Multiple Signal Classification for Determining Direction of Arrival of Frequency Hopping Spread Spectrum Signals.

摘要

This research extends a MUSIC algorithm to determine DOA of FHSS signals. All incident FHSS signals have unknown DOA and use PSK. Conventional MUSIC algorithm involves multiple MUSIC estimation for all frequency bins. On the other hand, the extended development is meant to execute a single MUSIC algorithm of observations on multiple frequency bins or hops. The new extension shows better performance compared to the conventional MUSIC execution at di erent SNR levels. Both have the same power accumulation at the true angles of arrival. However, the new development has lower side lobes and hence helps avoid false detections. In addition, the new development has lower side lobes variance resulting in lower error of false detections compared to the normal execution. Simulation results show that the new extension is sensitive to the SNR values and number of samples taken at each frequency bin. However, it is less sensitive to the possible number of frequency hops or hop set and number of array sensors.