Sidelobe suppression in correlated multipath estimates

Abstract

Some environments cause a propagating signal to split along several different paths. When such multipath propagation occurs, the covariance among signals traveling along rays emanating from a common source is expected to be larger than the covariance between signals generated by independent sources. An estimate of the covariance between signals arriving from two different directions is shown to be a bilinear form. The ability of the bilinear form to distinguish a correlated arrival from an independent source is studied using a simulated acoustic field. Peaks associated with the correlated pair of arrivals are 24 dB above an uncorrelated background, while no similar peaks occur for the independent source. The 8‐dB sidelobes are found to be a problem. An adaptive set of filter vectors, obtained from the classical minimum variance problem, produces a bilinear form which is found to minimize sidelobe interference to 2 dB. However, the peaks indicating a correlated arrival are reduced to 18‐dB output above the uncorrelated background. The constrained minimum variance estimate of the covariance between two arrivals of a multipath is apparently effective in suppressing sidelobe interference at the cost of a lower main peak. [Work supported by ONR.]