Pulse repetition interval classification based on decimated walsh-hadamard transform for electronic intelligence

Pulse Repetition Interval (PRI) classification is conducted in Electronic Intelligence (ELINT) to identify radar system based on the intercepted signal. By classifying the PRI according to their types, ELINT analysis can recognize the functional of radar system. However, in PRI classification, missing and spurious pulses can cause misclassification in PRI types which can lead to wrong identification of radar signal. Typical or common PRI types like constant, jittered and staggered are among the misclassified. Therefore, new method of PRI classification is developed based on decimated Walsh-Hadamard transform (DWHT). Decimation process reduces the sample to limit aliasing signal and noise which contribute to the misclassification. Template for constant, jittered and staggered PRI is developed based on the power
spectrum of WHT. Simulated signal to represent transmitted radar signal is used to test the developed method as it would be seen in the ELINT receiver. Gaussian noise is added to simulate noise and clutter. The DWHT template is used as classifier to classify the PRI types accordingly. For comparison with an established method, PRI classification based on statistical histogram is developed. Comparison between DHWT and histogram method shows DWHT has 9% to 30% higher accuracy than histogram depending on the PRI types. DWHT can deliver more than 86% accuracy in PRI classification even with the presence of missing and spurious pulses.