PULSE CUMULATING METHOD FOR MULTI-TARGETS' JOINT AZIMUTH-ELEVATION FREQUENCY-DELAY ESTIMATION

The Directions of Arrivals （DOAs）, speeds and distances of targets are all required for array signal processing. Based on the periodic phase shift of coherent pulse sequence waveform, a new estimation of multi-targets＇ 2-Dimentional （2-D） DOA angle, Doppler frequency shift and relative time-delay is proposed. Based on a virtual sensor array constructed by pulse cumulating, the estinaations of azimuth, elevation, Doppler frequency shift and time-delay can be obtained simultaneously, and the least number of pulses could be two. This method is computationally efficient even in heavier noised environment, and all estimations are automatically paired in calculation process with no used to any plane sensor array and deal with many spectrum searching. Further more, this algorithm can be targets at the same time only by few sensors. The targets number that can deal with simultaneously is several times to the sensor number, which is the upper limit for normal algorithms such as ESPRIT and MUSIC. These characteristics would be very useful, especially, for aerial systems. Simulations demonstrate the capabilities of this method efficiently.

The Directions of Arrivals (DOAs), speeds and distances of targets are all required for array signal processing. Based on the periodic phase shift of coherent pulse sequence waveform, a new estimation of multi-targets' 2-Dimentional (2-D) DOA angle, Doppler frequency shift and relative time-delay is proposed. Based on a virtual sensor array constructed by pulse cumulating, the estimations of azimuth, elevation, Doppier frequency shift and time-delay can be obtained simultaneously, and the least number of pulses could be two. This method is computationally efficient even in heavier noised environment, and all estimations are automatically paired in calculation process with no spectrum searching. Further more, this algorithm can be used to any plane sensor array and deal with many targets at the same time only by few sensors. The targets number that can deal with simultaneously is several times to the sensor number, which is the upper limit for normal algorithms such as ESPRIT and MUSIC. These characteristics would be very useful, especially, for aerial systems. Simulations demonstrate the capabilities of this method efficiently.