Font Size: a A A

The Research Of DOA Estimation Algorithm In Low Signal Noise Ratio

Posted on:2016-01-07Degree:MasterType:Thesis
Country:ChinaCandidate:Y LiuFull Text:PDF
GTID:2348330542476228Subject:Information and Communication Engineering
Abstract/Summary:PDF Full Text Request
In array signal processing,the technology is the most important problem of DOA estimation,DOA estimation is widely used in radar,communication,electronic,biomedical and other fields.So DOA estimation is a very important research in the array signal processing.The spectrum estimation based on MUSIC algorithm has high accuracy and resolution ratio,it is widely applied in the field of DOA estimation.Searching the spectral peak would take a long time by MUSIC algorithm which is impacted by Signal-to-noise ratio especially in 2D and 3D problems,the time will grow exponentially.MUSIC algorithm can't satisfy the requirement of high real-time.This paper has done a lot of research about spectrum peak searching of MUSIC technology.Firstly,this paper studies a lot of relevant theoretical on MUSIC algorithm,the existing array signal DOA estimation method is analyzed.This paper introduces the mathematical model and the MUSIC algorithm of spectrum peak searching in array signal,then expounds the idea and theoretical knowledge of Genetic and Invasive weed optimization Algorithm.In the third chapter studies several researches on the genetic algorithm and solves the problems.Traditional genetic algorithm exists a problem which is easy to fall into local optimal solution in low SNR.Then an improved genetic algorithm is presented,which improves Genetic Algorithm crossover operator _cp and mutation operator_mp.The improved algorithm is used in the spectral peak searching of MUSIC algorithm and verified by simulation experiments.In the last section of this chapter the paper discusses two signal source and three signal sources in space and verifies it has better global convergence.Finally,the fourth chapter introduces the principle of differential algorithm.In view of the traditional algorithm of weeds has poor convergence and low precision search in low SNR,this paper proposes an improved algorithm of weeds.Before new seed generates by traditional algorithm,descending order target function and introducing a number between 0~1 to change simple linear relationship when traditional algorithm produces weed seeds.After the production of new seeds,the mutation,crossover and selection operations of difference algorithm are introduced into the new algorithm.Then the improved algorithm are applied in the spectrum peak searching of MUSIC and verified by simulation experiment.In the last section of this chapter the paper discusses two signal source and three signal sources in space and verifies it has better real-time.
Keywords/Search Tags:MUSIC algorithm, Spectral peak searching, weed optimization Algorithm, genetic algorithm, convergence, real-time
PDF Full Text Request
Related items