Enhanced MUSIC Approach for Estimation of Directions of Arrival for Multiple Coherent Wideband Sources

Abstract

Direction of Arrival (DOA) is a term used in signal processing literature to describe the direction from which a propagating wave typically arrives at a place, which is usually where a set of sensors are located. Many signal processing applications need the estimation of a set of unknown characteristics based on data gathered from a variety of sensors. Electronic surveillance, sonar, and radar are only a few of many applications that require high resolution DOA estimate and among these applications, DOA estimation of narrow band signals is one that has drawn a lot of interest from researchers. The most promising class of approaches among those suggested to solve the DOA problem is signal subspace algorithms. These techniques attempt to divide the space spanned by the measured data into subspaces known as signal and noise by taking advantage of the underlying data model. The approach that has drawn the greatest interest and been examined the most within this class of algorithms The Multiple Signal Classification (MUSIC) technique can be used with arbitrary geometry [1]. This paper focusses on estimating the DOA using MUSIC method but one of the main issues in this method is its inability to estimate highly correlated signals due to loss of non-singularity property of the covariance matrix that is used in the signal model [2], therefore a developed method (Focused Method) will be proposed in this paper and used to overcome this problem. The direction of arrival estimation is simulated on a MATLAB software with a set of input parameters such as array elements, signal to noise ratio, number of snapshots and number of signal sources. An extensive simulation has been conducted and the results show that the MUSIC can achieve an accurate and efficient DOA estimation in coherent cases.