The estimation error performance of the Iterative Compressive Sensing Direction of Arrival (icsDOA) algorithm for signals received over a multipath channel

For decades, direction of arrival (DOA) sensing has been used all across the world. New and more powerful direction sensing techniques are developed every year. In 2016 a new method of applying compressive sensing techniques to direction of arrival estimation called the icsDOA was created. This algorithm performs much more accurately than any other direction sensing technology used in the modern era. This algorithm is named the 'Iterative Compressive Sensing Direction of Arrival', or icsDOA algorithm. The icsDOA algorithm uses the compressive sensing recovery method as its basis, but implements it multiple times in an iterative manner. The algorithm converges to a directional indicator in two iterations and eliminates angular quantisation error. This method only requires a single snapshot of complex voltage outputs from an antenna array to determine a highly accurate directional indication vector. The performance of this algorithm has been demonstrated for signal transmission over an additive Gaussian noise channel. In many practical circumstances of DOA sensing, there is a multipath signal component that degrades the performance of the direction of arrival algorithm. This thesis establishes the achievable DOA estimation accuracy of the icsDOA Algorithm for signals received over a multipath channel.