Active sonar detection using adaptive time-frequency feature

In the complex variable shallow water acoustic channel, the regular signal emitted by the active sonar would propagate through multiple paths and scattered by the target. Compared with the transmitted signal, the echo signal has a time delay and Doppler spread. This leads to a great decrease of the correlation between the copy vector of the matched filter and the target echo, which degrades target detection and estimation performance. This paper uses the time-frequency feature extracted from the target echo to adaptively weight copy vector of the matched filter. The target detection performance of the active sonar is improved by solving the mismatch between original transmitted signal and target echo.