An efficient set of features for pulse repetition interval modulation recognition

Pulse repetition interval (PRI) modulation recognition is one of the essential processes in Electronic Support (ES) receivers. The recognized PRI modulation type and other measured pulse train parameters usually reveal the functional purpose of the radar and are of good use in emitter identification. In this paper, a novel feature set for PRI modulation recognition is proposed. The selected features exploit both the statistical and the sequential information of pulse intervals to describe specific modulation types. After feature extraction, a relatively simple multi-layer neural network is employed for classifying different PRI modulation types. With simulations, the feature set is shown to be capable of identifying all well-known PRI types.