Detection performance analysis of sub-band processing continuous active sonar

Abstract

As compared to conventional pulsed active sonar (PAS), continuous active sonar (CAS) significantly improves the time-bandwidth product(processing gain) and targets information update rate. However, it is difficult to estimate how much better CAS performs than PAS in terms of detection. In addition, it is not easy to design the parameters, such as the cycle of continuous waveforms and the duration of each sub-pulse. In this work, we use frequency-hopping LFM (FH-LFM) waveforms for designing CAS transmitting waveforms. Moreover, we also apply sub-band processing to detect echoes. In order to improve detection performance, the matched filtering results of all sub-pulses are delayed sequentially and processed by noncoherent integration. The detection and false alarm probability of noncoherent integration are analyzed, and the receiver operating characteristic (ROC) curves of CAS and PAS are compared. The results make it evident that how much better the detection performance of CAS is than that of PAS. Furthermore, with the help of PAS design experiences, we determine the appropriate parameters of CAS and evaluate the detection performance.