Research On Directon Finding Techniques Based On MUSIC Algorithm

In the modern electronic warfare, the traditional active radar is easily reconnoitred and intercepted by enemy, so enhance the viability of the radar has become the primary problem of war defence. The research and development of the passive radar detection technology is particularly important. Passive direction finding is an extremely important manner in the non-cooperative passive detection. At the present time, direction of arrival (DOA) estimate is the most widely used in the passive direction finding. Direction of arrival estimate is the core of passive direction finding system, is animportant component of passive direction finding technology.At first this article introduce the research background and the significance of passive direction finding technology, and a further introduction of the research situation of spatial spectrum estimation in passive direction finding technology. Moreover introduce the delayed-addition algorithm and Capon minimum variance estimation, and the performances of them are analyzed and compared. As the resolution of tradional direction of arrival estimate algorithm is low and cannot distinguish the signals in one beam space, further introduce the high resolution MUSIC (Multiple Signal Classification) algorithm and root-MUSIC algorithm and their performances are simulated and compared. The tradional MUSIC algorithm has high computational complexity, its estimate performance sharp descent in colored noise environment and fail for coherent signal sources. This article focuses on the research of a frequency direction of arrival estimation under the colored noise environment and the improvement of modified MUSIC algorithm. The traditional MUSIC algorithm need the eigen-decomposition of the covariance matrix, when the array elements and the snapshots are more, it has high computational complexity and cannot eliminate the effection of colored noise. The frequency direction of arrival estimation in this article can not only eliminate the effection of colored noise, but reduce the computational complexity, more fit for real-time processing. The traditional MUSIC algorithm cannot estimate the direction of arrival of coherent signals, so research the MUSIC algorithm based on spatial smoothing. As the loss of array aperture, the estimate performance of this algorithm is not very good. Then further research the modified MUSIC algorithm and eigen-decomposite the covariance matrix of modified MUSIC algorithm. Because of the shortcomings of these algorithms this article provides a real-value eigen-decomposite the covariance matrix of modified MUSIC algorithm. First, real-value transform the modified covariance matrix, then singular value decomposition for it is given. Then reconstruct the covariance matrix of the data and singular value decompose it. The improved MUSIC algorithm in this article can not only estimate the direction of arrival of spatial independent signals but the direction of arrival of the spatial coherent signals. In addition, simulate the MUSIC algorithm based on forward/backward spatial smoothing, forward and backward spatial smoothing and modified MUSIC algorithm, the performances of these algorithms are detailed analyzed. The simulation results show that the higher the signal to noise ratio and the more the elements, the better the estimate performance of MUSIC algorithm; root MUSIC algorithm avoid the search of spatial spectrum, reduces the calculation, and its estimate performance is better than MUSIC algorithm; forward/backward spatial smoothing algorithm and modified MUSIC algorithm can not only estimate the direction of arrival of spatial independent signals but the direction of arrival of the spatial coherent signals.