Multiple source detection performance of linear sparse arrays

This paper considers detection performance of linear sparse arrays in the presence of multiple source Direction-Of-Arrivals (DOA). A number of source DOAs greater than the number of physical array elements are randomly generated, without restriction on the angular separation. Two augmentation algorithms are considered before application of traditional spatial spectrum estimation algorithms: MVDR and MUSIC. Detection and false alarm statistics are generated from the spatial spectrum. MUSIC requires the knowledge on the number of source, which can be estimated from existing source number estimation algorithms. The effect of source number estimation errors on the detection performance is evaluated. Monte Carlo simulations are performed to validate this detection scheme.