Abstract:

In order to estimate the frequency and DOA of an incident signal with sub-Nyquist spatio-temporal sampling,A novel joint estimation algorithm based on spectrum correction and Chinese remainder theorem(CRT)was proposed.Firstly,the incident signal was parallelly sampled at multiple sensors non-uniformly located in a sparse linear array.Then,AM spectrum corrector was employed to extract the rema of frequency and phase-difference from a small quantity of snapshots.Finally,implementing the close-form CRT on these remainders directly yields the estimates of frequency and DOA.In the proposed scheme,the spectrum corrector was shared by both the frequency estimator and the DOA es-timator,which contributed to a high utilization of the snapshots.Furthermore,compared to the existing CRT based esti-mators,the proposed estimator needs not to incorporate multiple times of temporal undersampling on a single sensor,which saves considerable time consumption and thus obtain high ability of tracking time-variant objects.Simulations verify the feasibility and the high-accuracy of the proposed joint estimator and further study shows that high reconstruc-tion probability can be realized even in low SNR cases,which present the vast potential for future development.

Key words: direction of arrival, frequency estimation, undersampl g, Chinese remainder theorem