New Method Based on Combination of the DAS and Min Processors for Spatial Spectral Estimation with Co-Prime Arrays

Abstract

The Co-prime Sensor Array (CSA) is a sparse array geometry that use two nonlinear product and min processors to estimate the spatial power spectral density (PSD). These processors achieve an asymptotically unbiased PSD estimate using far fewer sensors than a fully populated uniform linear array (full ULA), but at the cost of increased side lobe levels (SLLs). The SLLs can be reduced by adding more sensors to the CSA while keeping the inter-sensor spacing fixed, so-called extended CSA (ECSA) geometry. The ECSA is not feasible when the size of array is limited by the costs or the environmental constraints. In this paper, we develop a new method based on combination of the conventional delay and sum (DAS) processor and nonlinear min processor to reduce the SLLs of the conventional CSA. Therefore, in many practical applications where the number of sensors is limited, the proposed method would be very useful. The theoretical and simulation results show that the proposed processor achieves lower peak side lobe level (PSL) and integrated side lobe level (ISL) than the previous processors while approaching the directivity of the DAS processor.