When matching, that is, the sequence numbers in the two genomes are the same, calculate the average weight difference of the matching genes:

[0079]

[0080] Where K is the number of matching genes, W 1k and W 2k are the weights of the matching genes in the two sets of genomes. When there is no match, that is, one of the two sets of genomes exists but the other does not, the length of the sequence number groups of any two sets of chromosomes in order is s1 and s2 respectively, and the minimum length S = min(s1, s2) is taken. When the unmatched sequence number is less than S, it is mutually exclusive (disjoint) and recorded as D. When the unmatched sequence number is greater than S, it is excess (excess), recorded as E;

[0081] Step 4: Population division by topological similarity To distinguish different populations, the importance of the three factors is adjusted by coefficients c1, c2, and c3. The maximum length of the sequence number set in the genome I = max(s1, s2, ...) is used to normalize the size of the genome. The similarity thresholds δ are set respectively. t Divide the population and assume that each population shares a fitness;

[0082] Step.5 The elite strategy is that the population size with stronger fitness will grow with the number of evolved generations. The growth and contraction of a population depends on the change of the average fitness of the population adjustment:

[0083]

[0084] Among them, f gh is the fitness of the g-th gene in population h, is the average fitness of the population, N h ′ and N h Respectively represent the previous and new number of individuals in population h;

[0085] Step.6 Offspring inheritance is to select the top r% genes with the highest fitness in each population for mating to generate offspring to replace the original population.

[0086] In order to further demonstrate the technical effect of the present invention, the present invention has carried out experimental simulation:

[0087] Simulation conditions: Consider a scenario where a drone is equipped with an intelligent reflective surface to assist in covert communication, which includes four nodes, namely a legitimate transmitter, a legitimate receiver, a drone-equipped intelligent reflective surface relay, and an eavesdropper. In this example, the number of reflective units L of the IRS is 9.

[0088] Simulation content: The feasibility of the proposed method is verified by analyzing the fitness changes of the evolving neural network, the population changes during the evolution process, and the optimal network structure. The results are as follows: Figure 4 , Figure 5 and Figure 6 shown.

[0089] Figure 4 It is a schematic diagram of fitness change of the method for assisting covert communication using an intelligent reflective surface mounted on a UAV based on an evolutionary neural network provided by the present invention. As shown in the figure, the algorithm converges at the 147th generation and the optimal fitness reaches the maximum.

[0090] Figure 5 It is a schematic diagram of population changes during the evolution of the method for covert communication assisted by intelligent reflective surfaces on UAVs based on evolutionary neural networks provided by the present invention. The initial population is 50, with 100 in each generation. After 124 generations, the dominant population expands and reproduces to become the optimal population, and the algorithm converges.

[0091] Figure 6 is a schematic diagram of the optimal network structure of the method for assisting covert communication using an intelligent reflective surface mounted on a UAV based on an evolutionary neural network provided by the present invention, such as Figure 6 As shown, the optimal network (optimal neural network) is a three-layer network, in which the gray squares are input nodes including 19 electromagnetic environment parameters, and the numbers represent the input node numbers; the white circles are hidden layer nodes, and the numbers represent the weights; the blue circles are output nodes including 12 output nodes, and the numbers represent the output node numbers.

[0092] In summary, the present invention aims at the actual covert communication scenario of unmanned aerial vehicles equipped with intelligent reflective surfaces. It effectively implements covert communication by jointly optimizing the communication resources of legitimate users and intelligent reflective surfaces. It takes advantage of evolutionary neural networks and digitizes complex networks through chromosome encoding, so as to optimize the network by simulating the evolutionary process in nature using evolutionary algorithms. It can effectively solve the problems of neural network construction and covert communication under complex problems. It has the advantages of strong generalization ability and high processing efficiency, and can quickly adjust and deploy to handle covert communication tasks.

[0093] The present invention also provides a concealed communication system combining an unmanned aerial vehicle and an intelligent reflective surface, the device comprising:

[0094] The environmental parameter acquisition module is used to extract electromagnetic environment parameters as input of the neural network; the electromagnetic environment parameters include: three-dimensional position information of the legitimate user's transmitting end, three-dimensional position information of the legitimate user's receiving end, estimated activity range of the eavesdropper, channel distribution information of the link between the legitimate user's transmitting end and the intelligent reflecting surface, channel distribution information of the link between the intelligent reflecting surface and the legitimate user's receiving end, and detection threshold of the eavesdropper; wherein the intelligent reflecting surface is carried on the drone;

[0095] The network construction and training module is used to maximize the covert communication rate of legitimate users and the detection error probability of eavesdroppers as optimization goals, and to design a multi-objective fitness function with the transmission power, lossless reflection of the intelligent reflective surface, and the flight altitude of the UAV as constraints, and to optimize the neural network using an evolutionary algorithm to obtain the optimal neural network and the optimal covert communication scheme output by the optimal neural network;

[0096] The scheme execution module is used to control the movements of the UAV and the intelligent reflective surface in the communication area based on the optimal covert communication scheme; wherein the optimal covert communication scheme includes the position of the UAV and the reflection coefficient matrix of the intelligent reflective surface.

[0097] It should be noted that the covert communication system combining a drone and an intelligent reflective surface provided in an embodiment of the present invention can execute the covert communication method combining a drone and an intelligent reflective surface described in any of the above embodiments during specific operation, which will not be elaborated in this embodiment.

[0098] Figure 7 is a schematic diagram of the structure of the electronic device provided by the present invention, such as Figure 7As shown, the electronic device may include: a processor 710 , a communication interface 720 , a memory 730 and a communication bus 740 , wherein the processor 710 , the communication interface 720 , and the memory 730 communicate with each other via the communication bus 740 . The processor 710 can call the logic instructions in the memory 730 to execute the covert communication method combining the unmanned aerial vehicle and the intelligent reflecting surface, the method comprising: extracting electromagnetic environment parameters as input of the neural network; the electromagnetic environment parameters include: three-dimensional position information of the legitimate user transmitting end, three-dimensional position information of the legitimate user receiving end, estimated activity range of the eavesdropper, channel distribution information of the link between the legitimate user transmitting end and the intelligent reflecting surface, channel distribution information of the link between the intelligent reflecting surface and the legitimate user receiving end, and detection threshold of the eavesdropper; wherein the intelligent reflecting surface is carried on the unmanned aerial vehicle; taking the maximization of the covert communication rate of the legitimate user and the maximization of the detection error probability of the eavesdropper as optimization goals, taking the transmission power, the lossless reflection of the intelligent reflecting surface and the flight altitude of the unmanned aerial vehicle as constraints, designing a multi-objective fitness function, and optimizing the neural network using an evolutionary algorithm to obtain the optimal neural network and the optimal covert communication scheme output by the optimal neural network; based on the optimal covert communication scheme, controlling the actions of the unmanned aerial vehicle and the intelligent reflecting surface in the communication area; wherein the optimal covert communication scheme includes the position of the unmanned aerial vehicle and the reflection coefficient matrix of the intelligent reflecting surface.

[0099] In addition, the logic instructions in the above-mentioned memory 730 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0100] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the covert communication method combining a drone and an intelligent reflective surface provided in the above-mentioned embodiments, and the method includes: extracting electromagnetic environment parameters as input of a neural network; the electromagnetic environment parameters include: three-dimensional position information of a legitimate user's transmitting end, three-dimensional position information of a legitimate user's receiving end, an estimated activity range of an eavesdropper, channel distribution information of a link between a legitimate user's transmitting end and an intelligent reflective surface, and information about a link between an intelligent reflective surface and a legitimate user. The invention discloses a method for realizing a covert communication scheme of the UAV and a smart reflective surface. The method comprises the following steps: a) determining the channel distribution information of the link between the user receiving end and the detection threshold of the eavesdropper; b) the smart reflective surface is mounted on the UAV; c) taking the maximization of the covert communication rate of the legitimate user and the maximization of the detection error probability of the eavesdropper as the optimization goals, taking the transmission power, the lossless reflection of the smart reflective surface and the flight altitude of the UAV as the constraints, designing a multi-objective fitness function, and adopting an evolutionary algorithm to optimize the neural network to obtain the optimal neural network and the optimal covert communication scheme output by the optimal neural network; c) controlling the movement of the UAV and the smart reflective surface in the communication area based on the optimal covert communication scheme; and d) controlling the movement of the UAV and the smart reflective surface in the communication area. The optimal covert communication scheme includes the position of the UAV and the reflection coefficient matrix of the smart reflective surface.

[0101] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented when the processor executes the covert communication method combining the drone and the intelligent reflective surface provided in the above embodiments, the method comprising: extracting electromagnetic environment parameters as input of a neural network; the electromagnetic environment parameters comprising: three-dimensional position information of a legitimate user transmitting end, three-dimensional position information of a legitimate user receiving end, an estimated activity range of an eavesdropper, channel distribution information of a link between a legitimate user transmitting end and an intelligent reflective surface, channel distribution information of a link between an intelligent reflective surface and a legitimate user receiving end, and a detection threshold of an eavesdropper; wherein the intelligent reflective surface is mounted on a drone; taking maximization of the covert communication rate of the legitimate user and maximization of the detection error probability of the eavesdropper as optimization goals, taking transmission power, lossless reflection of the intelligent reflective surface, and flight altitude of the drone as constraints, designing a multi-objective fitness function, and optimizing the neural network using an evolutionary algorithm to obtain an optimal neural network and an optimal covert communication scheme output by the optimal neural network; based on the optimal covert communication scheme, controlling the actions of the drone and the intelligent reflective surface in the communication area; wherein the optimal covert communication scheme comprises the position of the drone and the reflection coefficient matrix of the intelligent reflective surface.

[0102] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0103] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A covert communication method combining a drone and an intelligent reflective surface, characterized in that: include: Extract electromagnetic environment parameters as input of neural network; The electromagnetic environment parameters include: three-dimensional position information of the legitimate user's transmitting end, three-dimensional position information of the legitimate user's receiving end, estimated activity range of the eavesdropper, channel distribution information of the link between the legitimate user's transmitting end and the intelligent reflecting surface, channel distribution information of the link between the intelligent reflecting surface and the legitimate user's receiving end, and detection threshold of the eavesdropper; wherein the intelligent reflecting surface is mounted on the drone; Taking the maximization of the covert communication rate of legitimate users and the maximization of the detection error probability of eavesdroppers as the optimization goals, and the transmission power, lossless reflection of the intelligent reflective surface and the flight altitude of the UAV as the constraints, a multi-objective fitness function is designed, and the evolutionary algorithm is used to optimize the neural network to obtain the optimal neural network and the optimal covert communication scheme output by the optimal neural network; Based on the optimal covert communication scheme, the movement of the UAV and the intelligent reflective surface in the communication area is controlled; wherein the optimal covert communication scheme includes the position of the UAV and the reflection coefficient matrix of the intelligent reflective surface; The neural network is a three-layer neural network. When optimizing the neural network, the number of hidden layer nodes and corresponding weight parameters of the neural network are optimized; The multi-objective fitness function is specifically: Among them, F1 represents the fitness value of the covert communication rate of the legitimate user, F2 represents the fitness value of the detection error probability of the eavesdropper, R is the covert communication rate, λ is the detection error probability of the eavesdropper, and R max is the maximum value of the covert communication rate, λ max is the maximum error probability, a and b are weighting coefficients; The evolutionary algorithm is used to optimize the neural network with fixed input nodes and output nodes, including the chromosome encoding operation of the neural network: A set of chromosomes represents a topological structure, each chromosome represents a connection in a network, and each chromosome has four genes: gene sequence number, connection vector, enable flag, and weight parameter; Among them, the gene sequence number is used to mark the connection vector; the connection vector is used to mark the input node and the output node; the enable mark is used to mark the enable state of the connection; the weight parameter is the weight parameter of the connection; After chromosome encoding of neural network, evolutionary algorithm also includes: The chromosomes generated by the code are mutated, crossover, population divided, elite strategy implemented, and offspring inheritance performed to optimize the neural network.

2. The covert communication method combining a drone and an intelligent reflective surface according to claim 1, characterized in that: The estimated activity range of the eavesdropper is estimated based on the distribution rule of the position of the eavesdropper; the position where the eavesdropper appears obeys Gaussian distribution.

3. A concealed communication system combining an unmanned aerial vehicle and an intelligent reflective surface, characterized in that: include: Environmental parameter acquisition module, used to extract electromagnetic environment parameters as input of neural network; The electromagnetic environment parameters include: three-dimensional position information of the legitimate user's transmitting end, three-dimensional position information of the legitimate user's receiving end, estimated activity range of the eavesdropper, channel distribution information of the link between the legitimate user's transmitting end and the intelligent reflecting surface, channel distribution information of the link between the intelligent reflecting surface and the legitimate user's receiving end, and detection threshold of the eavesdropper; wherein the intelligent reflecting surface is mounted on the drone; The network construction and training module is used to maximize the covert communication rate of legitimate users and the detection error probability of eavesdroppers as optimization goals, and to design a multi-objective fitness function with the transmission power, lossless reflection of the intelligent reflective surface, and the flight altitude of the UAV as constraints, and to optimize the neural network using an evolutionary algorithm to obtain the optimal neural network and the optimal covert communication scheme output by the optimal neural network; A scheme execution module, used to control the movement of the UAV and the intelligent reflective surface in the communication area based on the optimal covert communication scheme; wherein the optimal covert communication scheme includes the position of the UAV and the reflection coefficient matrix of the intelligent reflective surface; The neural network is a three-layer neural network. When optimizing the neural network, the number of hidden layer nodes and corresponding weight parameters of the neural network are optimized; The multi-objective fitness function is specifically: Among them, F1 represents the fitness value of the covert communication rate of the legitimate user, F2 represents the fitness value of the detection error probability of the eavesdropper, R is the covert communication rate, λ is the detection error probability of the eavesdropper, and R max is the maximum value of the covert communication rate, λ max is the maximum error probability, a and b are weighting coefficients; The evolutionary algorithm is used to optimize the neural network with fixed input nodes and output nodes, including the chromosome encoding operation of the neural network: A set of chromosomes represents a topological structure, each chromosome represents a connection in a network, and each chromosome has four genes: gene sequence number, connection vector, enable flag, and weight parameter; Among them, the gene sequence number is used to mark the connection vector; the connection vector is used to mark the input node and the output node; the enable mark is used to mark the enable state of the connection; the weight parameter is the weight parameter of the connection; After chromosome encoding of neural network, evolutionary algorithm also includes: The chromosomes generated by the code are mutated, crossover, population divided, elite strategy implemented, and offspring inheritance performed to optimize the neural network.

4. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the covert communication method combining the drone and the intelligent reflective surface as described in any one of claims 1 to 2 are implemented.

5. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the covert communication method combining a drone and an intelligent reflective surface as described in any one of claims 1 to 2 are implemented.

6. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the covert communication method combining a drone and an intelligent reflective surface as described in any one of claims 1 to 2 are implemented.

Images