Research On Detection Technology For Ultra Short Base Line Positioning System

Signal detection is an important task in underwater acoustic signal processing, and has wide applications in target acquisition and recognition, underwater communication, telemetry, ocean exploitation, and so on. The main work of this thesis focuses on the signal detection technique in USBL (Ultra-Short Base Line), including noise correlativity between different array cells of USBL system, responder signal detection technique based on polarity coincidence correlator, multi-target detection technique based on multi-channel adaptive notch filters and CFAR (Constant False Alarm Rate) processing, etc.The intensity and structure of noise field of receiving array have important influence in acoustic system performance. So if the characteristics of noise understood, combined receiving equipment designed and signal processing method chosen, the acoustic system performance can be improved. The noise spatio-temporal characteristics, restricted by the receiving noise frequency band, is given utilizing theoretical analysis and trial data, which play a active role in increasing detection performance.The signal detection for USBL system includes responder signal detection and array receiving signal detection. And the receiving circuit of responder needs both advantageous narrowband filter and low power consume performances, which can be implemented using polarity coincidence correlator. Its principle and frequency characteristic are analyzed, and evaluated using simulation and trial data processing. Multi-target detection always goes along with multi-target positioning, which may be a key technique in the future. The characteristic about parallel adaptive notch filter bank is studied and extract sine waves with different frequency at the same time.In signal detection systems, CFAR processing is used to provide the detection threshold avoiding the influences of noise and interference fluctuation, and make the false probability of target detection constant. Finally, four kinds of CFAR detectors are introduced and studied by numerical simulation in the condition of gauss noise and multi-target environment.