Modulation Signal Recognition Based on Information Entropy and Ensemble Learning

Abstract

Abstract:In this paper, information entropy and ensemble learning based signal recognition theory and algorithms have been proposed. We have extracted 16 kinds of entropy features out of 9 types of modulated signals. The types of information entropy used are numerous, including Rényi entropy and energy entropy based on S Transform and Generalized S Transform. We have used three feature selection algorithms, including sequence forward selection (SFS), sequence forward floating selection (SFFS) and RELIEF-F to select the optimal feature subset from 16 entropy features. We use five classifiers, including k-nearest neighbor (KNN), support vector machine (SVM), Adaboost, Gradient Boosting Decision Tree (GBDT) and eXtreme Gradient Boosting (XGBoost) to classify the original feature set and the feature subsets selected by different feature selection algorithms. The simulation results show that the feature subsets selected by SFS and SFFS algorithms are the best, with a 48% increase in recognition rate over the original feature set when using KNN classifier and a 34% increase when using SVM classifier. For the other three classifiers, the original feature set can achieve the best recognition performance. The XGBoost classifier has the best recognition performance, the overall recognition rate is 97.74% and the recognition rate can reach 82% when the signal to noise ratio (SNR) is −10 dB.

Automatic Modulation Classification Based on Novel Feature Extraction Algorithms

“…In [3], a new feature is constructed based on the fourth and sixth order cumulants and a neural network classifier is designed to classify nine kinds of signals including MSK and MQAM signals. More recently, authors combine the information entropy features of the signal with the feature selection algorithms and use five different ensemble learning classifiers to complete the classification of a variety of digital modulation signals [4]. In another work, compressive temporal higher order cyclostationary statistics is proposed in [5].…”

Few-Shot Modulation Classification Method Based on Feature Dimension Reduction and Pseudo-Label Training

“…In traditional handcrafted feature based methods domain, the authors of [1]- [3] classified modulation signals by using high-order cumulants, and the authors of [4], [5] classified signals by extracting cyclic spectrum features. The authors of [6], [7] used information entropy features for AMC. In deep learning based methods domain, authors summarized the typical AMC methods based on deep learning in recent years in [8].…”

Emitter signals modulation recognition based on discriminative projection and collaborative representation

“…With the development of signal processing and pattern recognition technology, non‐linear feature transformation (fuzzy function, wavelet transformation etc.) and machine learning technology has been widely applied in this field and obtains a lot of achievements [1–6]. In recent years, as motivated by big data and strong computing capacity, the emitter signal recognition based on artificial intelligence (AI) technology has gradually become a hot topic.…”