This research presents the development and performance analysis of space-time adaptive processing(STAP) algorithms for circular array radars. The algorithms developed here result in reduced training data support and computational requirements utilizing a priori information regarding the structure underlying the interference. Because these algorithms are based on interference consisting of a small number of strong low rank interferers plus white noise, they permit development of powerful data matrix-based methods that also offer potential for considerably reduced training data support and computational complexity. The algorithms afford the capability of a constant false alarm rate(CFAR) and permit the design of robust receiver structures. Performance analysis using realistic simulated data and real data from experiments using a UESA circular array is presented. Simulation techniques for data generation from an idealized circular array and a practical circular array, including relevant radar parameters, are developed. The algorithms resulting from this research permit rapid localization of dominant interference subspaces, enabling fast adaptation intervals which are fractions of a coherent processing interval(CPI). Consequently, these algorithms are suitable for use in severely non-homogeneous and nonstationary interference backgrounds. This work outlines the differences between uniform linear arrays and circular arrays and discusses the resulting impact on STAP signal processing.
Keywords: Circular Array, Stap, Cpi, Subset Selection, Cfar, Robust Receiver, Low Rank, Dominant Interference
