Design and Implementation of a Real-Time Low-Latency Automatic Modulation Classifier

Automatic signal modulation classifiers are important instruments in electronic warfare, cognitive radio, or spectrum management. In this context, a feature-based (FB) automatic modulation classifier (AMC) for pulsed radar signals has been developed based on low-complexity statistical features of the signal and hierarchical decision trees (HDTs). This AMC works on a block-by-block basis. A global classification of the analyzed signal is obtained by merging the information from all the blocks. This real-time classifier runs on a field-programmable gate array (FPGA)-based platform at sampling rates exceeding 1 Gsample/s. To achieve real-time operation (clock frequency <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\approx 150$ </tex-math></inline-formula> MHz and latency <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\lt 3.5~\mu $ </tex-math></inline-formula>s) while minimizing FPGA resources, several architectural techniques, mainly based on pipelining, have been used.