Blind signal-type classification using a novel robust feature subset selection method and neural network classifier

Abstract

Automatic modulation recognition plays an important role for many novel computer and communication technologies. Most of the proposed systems can only identify a few kinds of digital signal and/or low order of them. They usually require high levels of signal-to-noise ratio. In this paper, we present a novel hybrid intelligent system that automatically recognizes a variety of digital signals. In this recognizer, a multilayer perceptron neural network with resilient back propagation learning algorithm is proposed as the classifier. For the first time, a combination set of spectral features and higher order moments up to eighth and higher order cumulants up to eighth are proposed as the effective features. Then we have optimized the classifier design by bees algorithm (BA) for selection of the best features that are fed to the classifier. This optimization method is new for this area. Simulation results show that the proposed technique has very high recognition accuracy with seven features selected by BA.

A Two-Stage Hybrid Feature Selection Method of Communication Signals Based on Combined Scoring and Improved Binary Whale Optimization Algorithm

“…In [21], rough set (RST) was used to extract robust features from high-dimensional signal feature set in order to reduce the training and testing time of the classifier, but there are still some redundant features after rough set reduction. [22] used the Bee Algorithm (BA) to optimize the input features of the classifier, and [23]…”