Fast-converging maximum-likelihood interference cancellation

Abstract

The use of adaptive linear techniques to solve signal processing problems is needed particularly when the environment external to the radar is not known a priori. Due to the lack of knowledge of the external environment, adaptive techniques require a certain amount of data to cancel the interference external to the radar or communication system. The amount of data required so that the performance of the adaptive processor is close (nominally within 3 dB) to the optimum is called the convergence time of the processor. The minimization of the convergence time is important since in many environments the external interference changes with time. Although there are heuristic techniques such as the loaded sample matrix inversion (LSMI) in the literature that provide fast convergence for particular problems, there is currently not a general solution for arbitrary interference that is derived via classical theory. In this paper, the authors derive a maximum-likelihood (ML) solution for a structured covariance matrix. The only structure assumed is that the thermal noise variance is known. Using this ML estimate, simulations show that the convergence is on the order of twice the number of narrow-band interference sources.