Unmanned aerial vehicle narrowband communication frequency optimization method for improving fruit fly optimization algorithm
By constructing a dynamic interference model and improving the search strategy of the fruit fly optimization algorithm, the problem that traditional algorithms are prone to falling into local optimality in UAV narrowband communications is solved, more efficient frequency optimization is achieved, and the anti-interference ability and quality of UAV communications are improved.
Patent Information
- Application Number
- CN202510667508.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-22
- Publication Date
- 2025-09-12
AI Technical Summary
The traditional fruit fly optimization algorithm is prone to falling into local optimality in UAV narrowband communication, and the existing interference model assumes that the interference signal is static and cannot adapt to the dynamic changing actual communication environment.
A dynamic interference model is constructed to improve the olfactory search and visual search of the fruit fly optimization algorithm. The multi-objective optimization strategy is combined to optimize the frequency selection through adaptive step size and mutation operations.
The global search capability and convergence speed of the algorithm are significantly improved, and the anti-interference performance and communication quality of UAV narrowband communication are enhanced.
Smart Images
Figure CN120640418A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of swarm intelligence, and in particular relates to a method for optimizing the narrowband communication frequency of unmanned aerial vehicles (UAVs) by improving a fruit fly optimization algorithm. Background Art
[0002] With the advancement of science and technology and the maturity of technology, drones have been widely used in numerous fields. To improve the anti-interference capabilities of drone narrowband communications, numerous anti-interference technologies have been studied. Frequency selection is a key component, and its optimization algorithm is crucial for improving anti-interference performance. Currently, with the development of artificial intelligence, interference technology is showing a new intelligent trend. Existing technologies have begun to focus on intelligent solutions for communication anti-interference technology, such as existing technologies that combine game theory and machine learning, and use the Fruit Fly Optimization Algorithm (FOA). As a new type of swarm intelligence optimization algorithm, the Fruit Fly Optimization Algorithm has been widely used due to its simple principle and small number of parameters.
[0003] However, traditional fruit fly optimization algorithms are prone to falling into local optima in the later stages of the search. Furthermore, traditional interference models often assume that interference signals are static, whereas the frequency, amplitude, and phase of interference signals in real drone communication environments can change dynamically over time. Therefore, a fruit fly optimization algorithm that accounts for dynamic changes and is less susceptible to falling into local optima is urgently needed for optimizing drone narrowband signals. Summary of the Invention
[0004] To solve the above problems, the present invention provides a method for optimizing the narrowband communication frequency of UAVs by improving the fruit fly optimization algorithm.
[0005] To implement the above technology, the specific steps are as follows: S1. Collect UAV interference signals and useful signals in narrowband communication, build a UAV narrowband communication interference model including a received signal model and an interference signal model; and build an available signal-to-noise ratio (SNR) based on the useful signal power, interference signal power, and noise signal power; The received signal model consists of the useful signal, interference signal and noise signal, and the expression is as follows: Where, Indicates the received signal; Indicates useful signal; Indicates interference signal; represents Gaussian white noise signal; The interference signal model consists of the interference signal amplitude, interference signal frequency and interference signal initial phase, and the expression is as follows: Where, Indicates the interference signal amplitude; Indicates the frequency of the interference signal; represents the initial phase of the interference signal; The available signal-to-noise ratio (SNR) is constructed based on the useful signal power, interference signal power, and noise signal power to measure communication quality. The expression is as follows: Where, Indicates the useful signal power; represents the interference signal power; Represents the noise power.
[0006] S2. Based on the interference signal model in the UAV narrowband communication interference model, a dynamic interference model is constructed. The expression is as follows: Where, and Represent the first parameter of the model and the second parameter of the model respectively; and denote the autoregressive order and the moving average order respectively; express The frequency of the interference signal at all times; express White noise at all times.
[0007] S3, based on maximizing the signal-to-noise ratio, minimizing the communication signal fluctuation and minimizing the energy consumption, a weighted method is used to construct the objective function; The expression of the objective function constructed using the weighted method is as follows: Where, 、 and Represent the weight coefficients of signal-to-noise ratio, communication signal fluctuation and energy consumption respectively; represents the maximum signal-to-noise ratio; It means minimizing the fluctuation of communication signal; Indicates minimizing energy consumption.
[0008] S4. Improve the olfactory search and visual search in the traditional fruit fly optimization algorithm and construct an improved fruit fly optimization algorithm; Combining the dynamic interference model and the objective function, the olfactory search is improved by introducing an adaptive step size and dynamically adjusting the search step size according to the number of iterations and the change of the interference signal. The expression is as follows: Where, represents the distance between the fruit fly individual and the origin in the improved fruit fly optimization algorithm; represents the flavor concentration judgment value in the improved fruit fly optimization algorithm; represents the initial step length; Indicates the current iteration number; Indicates the maximum number of iterations; represents the adjustment factor; Indicates the change in the frequency of the interference signal; ''and ''represents the horizontal and vertical coordinate positions of the initialized fruit fly population; 、, Respectively represent the horizontal and vertical coordinates of the optimal individual position of the current population; In the visual search stage, mutation operation is introduced to increase population diversity and prevent the algorithm from falling into local optimality. At the same time, mutation judgment is performed based on the multi-objective optimization results of the objective function. The expression is as follows: Where, Represents a random number within (0,1); represents the probability of mutation; 、 、 、 They represent the minimum and maximum horizontal coordinate boundaries and the minimum and maximum vertical coordinate boundaries of the search space respectively.
[0009] S5. Use the improved fruit fly optimization algorithm to solve the problem with the goal of maximizing the objective function and obtain the optimal frequency solution; The solution steps include: 1) Initialize population parameters, including population size, maximum number of iterations, initial step size, mutation probability, and weight coefficient; 2) Randomly initialize the position of the fruit fly population; 3) Monitor interference signal parameters in real time based on the dynamic interference model, calculate the distance between individual fruit flies and the optimal individual in the population, determine the flavor concentration, and calculate the flavor concentration based on a multi-objective optimization strategy; 4) Find the individual with the best flavor concentration in the current population and update the population position; 5) Determine whether the maximum number of iterations has been reached. If so, output the optimal solution. Otherwise, return to step 3).
[0010] Beneficial effects of the present invention: This paper constructs a dynamic interference model and applies a multi-objective optimization strategy to improve the olfactory and visual search phases of the traditional fruit fly optimization algorithm, significantly enhancing the algorithm's global search capability and convergence speed. Furthermore, a communication interference model is established, transforming the frequency selection problem into an optimization problem. By improving the algorithm's search phase, algorithm performance is enhanced, boosting the anti-interference capabilities of unmanned narrowband communications.
[0011] The simulation results of the present invention show that the improved algorithm can effectively improve the anti-interference performance of UAV narrowband communication in an interference environment, and further improve the communication quality and reliability. BRIEF DESCRIPTION OF THE DRAWINGS
[0012] Figure 1 is a flow chart of the steps of the present invention; Figure 2 Schematic diagram of a 2.4 GHz frequency band dynamic interference model constructed in an embodiment of the present invention; Figure 3 Schematic diagram of the population frequency distribution changes in an embodiment of the present invention, where (a) shows the population frequency distribution when the number of iterations is 1; (b) shows the population frequency distribution when the number of iterations is 67; (c) shows the population frequency distribution when the number of iterations is 134; and (d) shows the population frequency distribution when the number of iterations is 200. Figure 4 Schematic diagram of the simulation convergence speed results in an embodiment of the present invention; Figure 5 This is a comparison chart of the experimental results of the present invention. DETAILED DESCRIPTION
[0013] The present invention is further described in detail below with reference to specific embodiments.
[0014] like Figure 1 As shown, a method for optimizing the narrowband communication frequency of a UAV using an improved fruit fly optimization algorithm includes the following steps: S1. Collect UAV interference signals and useful signals in narrowband communication, build a UAV narrowband communication interference model including a received signal model and an interference signal model; and build an available signal-to-noise ratio (SNR) based on the useful signal power, interference signal power, and noise signal power; In narrowband communications, interference signals are superimposed on useful signals, affecting signal transmission; The received signal model consists of the useful signal, interference signal and noise signal, and the expression is as follows: Where, Indicates the received signal; Indicates useful signal; Indicates interference signal; represents Gaussian white noise signal; The interference signal model consists of the interference signal amplitude, interference signal frequency and interference signal initial phase, and the expression is as follows: Where, Indicates the interference signal amplitude; Indicates the frequency of the interference signal; represents the initial phase of the interference signal; The available signal-to-noise ratio (SNR) is constructed based on the useful signal power, interference signal power, and noise signal power to measure communication quality. The expression is as follows: Where, Indicates the useful signal power; represents the interference signal power; Represents the noise power.
[0015] S2. Construct a dynamic interference model based on the interference signal model in the UAV narrowband communication interference model; Traditional interference models often assume that interference signals are static. However, in actual UAV communication environments, the frequency, amplitude, and phase of interference signals change dynamically over time. Therefore, this application constructs a dynamic interference model to monitor changes in interference signal parameters in real time. By establishing a time series model of the interference signal parameters, such as the autoregressive moving average (ARMA) model, the changing trend of the interference signal is predicted. The expression is as follows: Where, and Represent the first parameter of the model and the second parameter of the model respectively; and denote the autoregressive order and the moving average order respectively; express The frequency of the interference signal at all times; express White noise at all times.
[0016] S3, based on maximizing the signal-to-noise ratio, minimizing the communication signal fluctuation and minimizing the energy consumption, a weighted method is used to construct the objective function; In narrowband communication between drones, it is not only important to maximize the signal-to-noise ratio (SNR) to improve communication quality, but also to consider communication stability and energy consumption. Therefore, a multi-objective optimization strategy is introduced, with maximizing the SNR, minimizing communication signal fluctuations, and minimizing energy consumption as the objective functions. The expression for maximizing the signal-to-noise ratio is as follows: Where, Indicates the The power of the interfering signal; M Indicates the total number of interference signals; The variance of the signal is used to measure the degree of fluctuation. The expression for minimizing the fluctuation of the communication signal is as follows: Where, represents the variance of the signal; Indicates signal communication time; An index indicating the signal communication time; Indicates the The signals received at the moment include useful signals, interference signals and noise signals; Indicates the signal at the communication time The mean within The transmission power is and the communication time is Construct the minimum energy consumption, the expression is as follows: Where, Indicates energy consumption; The expression of the objective function constructed using the weighted method is as follows: Where, 、 and represent the weight coefficients of signal-to-noise ratio, communication signal fluctuation and energy consumption respectively, and , adjust the weight value according to the actual application requirements to balance the importance of different objectives.
[0017] S4. Improve the olfactory search and visual search in the traditional fruit fly optimization algorithm and construct an improved fruit fly optimization algorithm; The traditional fruit fly optimization algorithm simulates the foraging behavior of fruit flies, that is, in the search space, individual fruit flies perform olfactory and visual searches based on their own position information; The steps of the traditional fruit fly optimization algorithm include: Initialize the population: randomly initialize the position of the fruit fly population ''、 '', , is the population size; Olfactory search: calculating the distance between the fruit fly individual and the origin '', the expression is as follows: ; In the present invention, the coordinates of the fruit fly individual are expressed as ( '', ''), represents a candidate solution in the optimization problem, that is, a frequency selection scheme; the origin represents the reference point, that is, the initial center of the solution space, and represents the default frequency reference value in the present invention; Calculate the taste concentration judgment value, the expression is as follows: In the present invention, the flavor concentration judgment value is used to reflect the "attractiveness" of an individual location; Calculating flavor intensity '', flavor concentration ''In this invention, it represents the objective function, which is used to evaluate the quality of the solution; Visual search: Find the individual with the best flavor concentration in the current population and update the population position. The expression is as follows: Where, Represents a random value; , Respectively represent the horizontal and vertical coordinates of the optimal individual obtained by the search; 、, Respectively represent the horizontal and vertical coordinates of the optimal individual position of the current population; The optimal individual refers to the fruit fly individual with the highest fitness (highest flavor concentration) in the current population, representing the optimal solution currently found; Combining the dynamic interference model and the objective function, the olfactory search is improved by introducing an adaptive step size and dynamically adjusting the search step size according to the number of iterations and the change of the interference signal. The expression is as follows: Where, Indicates the distance between the fruit fly individual and the origin in the improved fruit fly optimization algorithm; Represents the flavor concentration judgment value in the improved fruit fly optimization algorithm; represents the initial step length; Indicates the current iteration number; Indicates the maximum number of iterations; represents the adjustment factor; Indicates the change in the frequency of the interference signal; In the visual search stage, mutation operation is introduced to increase population diversity and prevent the algorithm from falling into local optimality. At the same time, mutation judgment is performed based on the multi-objective optimization results of the objective function. The expression is as follows: Where, Represents a random number within (0,1); represents the probability of mutation; 、 、 、 They represent the minimum and maximum horizontal coordinate boundaries and the minimum and maximum vertical coordinate boundaries of the search space respectively; during mutation, individuals that make the multi-objective comprehensive fitness better are retained first.
[0018] S5. Use the improved fruit fly optimization algorithm to solve the problem with the goal of maximizing the objective function and obtain the optimal frequency solution; The maximization objective function is expressed as: ; Constraints include available frequency range, communication bandwidth, and transmit power limits; The solution steps include: 1) Initialize population parameters, including population size, maximum number of iterations, initial step size, mutation probability, and weight coefficient; 2) Randomly initialize the position of the fruit fly population; 3) Monitor interference signal parameters in real time based on the dynamic interference model, calculate the distance between individual fruit flies and the optimal individual in the population, determine the flavor concentration, and calculate the flavor concentration based on a multi-objective optimization strategy; 4) Find the individual with the best flavor concentration in the current population and update the population position; 5) Determine whether the maximum number of iterations has been reached. If so, output the optimal solution. Otherwise, return to step 3).
[0019] In order to verify the present invention, simulation verification was carried out, as follows: Simulations were performed in Matlab to set the drone's narrowband communication parameters, including center frequency and bandwidth. Interference signal parameters, including frequency and power, were also set, and the dynamic variation of the interference signal parameters was configured according to the dynamic interference model. The performance of the traditional and improved Fruit Fly optimization algorithms for anti-interference communication frequency selection was compared.
[0020] Construct a dynamic interference model in the 2.4G frequency band commonly used for drone communications, such as Figure 2 As shown; As the number of iterations increases, Figure 3 As shown in parts (a), (b) and (c), the frequency distribution is relatively fixed, that is, the optimized frequency finally selected, as shown in Figure 3 As shown in part (d); like Figure 4 As shown in the figure, the fitness values of the objective functions of different algorithms change with the increase of the number of iterations. Compared with the traditional FOA, the improved FOA reaches a higher objective function faster, is significantly better than the traditional FOA in convergence speed, and has a better optimization effect.
[0021] The simulation parameters are detailed in Table 1; Table 1: Simulation parameters Analysis of the experimental results shows that the improved fruit fly optimization algorithm converges faster, finding the optimal solution more quickly and avoiding local optima. This is because the dynamic interference model enables the algorithm to adapt to interference changes in a timely manner, and the multi-objective optimization strategy guides the algorithm to balance the search across multiple objectives.
[0022] The frequency scheme selected by the improved algorithm achieves a higher signal-to-noise ratio, smaller signal fluctuations, lower energy consumption, stronger anti-interference capabilities, and more stable communication quality. By adjusting the weight coefficient, the different communication quality and energy consumption requirements of different application scenarios can be met.
[0023] like Figure 5 As shown in the figure, the red area represents the frequency range of the interference signal; the red Indicates the interference center frequency; the blue dotted line indicates the optimal frequency optimized in an interference-free environment; the green Represents the optimal frequency after anti-interference optimization; the blue area represents the signal bandwidth selected by the fruit fly optimization algorithm (FOA) without interference; the green area represents the signal bandwidth selected by the improved fruit fly optimization algorithm; Figure 5 It can be seen that in the environment without interference optimization, FOA is selected based only on SNR, and the bandwidth is naturally selected in the blue area with a center frequency of 2430 MHz (blue dotted line). In the environment with interference optimization, the improved FOA is optimized through a multi-factor fitness function, and the bandwidth is selected in the green area with a center frequency of 2485 MHz (green The signal bandwidth optimized for anti-interference is also outside the green interference zone, indicating that the FOA successfully avoids the interference source through its fitness function. The frequency selected by the anti-interference FOA effectively avoids the interference zone, while traditional FOA relies solely on SNR, which may be affected in the presence of interference.
[0024] The frequency scheme selected by the improved algorithm has a higher communication signal-to-noise ratio, smaller communication signal fluctuations, lower energy consumption, stronger anti-interference ability, and more stable communication quality. By adjusting the weight coefficient, the different requirements of different application scenarios for communication quality and energy consumption can be met. The improved fruit fly optimization algorithm proposed in this paper is applied to frequency selection for narrowband anti-interference communications in unmanned aerial vehicles (UAVs). By constructing a dynamic interference model and introducing a multi-objective optimization strategy, the algorithm's search phase is improved, enhancing its global search capability and convergence speed. Simulation results demonstrate that the algorithm effectively improves the anti-interference performance of UAV narrowband communications in interference environments, providing new insights and methods for anti-interference technology in UAV narrowband communications. Future research is expected to further investigate the application of the algorithm in more complex interference environments, as well as its integration with other anti-interference technologies. Furthermore, more rational multi-objective weight allocation methods will be explored to accommodate a wider range of application needs.
[0025] It should be noted that the above are only preferred embodiments of the present application and do not limit the scope of patent protection of the present application. Any equivalent structure or equivalent process transformation made using the contents of the description and drawings of this application, or directly or indirectly applied in other related technical fields, are also included in the scope of patent protection of the present application.
Claims
1. A method for optimizing the narrowband communication frequency of unmanned aerial vehicles based on an improved fruit fly optimization algorithm, characterized in that: The following steps are involved: S1. Collect UAV interference signals and useful signals in narrowband communication, and build a UAV narrowband communication interference model including a received signal model and an interference signal model; Constructing an available signal-to-noise ratio based on the useful signal power, the interference signal power, and the noise signal power; S2. Construct a dynamic interference model based on the interference signal model in the UAV narrowband communication interference model; S3, based on maximizing the signal-to-noise ratio, minimizing the communication signal fluctuation and minimizing the energy consumption, a weighted method is used to construct the objective function; S4. Improve the olfactory search and visual search in the traditional fruit fly optimization algorithm and construct an improved fruit fly optimization algorithm; The olfactory search is improved by combining the dynamic interference model and the objective function, introducing an adaptive step size, and dynamically adjusting the search step size according to the number of iterations and the change of the interference signal to improve the olfactory search; The visual search improvements are: introducing mutation operations to increase population diversity and prevent the algorithm from falling into local optimality; S5. Use the improved fruit fly optimization algorithm to solve the problem with the goal of maximizing the objective function and obtain the optimal frequency solution.
2. The method for optimizing the narrowband communication frequency of a UAV using an improved fruit fly optimization algorithm according to claim 1, characterized in that: The method collects interference signals and useful signals of drones in narrowband communications, and constructs a drone narrowband communication interference model including a received signal model and an interference signal model. The received signal model is composed of a useful signal, an interference signal and a noise signal; the interference signal model is composed of an interference signal amplitude, an interference signal frequency and an initial phase of the interference signal.
3. The method for optimizing the narrowband communication frequency of a UAV using an improved fruit fly optimization algorithm according to claim 1, characterized in that: The expression for constructing the dynamic interference model based on the interference signal model in the UAV narrowband communication interference model is as follows: Where, and Represent the first parameter of the model and the second parameter of the model respectively; and denote the autoregressive order and the moving average order respectively; express The frequency of the interference signal at all times; express White noise at all times.
4. The method for optimizing the narrowband communication frequency of a UAV using an improved fruit fly optimization algorithm according to claim 1, wherein: The expression of the objective function constructed by the weighted method based on maximizing the signal-to-noise ratio, minimizing the communication signal fluctuation and minimizing the energy consumption is as follows: Where, 、 and Represent the weight coefficients of signal-to-noise ratio, communication signal fluctuation and energy consumption respectively; represents the maximum signal-to-noise ratio; It means minimizing the fluctuation of communication signal; Indicates minimizing energy consumption.
5. The method for optimizing the narrowband communication frequency of unmanned aerial vehicle (UAV) using an improved fruit fly optimization algorithm according to claim 1, characterized in that: The olfactory search and visual search in the traditional fruit fly optimization algorithm are improved. In constructing the improved fruit fly optimization algorithm, the olfactory search is improved. The expression is as follows: Where, represents the distance between the fruit fly individual and the origin in the improved fruit fly optimization algorithm; represents the flavor concentration judgment value in the improved fruit fly optimization algorithm; represents the initial step length; Indicates the current iteration number; Indicates the maximum number of iterations; represents the adjustment factor; Indicates the change in the frequency of the interference signal; ''and ''represents the horizontal and vertical coordinate positions of the initialized fruit fly population; 、, Respectively represent the horizontal and vertical coordinates of the optimal individual position of the current population; Improve the visual search, the expression is as follows: Where, Represents a random number within (0,1); represents the probability of mutation; 、 、 and They represent the minimum and maximum horizontal coordinate boundaries and the minimum and maximum vertical coordinate boundaries of the search space respectively.