Performance analysis of ambiguity function based broadband chirp direction of arrival estimation

Abstract

When estimating the direction of arrival (DOA) estimation of broadband sources with an array, the source is usually modeled as white noise. However, when the source signal has certain time-frequency structure, studies have shown that the DOA estimation performance can be improved by using the information about its structure. In our previous work [Ning Ma and Joo Thiam Goh, ‘‘DOA Estimation for Broadband Chirp Signals,’’ Proc. of ICASSP 2004, Vol. II, pp. 261–264], we have proposed two new methods for broadband chirp signal DOA estimation based on the ambiguity functions, namely Incoherent Broadband Chirp DOA estimation (CBD-I) and Coherent Broadband Chirp DOA estimation (BCD-C). Besides improving the accuracy of the DOA estimates, these algorithms work with signal frequencies that are higher than the array design frequency. The number of sources are also not constrained by the number of sensors. The performance of the proposed DOA estimation methods is analyzed theoretically and numerically in this work. The analysis results show that these methods can improve the output SNR and therefore the DOA estimation performance when input SNR is higher than certain level.