On the Implementation of the Linearly Constrained Minimum Variance Beamformer

摘要

The linearly constrained minimum variance (LCMV) method, which allows multiple linear constraints, is an extension of the classical minimum variance distortionless response filter. Depending on the spatial filter length and the desired frequency grid, a direct computation of the resulting spatial beam pattern may be prohibitive. This brief exploits the rich structure of the LCMV expression to find a nonrecursive computationally efficient implementation of the LCMV beamformer with fixed constraints. The implementation is formed via the use of the matrix inversion lemma and the fast Fourier transform. Numerical simulations indicate a dramatic computational gain, especially for fine frequency grids and multiple constraints