Research On Joint Parameter Estimation Algorithm In Time-Frequency And Space Domain

With the development in the field of electronic and telecommunication and the improvement of electronic warfare signal processing technology, many kinds of signals distribute in a large span of time, frequency and space domain. Under this condition, higher requirement for signal reconnaissance and processing technology is proposed and it is one of the leading topics in this subject that researching on the joint parameter estimation method in time-frequency and space domain which can adapt to modern electronic warfare signal environment and have excellent performance. In this dissertation, under the array antenna environment, the joint parameter estimation methods in time-frequency and space domain are studied, which are mainly for linear frequency modulated(LFM) signals. The major contributions of this dissertation can be summarized as in a few aspects as follow:1. The method to estimate time-frequency parameters using time-frequency distributions and classical methods for direction of arrival(DOA) estimation are briefly introduced in this dissertation. On one hand, through the analysis of cross-terms suppression method in Wigner-Ville Distribution(WVD), the necessity of using spatial information to improve the performance of time-frequency distribution is illustrated. On the other hand, by analyzing the deficiencies and defects of the classical subspace decomposition DOA estimation method, the main idea and the advantages of subspace decomposition DOA estimation algorithm based on spatial time-frequency distribution(STFD) is proposed.2. The joint parameter estimation method in time-frequency and space domain based on WVD is studied in this dissertation. First, the array signal model for non-stationary signals is established and the ways to suppress the cross-terms in WVD by using spatial information are described, which shows a better performance to the processing only in time-frequency domain. Then, the subspace decomposition DOA estimation algorithms based on STFD are detailed analyzed, which show better performances to the classical subspace decomposition DOA estimation methods through theory and simulations. Finally, an improved WVD-MUSIC algorithm is proposed by amending the construction of the STFD matrix. The results of simulations illustrate that the proposed method can overcome the shortcomings of the original algorithm in a certain degree and shows more robustness for the variation of chirp rate of LFM signal.3. The joint parameter estimation method in time-frequency and space domain based on Fractional Fourier Transform(Fr FT) is studied in this dissertation. First, the fast discrete Fr FT for reference array sensor and the estimation formula for time-frequency parameter are derived and after that, the Fr FT-MUSIC algorithm for DOA estimation is studied. Finally, the steps and processes of the joint parameter estimation method based on Fr FT are summarized and the DOA estimation performance of WVD-MUSI and Fr FT-MUSIC are comparative analyzed by simulations.