Novel modulation detection scheme for underwater acoustic communication signal through short-time detailed cyclostationary features

In this paper, we propose a new method for blind modulation detection based on detailed second order cyclostationary features for underwater acoustic communication. Due to the severe and fast varying Doppler shift and phase noise in the underwater acoustic communication channel, much shorter signal length can be used to detection the modulation scheme, leading to less accuracy in the modulation detection performance. Here we propose a brand new method to perform modulation detection: by analyzing the zoomed-in spectral coherence function (SOF) of short time signal, we observe distinctive detailed features for different modulations. Taking advantage of such detailed features in short time SOF, we design a novel blind modulation detection algorithm for underwater acoustic communication system. To the best of our knowledge, our paper is the first to employ such detailed, in some sense hidden, features of cyclostationary analysis to conduct modulation detection. Coupled with our previous work on blind carrier frequency estimation and symbol rate estimation algorithms, the proposed modulation detection algorithm enjoys very high detection accuracy, high speed, and low complexity. Real experimental data collected at sea are used to validate the effectiveness of the proposed algorithm. It is also important to note that the proposed method does not assume any a priori knowledge of the target signal.