Research On Wide-Band Signals Direction Finding Algorithms

As an important part of the modern signal processing, high-resolution direction of arrival estimation technique has a promising perspective in many fields such as radar and sonar, with the development of the microelectronic and digital techniques. At first, high-resolution direction of arrival estimation algorithm is presented in the narrow-band environment. Wideband signals include more affluent information and make for the detection of target, the parameter estimation and the abstraction of target characteristic, so more and more wide-band signal processing is researched. Direction of arrival of sources for wideband signals become an important part of wideband array signal processing. The dissertation studies the theories of high-resolution DOA (Direction of Arrival) estimation for wideband array signals processing. The main results and achievements are summarized as follows:(1)The modified wide-band propagator algorithm is presented. The conjugate data being rearranged Toeplitz (TOP) method is amplified on the wide-band propagator algorithm. The Tightness of algorithm is verified in theory. Through theory analysis and simulation, this algorithm can obviously improve the performances of DOA estimation when the snapshot number is limited and the SNR (Signal Noise Ratio) is low. Moreover its performance is better than OPM (Orthogonal Propagator Method) algorithm, and its computational complexity is nearly about equal to that of the typical propagator algorithm.(2)A novel wide-band Direction of Arrival estimation method is proposed. The data vector model is constructed in time-frequency domain by computing the cross-spectra of continuous wavelet of the reference sensor and other sensors. Making use of part narrow-band characteristic of Wide-band Chirp signal, traditional array correlation matrix is replaced by the time-frequency correlation matrix for each source that is formed by choosing the time-frequency points residing on its the source energy to realize signal DOA estimation by characteristic-decomposition. The algorithm can be processed in both spatial and time-frequency domain, and make the best of localizing the source energy in the time-frequency domain. The method not only separates nonstationary signals in time-frequency domain, but also improves SNR efficiently and promotes the robust against noises. The computer simulation verifies the efficiency of the new method.