Parameter Estimation Of Chirp Signal Based On Group Intelligence Optimization

In recent decades,with the research on radar target characteristics and the rapid advancement of broadband microwave technology,wide band gap semiconductor technology,large-scale integrated circuits and computer application technology,combined with the characteristics of multi-discipline such as pattern recognition theory and machine vision,radar technology has been obtained.With its vigorous development,its functions include not only target detection and positioning,but also tracking,recognition,imaging and classification of targets.Among the many applications of radar,the most important is the radar reconnaissance.Radar detection technology can not only capture,measure,analyze,identify,and lock enemy radar signals,but also obtain useful technical parameters such as technical parameters,operational deployment,and geographic coordinates.However,the captured radar signals are often mixed with noise interference.Therefore,in the background of noise-containing interference,accurate,fast and efficient parameter estimation of chirp signals has important research significance.In addition,in other practical applications,the parameter estimation of chirp signals is also widely used,such as seismic survey,EEG signals,bat sonar signals.When processing these chirp signals,the observation data is often interfered by noise.Therefore,the parameter estimation problem of chirp signals with additional noise interference is an important research topic.The most widely used in chirp signal processing is the maximum likelihood estimation method.The maximum likelihood estimation method can theoretically achieve the Cramer-Rao lower bound,and has the highest estimation accuracy in the parameter estimation application.However,due to the requirement of the two-dimensional search for the maximum likelihood estimation method,the storage amount and the calculation amount are large,which is very To a large extent,its application and development are limited.Therefore,how to reduce the amount of search calculation of the maximum likelihood method in parameter estimation is an important issue.In this paper,the particle swarm optimization algorithm and the artificial bee colony algorithm are used to study this problem.Firstly,the particle swarm optimization algorithm is introduced to optimize the parameter estimation under the maximum likelihood estimation optimization of multi-component chirp signals.The main work is as follows:(1)In order to greatly reduce the computational complexity of maximum likelihood parameter estimation,a global mode PSO algorithm for maximum likelihood estimation parameter estimation of multi-component chirp signals is proposed.The method has a simple structure and a fast convergence speed,and can effectively reduce the calculation amount.(2)In order to realize the effective estimation of parameters under the condition of lower signal-to-noise ratio,this paper proposes a local mode PSO algorithm for estimating the maximum likelihood estimation parameters of multi-component chirp signals.The algorithm has good search stability and is not easy to fall into local optimum,and has high noise resistance.(3)In order to improve the parameter estimation success rate and search accuracy,this paper proposes a mixed-mode PSO algorithm for multi-component chirp signal maximum likelihood estimation parameter estimation based on the neighborhood structure characteristics of global mode and local mode particle swarm optimization.In order to further improve the parameter estimation resolution,this paper introduces the artificial bee colony algorithm to optimize the parameter estimation under the maximum likelihood estimation of multi-component chirp signal parameters,mainly from the following three points:(1)Based on the characteristics of artificial bee colony algorithm with few control parameters,easy implementation and strong robustness,this paper proposes a basic artificial bee colony algorithm for estimating the maximum likelihood estimation parameters of multicomponent chirp signals.The algorithm has strong global search and development capabilities and high resolution of signal parameters.(2)In order to speed up the convergence of the algorithm,this paper adds global crossover operation and multi-dimensional parallel search mechanism,and proposes a multi-dimensional parallel search global artificial bee colony algorithm for multi-component chirp signal maximum likelihood estimation parameter estimation.The algorithm has good convergence and fast convergence.(3)In order to further improve the accuracy of the algorithm,this section combines the characteristics of the basic artificial bee colony algorithm and the multi-dimensional parallel search global artificial bee colony algorithm,and proposes a multi-dimensional parallel search hybrid artificial bee colony for multi-component chirp signal parameter estimation.The algorithm has good stability,is not easy to fall into local optimum,and has high search precision.