Adaptive Detection of an Unknown FH Signal Based on Image Features

We introduce an adaptive detection structure for unknown frequency hopping (FH) signal. After getting the short time Fourier transform (STFT) of received signal, we process the time-frequency spectrogram in terms of image operations, and extract multiple image features. Two kinds of method are employed for combinative detection. We firstly establish the FH signal verification and joint parameters estimation (FHSVJPE) model based on these image features, and estimate some FH parameters while resolving the model. The FH detection threshold used in the model can be adjusted adaptively according to complicated time varying environments. Then we propose a time frequency (TF) matching method to further detect the jammed hops and estimate other FH parameters. Simulation results demonstrate that this detection algorithm is feasible and efficient.