A new real time radio frequency direction finding algorithm for Gaussian and non-Gaussian noise environments

In this paper, a new superresolution direction finding (SRDF) algorithm for multiple incident radio waves is proposed. Superresolution methods enable resolution of signals separated by less than the natural beamwidth of the array. This ability enables the algorithms to separate the closely spaced signals encountered in a multipath environment. The algorithm is termed loaded capon and is shown to be capable of operating on data sets containing a limited number of data points. The new algorithm is shown to be robust in both Gaussian and non-Gaussian noise environments. Simulated and measured data, recorded on a multi-channel direction finding system, are used to demonstrate the algorithm's superior performance robustness over both the standard MVE and eigen-based techniques such as MUSIC.