Co-Prime Arrays with Reduced Sensors (CARS) for Direction-Of-Arrival Estimation

Mingyang Chen,Lu Gan,Wenwu Wang

DOI: https://doi.org/10.1109/sspd.2017.8233226

2017-01-01

Abstract:In traditional direction-of-arrival (DoA) estimation, the number of sources to be detected is widely assumed to be smaller than the number of physical sensors. In practice, however, the number of sources encountered may be greater than that of the sensors. To address this problem, nested array structure can be considered, however, the effects of mutual coupling between sensors may become significant when the sensors are located closely. Recently, the co-prime array structure has been developed to further enhance the uniform degrees-of-freedom (DoF) with less mutual coupling, by utilizing a co-prime pair of uniform linear subarrays. This paper proposes a pair of co-prime arrays with reduced sensors (CARS) structure, where the first array is shifted until symmetrical to the center (reference sensor) and the number of sensors in the second array can be reduced as compared to existing optimized co-prime array configurations, according to the odd and even of the co-prime number pair. Furthermore, with the CARS structure, much more uniform DoF can be detected. The feasibility of the proposed CARS structure is demonstrated by numerical results of DoA estimations for multiple stationary sources with noise.