Classification of communication signals using an optimized classifier and efficient features

Abstract EN

Automatic classification of communication signals plays an important role in various applications.

This paper presents an efficient method that classifies a variety of communication signals.

In this method, a classifier based on a support vector machine is proposed as the multiclass classifier.

A combination set of the higher order statistics are proposed as the effective features for representation of the different kinds of communication signals.

An optimization algorithm is used for selecting the parameters of the classifier.

Simulation results show that the proposed recognizer has high performance even at very low SNRs.