Method and system for assisting anti-interference communication of unmanned aerial vehicle based on intelligent reflecting surface

Abstract

The invention belongs to the technical field of unmanned aerial vehicle communication. The method comprises the following steps: acquiring a user signal and a malicious interference signal, and constructing a non-convex objective function and a non-convex constraint condition; converting the non-convex objective function into a convex objective function and a convex constraint condition by using a continuous convex approximation algorithm and the non-convex constraint condition; processing the convex objective function and the convex constraint condition by utilizing a fractional programming algorithm to obtain optimized transmitting power distribution, an intelligent reflecting surface reflection coefficient and an unmanned aerial vehicle track; performing iterative optimization processing on the optimized transmitting power distribution, the reflection coefficient of the intelligent reflecting surface and the track of the unmanned aerial vehicle by using an S-process alternating optimization algorithm to obtain converged target transmitting power distribution, a target intelligent reflecting surface reflection coefficient and a target track of the unmanned aerial vehicle; and obtaining a suboptimal solution for maximizing the energy efficiency of the unmanned aerial vehicle by using a convergence mode.

Description

Technical Field

[0001] This application belongs to the technical field of UAV communication, and particularly relates to a method and system for anti-jamming communication of UAVs assisted by intelligent reflecting surfaces. Background Art

[0002] With the rapid development of Unmanned Aerial Vehicle (UAV) technology, UAVs are increasingly widely used in fields such as communication, reconnaissance, and logistics. However, UAV communication systems face many challenges in practical applications. One of the most prominent problems is limited energy efficiency and vulnerability to malicious interference attacks. Due to the limited energy reserve of UAVs, how to reduce energy consumption while ensuring communication quality has become a key issue in the design of UAV communication systems. In addition, when UAVs fly in a complex electromagnetic environment, they are easily attacked by malicious interference signals, resulting in a decline or even interruption of communication performance, seriously affecting the reliability and safety of UAVs.

[0003] Traditional UAV communication anti-jamming technologies mainly rely on increasing the transmission power or using complex signal processing algorithms to suppress interference. However, these methods often lead to an increase in energy consumption, further exacerbating the problem of limited energy of UAVs. In addition, traditional methods have limited anti-jamming effects when facing dynamic interference and inaccurate interference positions, and it is difficult to ensure the stability and efficiency of communication in a complex environment.

[0004] In view of this, there is an urgent need for a method and system for anti-jamming communication of UAVs assisted by intelligent reflecting surfaces to solve the above technical problems. Summary of the Invention

[0005] Based on this, it is necessary to provide a method and system for anti-jamming communication of UAVs assisted by intelligent reflecting surfaces for the above technical problems.

[0006] In a first aspect, this application provides a method for anti-jamming communication of UAVs assisted by intelligent reflecting surfaces, including:

[0007] Obtain user signals and malicious interference signals, and construct a non-convex objective function and non-convex constraint conditions;

[0008] Use the successive convex approximation algorithm and the non-convex constraint conditions to convert the non-convex objective function into a convex objective function and convex constraint conditions;

[0009] Use the fractional programming algorithm to process the convex objective function and the convex constraint conditions to obtain optimized transmission power allocation, intelligent reflecting surface reflection coefficients, and UAV trajectories;

[0010] An alternating optimization algorithm using the S-procedure is employed to iteratively optimize the optimized transmit power allocation, intelligent reflecting surface reflection coefficient, and UAV trajectory, yielding a converged target transmit power allocation, target intelligent reflecting surface reflection coefficient, and target UAV trajectory.

[0011] Using the converged target transmit power allocation, target intelligent reflecting surface reflection coefficient, and target UAV trajectory, a suboptimal solution for maximizing the energy efficiency of the UAV is obtained.

[0012] In some implementable ways, the obtaining of the user signal and malicious interference signal, and the construction of the non-convex objective function and non-convex constraint conditions include the following steps:

[0013] Construct a signal model;

[0014] Using the signal model, process the user signal and the malicious interference signal to obtain the channel parameters of the user signal and the malicious interference signal;

[0015] Using the signal model and the channel parameters of the user signal and the malicious interference signal, construct a non-convex objective function;

[0016] Construct non-convex constraint conditions, where the non-convex constraint conditions include transmit power allocation, intelligent reflecting surface reflection coefficient, and UAV trajectory.

[0017] In some implementable ways, the step of converting the non-convex objective function into a convex objective function and convex constraint conditions by using the sequential convex approximation algorithm and the non-convex constraint conditions includes:

[0018] Using auxiliary variables and the non-convex constraint conditions, perform convex approximation processing on the non-convex terms in the non-convex objective function to obtain the convex objective function;

[0019] Introduce slack variables and use the sequential convex approximation technique to perform first-order Taylor expansion on some of the non-convex constraint conditions to obtain the convex constraint conditions.

[0020] In some implementable ways, the step of using auxiliary variables and the non-convex constraint conditions to perform convex approximation processing on the non-convex terms in the non-convex objective function to obtain the convex objective function includes:

[0021] Using the auxiliary variables, perform convex approximation processing on the non-convex terms in the non-convex objective function to obtain a solvable convex optimization form result;

[0022] Using the semidefinite relaxation method, convert the rank constraint into a semidefinite constraint, making the non-convex problem into a convex optimization problem and solving it through a convex optimization algorithm to obtain the solution result;

[0023] Using the Gaussian randomization or eigenvalue decomposition method, calculate the solution result to obtain a convex objective function.

[0024] In some implementable ways, the step of introducing slack variables and using the continuous convex approximation technique to perform first-order Taylor expansion on some of the non-convex constraint conditions to obtain the convex constraint conditions includes:

[0025] Introduce slack variables to linearly constrain the non-convex constraint conditions to obtain non-convex constraint conditions with linear constraints;

[0026] Using the continuous convex approximation technique, perform first-order Taylor expansion on some of the non-convex constraint conditions with linear constraints to obtain convex constraint conditions.

[0027] In some implementable ways, the step of using the fractional programming algorithm to process the convex objective function and the convex constraint conditions to obtain the optimized transmit power allocation, intelligent reflecting surface reflection coefficient, and UAV trajectory includes:

[0028] Obtain the initialized convex constraint conditions;

[0029] Substitute the convex objective function and the initialized convex constraint conditions into the fractional programming algorithm respectively for calculation to obtain the iteratively optimized transmit power allocation, intelligent reflecting surface reflection coefficient, and UAV trajectory;

[0030] Update the iteratively optimized transmit power allocation, intelligent reflecting surface reflection coefficient, and UAV trajectory until convergence to obtain the optimized transmit power allocation, intelligent reflecting surface reflection coefficient, and UAV trajectory.

[0031] In some implementable ways, the step of using the alternating optimization algorithm of S-procedure to perform iterative optimization processing on the optimized transmit power allocation, intelligent reflecting surface reflection coefficient, and UAV trajectory to obtain the converged target transmit power allocation, target intelligent reflecting surface reflection coefficient, and target UAV trajectory includes:

[0032] Using the alternating optimization algorithm of S-procedure, optimize each variable in the iterative optimization processing of the optimized transmit power allocation, intelligent reflecting surface reflection coefficient, and UAV trajectory to obtain the converged target transmit power allocation, target intelligent reflecting surface reflection coefficient, and target UAV trajectory.

[0033] In some implementable ways, the step of using the converged target transmit power allocation, target intelligent reflecting surface reflection coefficient, and target UAV trajectory to obtain a suboptimal solution for maximizing the UAV energy efficiency includes:

[0034] Calculate the energy efficiency index using the converged target transmit power allocation, target intelligent reflecting surface reflection coefficient, and target UAV trajectory to obtain the index calculation result;

[0035] According to the index calculation result, determine whether the current optimization result is a sub-optimal solution for maximizing energy efficiency;

[0036] If the energy efficiency index meets a preset threshold or optimization goal, obtain the current sub-optimal solution for maximizing energy efficiency.

[0037] In a second aspect, a smart reflecting surface-assisted UAV anti-jamming communication system is provided, which is applied to the aforementioned smart reflecting surface-assisted UAV anti-jamming communication method. The system includes:

[0038] An acquisition unit for acquiring user signals and malicious interference signals and constructing a non-convex objective function and non-convex constraint conditions;

[0039] A first processing unit for converting the non-convex objective function into a convex objective function and convex constraint conditions by using the sequential convex approximation algorithm and the non-convex constraint conditions;

[0040] A second processing unit for processing the convex objective function and the convex constraint conditions by using the fractional programming algorithm to obtain the optimized transmit power allocation, intelligent reflecting surface reflection coefficient, and UAV trajectory;

[0041] A third processing unit for performing iterative optimization processing on the optimized transmit power allocation, intelligent reflecting surface reflection coefficient, and UAV trajectory by using the alternating optimization algorithm of S-procedure to obtain the converged target transmit power allocation, target intelligent reflecting surface reflection coefficient, and target UAV trajectory;

[0042] A result unit for obtaining the sub-optimal solution for maximizing the energy efficiency of the UAV by using the converged target transmit power allocation, target intelligent reflecting surface reflection coefficient, and target UAV trajectory

[0043] In a third aspect, the present application provides a computer program, which when executed by a processor implements the steps of the aforementioned smart reflecting surface-assisted UAV anti-jamming communication method.

[0044] Beneficial effects: The present application provides a method for intelligent reflecting surface-assisted anti-jamming communication of unmanned aerial vehicles (UAVs). The method includes: acquiring user signals and malicious jamming signals, and constructing a non-convex objective function and non-convex constraint conditions; using the sequential convex approximation algorithm and the non-convex constraint conditions to convert the non-convex objective function into a convex objective function and convex constraint conditions; using the fractional programming algorithm to process the convex objective function and the convex constraint conditions to obtain optimized transmit power allocation, intelligent reflecting surface reflection coefficients, and UAV trajectories; using the alternating optimization algorithm of S-procedure to perform iterative optimization processing on the optimized transmit power allocation, intelligent reflecting surface reflection coefficients, and UAV trajectories to obtain convergent target transmit power allocation, target intelligent reflecting surface reflection coefficients, and target UAV trajectories; using the convergent target transmit power allocation, target intelligent reflecting surface reflection coefficients, and target UAV trajectories to obtain a sub-optimal solution for maximizing the energy efficiency of the UAV. Through the above method, it is possible to continuously iteratively optimize the transmit power allocation, intelligent reflecting surface reflection coefficients, and UAV trajectories by using the sequential convex approximation, fractional programming, and alternating optimization algorithm of S-procedure to continuously receive the energy efficiency of the UAV until convergence, realizing anti-jamming of the intelligent reflecting surface-assisted UAV under the condition of inaccurate interference positions. Description of the Drawings

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present application or in the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0046] Figure 1 It is a flowchart of a method for intelligent reflecting surface-assisted anti-jamming communication of unmanned aerial vehicles in an embodiment. Detailed Embodiments

[0047] To facilitate the understanding of the present application, the present application will be described more comprehensively below with reference to the relevant drawings. Embodiments of the present application are shown in the drawings. However, the present application can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present application more thorough and comprehensive.

[0048] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present application belongs. The terms used in the description of the present application in this specification are only for the purpose of describing specific embodiments and are not intended to limit the present application. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0049] It is understood that the terms "first", "second", etc. used in the present application may be used herein to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from another element.

[0050] Some nouns involved in the present application are explained below for better understanding of the present application:

[0051] The Fractional Programming Algorithm is an algorithm for solving optimization problems with a fractional objective function. Its core idea is to transform the fractional programming problem into a form that is easier to solve, so as to find the optimal solution. Fractional programming refers to an optimization problem where the objective function is the ratio of two functions.

[0052] The S-procedure is a commonly used mathematical tool in the fields of control theory and optimization, mainly used to handle constraint conditions involving quadratic inequalities. The alternating optimization algorithm is an iterative optimization method that gradually optimizes the objective function by alternately updating multiple variables. The alternating optimization algorithm combining the two, namely the S-procedure, is an algorithm that uses the S-procedure to handle constraint conditions and solves complex optimization problems through the alternating optimization method.

[0053] The continuous convex approximation algorithm is an iterative method for solving non-convex optimization problems. It gradually approximates the non-convex problem into a series of convex optimization problems and solves these convex problems in each iteration, so as to approximate the optimal solution of the original non-convex problem.

[0054] UAV, that is, Unmanned Aerial Vehicle, is an aircraft that can fly without an on-board pilot. UAVs can fly through remote control or pre-set programs and are widely used in military, civilian, and commercial fields.

[0055] CVX is a high-level modeling language and solver for convex optimization problems. It allows users to express and solve convex optimization problems in a natural way. CVX is a free software package based on MATLAB.

[0056] In a first aspect, the present application provides an intelligent reflecting surface-assisted UAV anti-jamming communication method, and the method includes:

[0057] S100, obtain user signals and malicious interference signals, and construct a non-convex objective function and non-convex constraint conditions.

[0058] Among them, constructing the non-convex objective function and non-convex constraint conditions may include the following steps:

[0059] S101, construct a signal model.

[0060] Specifically, a propagation model of user signals and malicious interference signals in the UAV communication system is established.

[0061] Exemplarily, constructing a signal model is the basis for the analysis of the entire system. The signal model is used to describe the characteristics of user signals and malicious interference signals, and may include the generation, propagation, and reception processes of user signals and malicious interference signals.

[0062] Specifically, the user signal can be a communication signal transmitted from the ground station to the UAV, and its mathematical representation can be a complex signal, including information such as amplitude, frequency, and phase:

[0063]

[0064] where s u (t) represents the user communication signal, A u represents the amplitude of the user signal, f u represents the frequency of the user signal, represents the phase of the user signal, e represents the base of the natural logarithm, j represents the imaginary unit, 2πf u t represents the frequency part of the signal, and t represents time.

[0065] The malicious interference signal is usually transmitted by an attacker, aiming to interfere with the communication signal between the ground station and the UAV and affect the communication quality. Its expression is similar to that of the user signal:

[0066]

[0067] where s d (t) represents the interference signal, A d represents the amplitude of the interference signal, f d represents the frequency of the interference signal, represents the phase of the interference signal, e represents the base of the natural logarithm, j represents the imaginary unit, 2πf d t represents the frequency part of the signal, and t represents time.

[0068] Channel model. During the propagation process, the signal will be affected by path loss, multipath effect, and noise. The channel model can quantify these effects. For wireless communication, the free space path loss model or the Rayleigh fading model can be used to describe it.

[0069] The received signal model of the user signal is:

[0070] r u (k) = h u ·s u (t) + n u (t);

[0071] Among them, r u (t) represents the instantaneous power of the user signal h u represents the channel gain of the user signal, for example, the channel gain determined by environmental factors; n u (t) represents the noise term, for example, Gaussian noise; s u (t) represents the user communication signal.

[0072] The received signal model of the interference signal is:

[0073] r d (t) = h d ·s d (t) + n d (t)

[0074] Among them, r d (t), represents the instantaneous power of the interference signal, h d represents the channel gain of the user signal, for example, the channel gain determined by environmental factors; n d (t) represents the noise term, for example, Gaussian noise; s d (t) represents the interference signal.

[0075] The above method constructs a signal model describing the user signal and the malicious interference signal, and defines influencing factors such as channel gain and noise in signal propagation.

[0076] S102, using the signal model, process the user signal and the malicious interference signal to obtain the channel parameters of the user signal and the malicious interference signal.

[0077] Through the signal model, the key parameters of the wireless channel can be further extracted, which may include channel gain, path loss, noise power, etc. These channel parameters are the basis for communication performance analysis and optimization.

[0078] Channel gain: The channel gains h u and h d reflect the attenuation of the signal during propagation due to factors such as distance, obstacles, and weather. For the free space path loss model, the channel gain can be calculated by the following formula:

[0079]

[0080] Among them, d u and d d are the distances between the user and the receiving end, and the interference source and the receiving end respectively, and α represents the path loss exponent, for example, taking values between 2 and 4.

[0081] Noise power: The noise power can be known or obtained through system testing. For example, it is expressed as the white noise power spectral density σ2 , Gaussian noise is commonly used to approximate in wireless communication:

[0082] σ 2 (noise power)

[0083] Interference power: The power P of the interference signal d is determined jointly by the signal amplitude A of the interference source d and the channel gain h d and can be calculated by the following formula:

[0084]

[0085] where, A d represents the amplitude of the interference signal, and h d represents the channel gain of the interference signal.

[0086] Through the processing of the signal model, channel parameters are obtained. The channel parameters can include channel gain, noise power, and the power of the interference signal, and these parameters provide a basis for the construction of the objective function and constraints.

[0087] S103, construct a non-convex objective function using the signal model and the channel parameters of the user signal and the malicious interference signal.

[0088] Specifically, construct a non-convex objective function using the signal model and the channel parameters of the user signal and the malicious interference signal, especially in the scenario of interference suppression, to maximize the SINR (signal-to-interference-plus-noise ratio).

[0089] The definition of SINR is:

[0090]

[0091] where, P u represents the power of the user signal; P d represents the power of the interference signal; σ 2 represents the noise power

[0092] The objective function is expressed as:

[0093]

[0094] where, N represents the total number of users or signal links; the objective function is non-convex because the logarithmic function and the fractional form of SINR result in non-convexity; SINR i represents the signal-to-interference-plus-noise ratio of the i-th signal link, which measures the quality of the i-th signal at the receiving end.

[0095] By using channel parameters, a non-convex objective function is constructed, which aims to maximize the SINR of the system and thus improve the performance of the communication system.

[0096] S104, construct non-convex constraint conditions.

[0097] Among them, the constraint conditions include transmit power allocation, intelligent reflecting surface reflection coefficient, UAV trajectory, and auxiliary variables.

[0098] Specifically, the constraint conditions are used to ensure that the optimization process of the system is carried out within a practical and feasible range and follows technical and physical limitations. These constraint conditions include transmit power allocation, intelligent reflecting surface reflection coefficient, and UAV trajectory.

[0099] Transmit power allocation: The transmit power of the ground node needs to meet the total power upper limit, and the power allocation needs to be dynamically optimized to balance communication quality and energy consumption.

[0100] Intelligent reflecting surface reflection coefficient: The phase parameter of the reflecting unit needs to meet the unit modulus constraint to ensure that only the phase is adjusted without changing the signal amplitude.

[0101] UAV trajectory: The flight path needs to comply with dynamic constraints (such as speed, acceleration) and airspace limitations, and at the same time avoid collisions with obstacles.

[0102] It should be noted that for the quantization of the constraint conditions, the limit value of the device or the set limit value can be used as the quantization standard. For example, the transmit power allocation needs to be less than or equal to the total power upper limit that the transmit power of the ground node needs to meet, and |v[n]| = 1 in the intelligent reflecting surface reflection coefficient, where |v[n]| is the phase coefficient of the nth reflecting unit, etc.

[0103] S200, use the sequential convex approximation algorithm and the non-convex constraint conditions to convert the non-convex objective function into a convex objective function and convex constraint conditions.

[0104] Specifically, the conversion into a convex objective function and convex constraint conditions may include the following steps:

[0105] S201, use auxiliary variables and the non-convex constraint conditions to perform convex approximation processing on the non-convex terms in the non-convex objective function to obtain the convex objective function.

[0106] Specifically, by introducing auxiliary variables, the non-convex part in the non-convex objective function can be transformed into a convex objective function. For example, a non-convex quadratic term can be transformed into a linear or quadratic convex function by introducing new variables (auxiliary variables). In this way, the originally difficult-to-optimize non-convex objective function is transformed into an easily-optimized convex objective function.

[0107] It should be noted that obtaining the convex objective function may include the following steps:

[0108] S2011. Use the auxiliary variable to perform convex approximation processing on the non-convex terms in the non-convex objective function to obtain a solvable convex optimization form result.

[0109] S2012. Use the semidefinite relaxation method to transform the rank constraint into a semidefinite constraint, making the non-convex problem become a convex optimization problem, and solve it through a convex optimization algorithm to obtain a solution result.

[0110] S2013. Use the Gaussian randomization or eigenvalue decomposition method to calculate the solution result to obtain the convex objective function.

[0111] Exemplarily, in the non-convex objective function, some parts may be non-linear or contain non-convex terms such as logarithms, exponentials, fractions, etc. To transform these non-convex terms into a convex objective function, the method of introducing auxiliary variables can be adopted.

[0112] If there is a non-convex quadratic term x 2 in the objective function, it is not easy to optimize directly. An auxiliary variable z can be introduced to transform this quadratic term into a linear constraint:

[0113] z = x 2 ;

[0114] At this time, the objective function becomes z, and z is convex (a linear function). In this way, the original non-convex objective function is transformed into a convex function, and x 2 is simplified to an optimization problem of the auxiliary variable.

[0115] Through this method, the complex non-convex objective function is transformed into a convex objective function that is easier to optimize.

[0116] In the UAV communication system, the goal is to maximize the sum of log(1 + SINR i ) of all signal links. Introduce an auxiliary variable z i , such that z i ≥ log(1 + SINR i ), and the objective function becomes ∑z i , which is a convex objective function.

[0117] The optimization problem may include a constraint that the rank of matrix V is 1, i.e., rank(V) = 1, which is non-convex. Through semidefinite relaxation, ignoring the rank constraint and treating V as a positive semidefinite matrix, the problem becomes a semidefinite programming problem, which is convex. Use a convex optimization tool (such as CVX) to solve the semidefinite programming problem to obtain the solution V * .

[0118] If V* has a rank of 1, and the original vector u can be directly obtained through eigenvalue decomposition, that is, V * = uu H , where u represents a complex vector, the phase coefficient of the intelligent reflecting surface, and u H represents the conjugate transpose of the vector u, and V * represents the optimal positive semi - definite matrix obtained through semidefinite relaxation.

[0119] If the rank of V * is greater than 1, an approximate u - vector is extracted from V * using the Gaussian randomization method.

[0120] The finally obtained u is used to construct the phase coefficient of the intelligent reflecting surface:

[0121]

[0122] where represents the phase coefficient of the i - th intelligent reflecting unit at time slot n, and e jθ is the exponential form of a complex number, representing a complex number with a phase of θ, j is the imaginary unit, and ∠ extracts the phase (angle) of its argument, is the phase ratio of the i - th and the reference unit at time slot n, and n indicates that this formula is applicable to all time slots from 1 to M.

[0123] S202. Introduce slack variables and use the continuous convex approximation technique to perform a first - order Taylor expansion on some of the non - convex constraint conditions to obtain the convex constraint conditions.

[0124] Specifically, slack variables are used to relax some non - convex constraints, allowing the problem to become easier to handle. Using the continuous convex approximation technique, especially the first - order Taylor expansion, some non - convex constraints can be linearized, further transforming the non - convex constraints into convex constraints.

[0125] It should be noted that obtaining the convex constraint conditions may include the following steps:

[0126] S2021. Introduce slack variables and perform linear constraints on the non - convex constraint conditions to obtain non - convex constraint conditions with linear constraints.

[0127] S2022. Use the continuous convex approximation technique to perform a first - order Taylor expansion on some of the non - convex constraint conditions with linear constraints to obtain the convex constraint conditions.

[0128] In some cases, some constraint conditions are non - convex, such as those with product terms or logarithmic terms. To handle these non - convex constraint conditions, the problem can be simplified by introducing slack variables and the continuous convex approximation technique.

[0129] Slack variables can be used to relax the constraints, transforming the originally strict non-convex constraints into a relaxed form that is easier to optimize. Then, first-order Taylor expansion can be used to approximate non-convex functions and linearize some complex non-convex constraints.

[0130] Exemplarily, a certain non-convex constraint is:

[0131] log(x) ≤ b;

[0132] This constraint is non-convex and can be linearized using Taylor expansion. Performing a first-order Taylor expansion at the x 0 point gives:

[0133]

[0134] In this way, the non-convex constraint log(x) ≤ b is linearized to:

[0135]

[0136] This constraint becomes linear and is thus transformed into a convex constraint, which is easier to handle.

[0137] By introducing auxiliary variables to perform convex approximation on non-convex terms in the objective function and using slack variables and first-order Taylor expansion to linearize non-convex constraints, the originally complex non-convex optimization problem can be transformed into a convex optimization problem that is easier to optimize, thereby improving the efficiency and accuracy of optimization.

[0138] Exemplarily, in an optimization problem, the goal is to maximize the SINR (Signal-to-Interference-plus-Noise Ratio) in a drone communication system, and the system contains a non-convex objective function and constraints:

[0139]

[0140] where, Objective represents the non-convex objective function, P u,i represents the power of the i-th user signal, P d,i represents the power of the interference signal, and σ 2 is the noise power.

[0141] If the SINR expression in the objective function is non-convex, therefore, auxiliary variables can be introduced for convex approximation.

[0142] Introduce auxiliary variables:

[0143] Introduce the auxiliary variable z i to replace the non-convex term log(1 + SINR i ), making the objective function become:

[0144]

[0145] This transforms the objective function into a convex objective function that includes an auxiliary variable z i .

[0146] Introduce slack variables and linearize the constraints:

[0147] Set a non-convex constraint, such as log(x) ≤ b, and linearize it using Taylor expansion:

[0148]

[0149] Convert the non-convex constraint into a linear form, making the optimization problem convex, so that it can be solved by standard convex optimization algorithms.

[0150] Maximize the sum of logarithms of all SINRs while satisfying the linearized constraint conditions. Through these steps, the original non-convex optimization problem is transformed into a feasible convex optimization problem, which can be effectively solved to obtain the optimal solution.

[0151] S300, using the fractional programming algorithm, process the convex objective function and the convex constraint conditions to obtain the optimized transmit power allocation, intelligent reflecting surface reflection coefficient, UAV trajectory, and auxiliary variable.

[0152] Specifically, obtaining the optimized transmit power allocation, intelligent reflecting surface reflection coefficient, UAV trajectory, and auxiliary variable may include the following steps:

[0153] S301, obtain the initialized convex constraint conditions.

[0154] Specifically, obtain the initialized transmit power allocation P (0) , intelligent reflecting surface reflection coefficient v (0) , and UAV trajectory Q (0) . Determine the initial state of the optimization problem, including the convex objective function and the convex constraint conditions.

[0155] S302, substitute the convex objective function and the initialized convex constraint conditions into the fractional programming algorithm respectively for calculation to obtain the iteratively optimized transmit power allocation, intelligent reflecting surface reflection coefficient, and UAV trajectory.

[0156] Specifically, the fractional programming algorithm is used to process optimization problems where the convex objective function contains a fractional form. Substitute the convex objective function and the convex constraint conditions into the fractional programming algorithm for optimization processing. Through the fractional programming algorithm, gradually iteratively optimize the transmit power allocation, intelligent reflecting surface reflection coefficient, and UAV trajectory.

[0157] In each iteration step, update the optimization variables until the convergence condition is met.

[0158] S303. Update and iterate the optimized transmit power allocation, intelligent reflecting surface reflection coefficient, and UAV trajectory until convergence to obtain the optimized transmit power allocation, intelligent reflecting surface reflection coefficient, and UAV trajectory.

[0159] Specifically, according to the optimization result, update the transmit power allocation P (k+1) , ensuring that the updated transmit power allocation meets the requirements of the total power limit and dynamic optimization. Update the intelligent reflecting surface reflection coefficient v (k+1) , ensuring that the updated reflection coefficient meets the unit modulus constraint. Update the UAV trajectory Q (k+1) , ensuring that the updated trajectory complies with the dynamic constraints and airspace limitations. Check whether the optimization variables converge, that is, check P (k+1) , v (k+1) , Q (k+1) , whether it is close enough to the result of the previous iteration. If it converges, stop the iteration; otherwise, return to the second step to continue the optimization, where k represents the number of iterations.

[0160] S400. Use the alternating optimization algorithm of S-procedure to perform iterative optimization processing on the optimized transmit power allocation, intelligent reflecting surface reflection coefficient, and UAV trajectory to obtain the converged target transmit power allocation, target intelligent reflecting surface reflection coefficient, and target UAV trajectory.

[0161] Specifically, use the alternating optimization algorithm of S-procedure to optimize each variable in the iterative optimization processing of the optimized transmit power allocation, intelligent reflecting surface reflection coefficient, and UAV trajectory to obtain the converged target transmit power allocation, target intelligent reflecting surface reflection coefficient, and target UAV trajectory.

[0162] Exemplarily, in the foregoing steps, the optimized transmit power allocation, intelligent reflecting surface reflection coefficient, and UAV trajectory have been obtained.

[0163] Fix some variables and optimize other variables:

[0164] Optimize the transmit power allocation: Fix the intelligent reflecting surface reflection coefficient v (k) and the UAV trajectory Q (k) , and optimize the transmit power allocation P (k+1) .

[0165] Use a convex optimization algorithm (such as CVX) to solve the optimization problem to obtain P (k+1) .

[0166] Optimize the intelligent reflecting surface reflection coefficient:

[0167] Fix the transmit power allocation P (k+1) and the UAV trajectory Q (k), optimize the reflection coefficient v of the intelligent reflecting surface (k+1) .

[0168] Use the convex optimization algorithm to solve the optimization problem and obtain v (k+1) .

[0169] Optimize the UAV trajectory:

[0170] Fix the transmit power allocation P (k+1) and the reflection coefficient v of the intelligent reflecting surface (k+1) , and optimize the UAV trajectory Q (k+1) .

[0171] Use the convex optimization algorithm to solve the optimization problem and obtain Q (k+1) .

[0172] Update the optimization variables:

[0173] Take the optimized variables P (k+1) , v (k+1) , Q (k+1) as the input for the next iteration.

[0174] Calculate the difference between the variables in the current iteration and those in the previous iteration:

[0175] ||P (k+1) - P (k) || ≤ ∈P;

[0176] ||v (k+1) - v (k) || ≤ ∈v;

[0177] ||Q (k+1) - Q (k) || ≤ ∈Q;

[0178] If the change in all variables is less than the preset thresholds ∈P, ∈v, ∈Q, it is considered that the algorithm has converged. If not, take P (k+1) , v (k+1) , Q (k+1) as the new initial values, return to continue the optimization until convergence, and obtain the final optimization result.

[0179] S500, using the converged target transmit power allocation, target intelligent reflecting surface reflection coefficient, target UAV trajectory, and target auxiliary variables, obtain the suboptimal solution for maximizing the UAV energy efficiency.

[0180] Specifically, the steps to obtain the suboptimal solution for maximizing the UAV energy efficiency may include the following:

[0181] S501, using the converged target transmit power allocation, target intelligent reflecting surface reflection coefficient, and target UAV trajectory, calculate the energy efficiency index to obtain the index calculation result.

[0182] Specifically, using the converged target transmit power allocation P * , the target intelligent reflecting surface reflection coefficient v * and the target UAV trajectory Q * , the system throughput is calculated.

[0183] The system throughput can be measured by the signal-to-interference-plus-noise ratio (SINR), and the specific calculation formula is:

[0184]

[0185] where P * represents the converged transmit power allocation, h i represents the channel gain, and σ 2 represents the noise power.

[0186] Calculate the total energy consumption, including the transmit power and the flight energy consumption.

[0187] Transmit power energy consumption:

[0188]

[0189] The flight energy consumption can be calculated according to the UAV flight trajectory Q * and the flight time. The specific formula is:

[0190]

[0191] where m is the mass of the UAV, u(t) is the speed of the UAV, a(t) is the acceleration of the UAV, k is the flight energy consumption coefficient, T is the flight time, and dt represents the infinitesimal increment of time t for integral operation.

[0192] Calculate the energy efficiency index:

[0193] The energy efficiency index is defined as the ratio of the system throughput to the total energy consumption:

[0194]

[0195] Thus, the index calculation result is obtained.

[0196] S502. According to the index calculation result, determine whether the current optimization result is a sub-optimal solution for maximizing the energy efficiency.

[0197] Compare the calculated energy efficiency index with a preset threshold or optimization target. If the energy efficiency index meets the preset threshold or optimization target, it is considered that the current optimization result is a sub-optimal solution for maximizing the energy efficiency. If the energy efficiency index does not meet the preset threshold or optimization target, further optimization or parameter adjustment is required.

[0198] S503. If the energy efficiency index meets a preset threshold or optimization goal, obtain a sub-optimal solution that maximizes the current energy efficiency.

[0199] If the energy efficiency index meets the preset threshold or optimization goal, output the current optimization result, including the target transmit power allocation P * , the target intelligent reflecting surface reflection coefficient v * and the target UAV trajectory Q * , and record the current optimization result.

[0200] Embodiment

[0201] Under the conditions of the transmit power of the given ground node and the UAV trajectory, robustly design the reflection coefficient of the intelligent reflecting surface under imprecise interference. For the given UAV trajectory Q and the transmit power P of the ground node, the optimization problem can be equivalently expressed as:

[0202]

[0203] Since the objective function is relatively complex and multiple variables are coupled together, it is difficult to directly solve the above problem. Because the function is monotonically increasing, find a set of phase matrices to maximize the throughput of the objective function in each time slot, and finally maximize the average rate of the system. Use the semidefinite relaxation method, and then obtain a feasible solution through the Gaussian randomization or eigenvalue decomposition method. The convex optimization problem can be effectively solved by CVX. However, the rank-1 constraint condition cannot be guaranteed to be achieved at this time. If the rank of V[n] is 1, then v[n] can be directly obtained through eigenvalue decomposition. Otherwise, v[n] needs to be approximately obtained through Gaussian randomization. Finally, the intelligent reflecting surface phase coefficient is:

[0204]

[0205] By introducing slack variables and using the continuous convex approximation technique, perform a first-order Taylor expansion on some non-convex constraints, the problem can be approximately transformed into a convex problem, and solved by the interior point method to obtain the UAV vertical trajectory. Continuously iterate and optimize until the UAV energy efficiency converges. By repeatedly predicting the interference position and solving the sub-problem, continuously iterate and update the UAV energy efficiency, and finally obtain a sub-optimal solution that maximizes the UAV energy efficiency.

[0206] In a second aspect, there is provided an intelligent reflecting surface-assisted UAV anti-interference communication system, which is applied to the foregoing intelligent reflecting surface-assisted UAV anti-interference communication method. The system includes:

[0207] An acquisition unit, configured to acquire user signals and malicious interference signals, and construct a non-convex objective function and non-convex constraint conditions;

[0208] A first processing unit, configured to use a sequential convex approximation algorithm and the non-convex constraint conditions to convert the non-convex objective function into a convex objective function and convex constraint conditions;

[0209] A second processing unit, configured to use a fractional programming algorithm to process the convex objective function and the convex constraint conditions to obtain optimized transmit power allocation, intelligent reflecting surface reflection coefficients, and UAV trajectories;

[0210] A third processing unit, configured to use an alternating optimization algorithm of S-procedure to perform iterative optimization processing on the optimized transmit power allocation, intelligent reflecting surface reflection coefficients, and UAV trajectories to obtain convergent target transmit power allocation, target intelligent reflecting surface reflection coefficients, and target UAV trajectories;

[0211] A result unit, configured to use the convergent target transmit power allocation, target intelligent reflecting surface reflection coefficients, and target UAV trajectories to obtain a sub-optimal solution for maximizing the energy efficiency of the UAV

[0212] In a third aspect, the present application provides a computer program, which when executed by a processor implements the steps of the foregoing method for intelligent reflecting surface-assisted UAV anti-jamming communication.

[0213] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided by the present application can include non-volatile and / or volatile memories. The non-volatile memory can include a read-only memory (ROM), a programmable ROM (PROM), an electrically programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), or a flash memory. The volatile memory can include a random access memory (RAM) or an external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0214] Each embodiment in the present disclosure is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other, and the differences between each embodiment and other embodiments are emphasized.

[0215] The protection scope of the present disclosure is not limited to the above embodiments. Obviously, those skilled in the art can make various changes and modifications to the present disclosure without departing from the scope and spirit of the present disclosure. If these changes and modifications fall within the scope of the claims of the present disclosure and their equivalent technologies, the intention of the present disclosure also includes these changes and modifications.

Claims

1. A method for anti-interference communication of unmanned aerial vehicles based on intelligent reflective surface assistance, characterized in that: Methods include: Obtain user signals and malicious interference signals, and construct non-convex objective functions and non-convex constraints; Using a continuous convex approximation algorithm and the non-convex constraint condition, the non-convex objective function is converted into a convex objective function and a convex constraint condition; The convex objective function and the convex constraint condition are processed by using a fractional programming algorithm to obtain optimized transmission power allocation, intelligent reflective surface reflection coefficient and UAV trajectory; Using the alternating optimization algorithm of S-procedure, the optimized transmission power distribution, the reflection coefficient of the intelligent reflection surface and the trajectory of the UAV are iteratively optimized to obtain the converged target transmission power distribution, the target intelligent reflection surface reflection coefficient and the target UAV trajectory; By utilizing the converged target transmit power allocation, the reflection coefficient of the target intelligent reflective surface and the target UAV trajectory, a suboptimal solution for maximizing the UAV energy efficiency is obtained.

2. The method for anti-interference communication of unmanned aerial vehicles based on intelligent reflective surface assistance according to claim 1 is characterized in that: The method of obtaining user signals and malicious interference signals and constructing a non-convex objective function and non-convex constraint conditions includes the following steps: Build signal model; Using the signal model, the user signal and the malicious interference signal are processed to obtain channel parameters of the user signal and the malicious interference signal; Constructing a non-convex objective function using the signal model and the channel parameters of the user signal and the malicious interference signal; Construct non-convex constraints, wherein the non-convex constraints include transmission power allocation, smart reflective surface reflection coefficient, and UAV trajectory.

3. The method for anti-interference communication of unmanned aerial vehicles based on intelligent reflective surface assistance according to claim 1 is characterized in that: The step of converting the non-convex objective function into a convex objective function and a convex constraint condition by using a continuous convex approximation algorithm and the non-convex constraint condition comprises: Using auxiliary variables and the non-convex constraint conditions, a convex approximation process is performed on the non-convex terms in the non-convex objective function to obtain the convex objective function; Slack variables are introduced, and continuous convex approximation technology is used to perform first-order Taylor expansion on part of the non-convex constraints to obtain the convex constraints.

4. The method for anti-interference communication of unmanned aerial vehicles based on intelligent reflective surface assistance according to claim 3 is characterized in that: The step of using the auxiliary variables and the non-convex constraint conditions to perform convex approximation processing on the non-convex terms in the non-convex objective function to obtain the convex objective function includes: Using the auxiliary variables, a non-convex term in the non-convex objective function is subjected to convex approximation processing to obtain a solvable convex optimization form result; Using a semidefinite relaxation method, the rank constraint is converted into a semidefinite constraint, so that the non-convex problem becomes a convex optimization problem, and the problem is solved by a convex optimization algorithm to obtain a solution result; The solution is calculated using Gaussian randomization or eigenvalue decomposition method to obtain a convex objective function.

5. The method for anti-interference communication of unmanned aerial vehicles based on intelligent reflective surface assistance according to claim 1 is characterized in that: The step of introducing slack variables and using continuous convex approximation technology to perform first-order Taylor expansion on part of the non-convex constraints to obtain the convex constraints includes: Introduce slack variables to linearly constrain non-convex constraints and obtain non-convex constraints for linear constraints; By using continuous convex approximation technology, first-order Taylor expansion is performed on the non-convex constraint conditions of some of the linear constraints to obtain convex constraint conditions.

6. The method for anti-interference communication of unmanned aerial vehicles based on intelligent reflective surface assistance according to claim 1 is characterized in that: The step of using the fractional programming algorithm to process the convex objective function and the convex constraint condition to obtain optimized transmission power allocation, intelligent reflection surface reflection coefficient and UAV trajectory includes: Obtaining the initialized convex constraint condition; Substituting the convex objective function and the initialized convex constraint condition into the fractional programming algorithm respectively, performing calculations, and obtaining iteratively optimized transmission power allocation, intelligent reflective surface reflection coefficient, and UAV trajectory; The optimized transmission power distribution, intelligent reflection surface reflection coefficient and UAV trajectory are updated and iterated until convergence to obtain the optimized transmission power distribution, intelligent reflection surface reflection coefficient and UAV trajectory.

7. The method for anti-interference communication of unmanned aerial vehicles based on intelligent reflective surface assistance according to claim 1 is characterized in that: The step of using the alternating optimization algorithm of the S-procedure to iteratively optimize the optimized transmission power distribution, the reflection coefficient of the intelligent reflection surface, and the trajectory of the UAV to obtain the converged target transmission power distribution, the target intelligent reflection surface reflection coefficient, and the target UAV trajectory includes: By using the alternating optimization algorithm of S-procedure, the variables in the iterative optimization process are optimized one by one for the optimized transmit power distribution, intelligent reflective surface reflection coefficient and UAV trajectory to obtain the converged target transmit power distribution, target intelligent reflective surface reflection coefficient and target UAV trajectory.

8. The method for anti-interference communication of unmanned aerial vehicles based on intelligent reflective surface assistance according to claim 1 is characterized in that: The step of obtaining a suboptimal solution for maximizing the energy efficiency of the UAV by using the converged target transmission power allocation, the target intelligent reflective surface reflection coefficient, and the target UAV trajectory comprises: The energy efficiency index is calculated using the converged target transmission power allocation, the target intelligent reflective surface reflection coefficient and the target UAV trajectory to obtain the index calculation result; According to the calculation results of the indicators, determining whether the current optimization result is a suboptimal solution for maximizing energy efficiency; If the energy efficiency index meets a preset threshold or optimization target, a suboptimal solution that maximizes the current energy efficiency is obtained.

9. An anti-interference communication system for unmanned aerial vehicles based on intelligent reflective surfaces, characterized in that: The method for anti-interference communication of unmanned aerial vehicle assisted by intelligent reflective surface applied to any one of claims 1 to 8, the system comprising: An acquisition unit, used to acquire user signals and malicious interference signals, and construct a non-convex objective function and non-convex constraint conditions; A first processing unit is used to convert the non-convex objective function into a convex objective function and a convex constraint condition by using a continuous convex approximation algorithm and the non-convex constraint condition; A second processing unit is used to process the convex objective function and the convex constraint condition by using a fractional programming algorithm to obtain an optimized transmission power allocation, a smart reflective surface reflection coefficient, and a UAV trajectory; The third processing unit is used to perform iterative optimization processing on the optimized transmission power distribution, intelligent reflection surface reflection coefficient and UAV trajectory by using the alternating optimization algorithm of S-procedure to obtain converged target transmission power distribution, target intelligent reflection surface reflection coefficient and target UAV trajectory; The result unit is used to obtain the suboptimal solution for maximizing the energy efficiency of the UAV by utilizing the converged target transmission power allocation, the reflection coefficient of the target intelligent reflective surface and the target UAV trajectory.

10. A computer program, characterized in that When the computer program is executed by a processor, the steps of the anti-interference communication method based on intelligent reflective surface assisted unmanned aerial vehicle according to any one of claims 1 to 8 are implemented.

Images