Time-Frequency DOA Estimate Algorithm Based on SPWVD
Hu Hang · 2006
In studies on DOA estimate based on time-frequency analysis, for the case of multi-signal sources, eliminating cross interference terms is an important problem. It is the key to improve DOA estimate performance. WVD is the most commonly used time-frequency analysis tool, but it has a very big bug that it will generate cross interference terms in case of multi-signal sources. So PWVD is used to suppress cross terms in many cases. However, the effect of PWVD suppressing cross terms doesn't be so ideal and the estimated results are not so reliable. This paper employ SPWVD which effectively suppress cross terms by using smoothing for two times in both time domain and frequency domain, consequently it improves the accuracy of DOA estimate much more greatly. Simulation results demonstrate in the background of multi sources the performance of DOA estimate algorithm based on SPWVD improves compared with that based on PWVD.