Weight and layout optimized arrays using least angle regression

We utilize a variable selection method called least angle regression for finding weight and layout optimized sparse wideband arrays. As opposed to previously reported-methods for finding sparse arrays, the proposed method is attractive in that it optimizes both weights and layout simultaneously, and finds a solution in bounded time proportional to the number of elements and excitation samples of the array. In the case of no sparsing the method yields weights whose beampattern is equal to the least squares solution. For a 64-element linear array, with parameters based on a realistic medical ultrasound imaging system, a 6 dB improvement in the peak sidelobe level was observed for an array sparsed to 48 elements compared to the fully sampled unity-weighted array.