ICA based DoA estimation with minimum input data

Abstract

This paper presents a method for locating electromagnetic sources with an optimized number of measurements by using Independent Component Analysis method. This method can detect the Direction of Arrival (DoA) of sources in special cases such as application of equal number of sensors and sources and detection of sources placed one behind another (overlapped). Simulations and measurements were performed to evaluate the performance of the presented method as well as the comparison to MUSIC and ESPRIT classical algorithms. The ICA method does not only estimate the number of sources, but also has the capability to correlate the DoA of the estimated signal to its source. The advantages and disadvantages of the three methods show the high potential application of the ICA as DoA algorithm.