Comparison of conventional and subspace based algorithms to estimate Direction of Arrival (DOA)

The Direction of Arrival (DOA) technology and researches are evolving areas in array signal processing. Biomedicine, sonar, radar, seismology, astronomy communication and earthquake widely use this estimation. We receive an incoming wave by sensing an array or a set of arrays, this sensed wave can be processed to derive many related estimation information of the associated wave. The Direction of arrival angle even known as the arrival angle estimation is one such vital parameter. The two broad classifications of estimations that can be employed to find out this direction of arrival are the conventional method of estimation and the subspace based methods. The beam forming and null steering forms the base for the conventional methods whereas the subspace based methods continuously exploit the data matrix of the narrowband received signal. The conventional methods are limited with resolution, so we need to move towards the subspace methods. The two conventional estimation approaches discussed here are the delay and sum method and the Capon's method. A comparison plot for these two conventional approaches is also carried out. Multiple Signal Classification (MUSIC) and ESPRIT (Estimation of Signal Parameter through Rotational Invariance Technique) are the two widely used subspace based estimation methods. In this paper the concepts of the arrival angle estimation and some MATLAB simulations to highlight the factors that can especially improve accuracy of the MUSIC algorithm are discussed and attempts are made to bring about better resolution and contrast.