The Application Of Subspace Decomposition Algorithm In Parameter Estiomation

Subspace decomposition algorithm is a high resolution method after the1970s. It can accurately estimate signal parameters (frequency, orientation, etc). Its performance is ideal. its distinguishing ability and estimating precision are higher than the traditional method. It has been popularized to the most technical fields. Its characteristic is that divide the receiving data into two independent and orthogonal subspace parts through mathematical transformation and use its characteristic to assess parameters. So it can be divided into two types of algorithms, signal subspace algorithm and noise subspace algorithm. The former is represented by rotation invariant subspace algorithm, as follows ESPRIT, LS-ESPRIT, TLS-ESPRIT, TAM, etc. The latter is represented by Muitiple signal Classification algorithm, as follows MUSIC, ROOT-MUSIC, Improved Toeplitz algorithm, MD-MUSIC, the MNM algorithm, etc. Subspace decomposition algorithm has been accepted by scientists for its good properties.Frequency estimation is an important content of the signal processing technology. frequency estimation technology research under the background of noise has become a scientific research and is applied in many industries. Exploring higher accurate frequency technology has a significant scientific research and application value. Space spectrum estimation is an important content of array signal processing technology research. it broke the Rayleigh restriction and can accurately assess multiple source location at the same time. DOA technology is hot direction in this field and has good prospect, but classic DOA technology often doesn’t distinguish location or happens performance deterioration. So exploring the algorithm which has excellent performance and higher precision is very necessary.This paper mainly describes the application of subspace decomposition algorithm in parameter estimation, especially in frequency estimation and space spectrum techology. In the paer, isagoge firstly introduces the frequency estimation and space spectrum technology’s background, significance, scientific research process and development trend. Secondly, the2,3,4chapter take ESPRIT algorithm and MUSIC algorithm as representative algorithm and respectively introduce the application of subspace decomposition algorithm in the frequency estimation, one-dimensional Direction of Arrival estimation, two-dimensional Direction of Arrival estimation, the paper indicates algorithm’s feasibility and simulation performance from experimental and theoretical two angles. According to the principle of MUSIC algorithm, Put forward its popularization, Toeplitz optimization method and two-dimensional MNM algorithm, and show that their feasibility and simulation performance by flowchart and MATLAB software. Finally it makes a comprehensive comparision for two kinds of subspace decomposition algorithm in the last of each chapter. Analysis results show they both have advantages and disadvantages in different parameter estimate fields, the computation of ESPRIT algorithm is smaller, the accuracy of MUSIC algorithm is better. their performance relate with different parameters, they are suitable for different places. The5chapter summarizes research work of full text and points out the future research direction of subspace decomposition algorithm.The paper has three innovation points, as follows:1.ESPRIT algorithm and MUSIC algorithm are applied to frequency estimation. Show their feasibility and simulation performance with flow chart and MATLAB software and make a comprehensive comparison for their performance.2.The improvement of1D-MUSIC algorithm-The improved Toeplitz algorithm is proposed. The paper reveals its feasibility and simulation performance by flow chart and MATLAB software and makes a comprehensive comparison for the performance of1D-ESPRIT algorithm and1D-MUSIC algorithm from experimental and theoretical two angles. 3.The popularization of2D-MUSIC algorithm-The two dimensional MNM algorithm is proposed. The paper reveals its feasibility and simulation performance by flow chart and MATLAB software and makes a comprehensive comparison for the performance of2D-ESPRIT algorithm and2D-MUSIC algorithm from experimental and theoretical two angles.