The performance of adaptive beamforming methods may degrade in the presence of even slight mismatches between the actual and presumed array responses to the desired signal. This paper addresses the problem of robust adaptive beamforming in the presence of unknown arbitrary (yet norm-bounded) mismatches of such type as well as interference-plus-noise covariance matrix mismatch. Our approach is developed for the case of an arbitrary dimension of the signal subspace and, therefore, it can be applied to both rank-one and higher-rank signal models. The proposed beamformer is based on the optimization of the worst-case signal-to-interference-plus-noise ratio (SINR). The obtained closed-form solution combines two different types of diagonal loading (DL) applied to the signal and data covariance matrices. An efficient on-line implementation of our beamformer is developed. Simulations validate substantial performance improvements relative to other popular adaptive beamforming techniques.
|Original language||English (US)|
|Number of pages||4|
|Journal||ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings|
|State||Published - Sep 25 2003|
|Event||2003 IEEE International Conference on Accoustics, Speech, and Signal Processing - Hong Kong, Hong Kong|
Duration: Apr 6 2003 → Apr 10 2003