Ensemble Classifier Based Spectrum Sensing in Cognitive Radio Networks

Abstract

Spectrum sensing is one of the most important and challenging tasks in cognitive radio. To develop methods of dynamic spectrum access, robust and efficient spectrum sensors are required. For most of these sensors, the main constraints are the lack of information about the primary user's (PU) signal, high computational cost, performance limits in low signal-to-noise ratio (SNR) conditions, and difficulty in finding a detection threshold. This paper proposes a machine learning based novel detection method to overcome these limits. To address the first constraint, detection is achieved using cyclostationary features. The constraints of low SNR, finding detection threshold, and computational cost are addressed by proposing an ensemble classifier. First, a dataset is generated containing different orthogonal frequency-division multiplexing signals at different SNRs. Then, cyclostationary features are extracted using FFT accumulation method. Finally, the proposed ensemble classifier has been trained using the extracted features to detect PU's signal in low SNR conditions. This ensemble classifier is based on decision trees and AdaBoost algorithm. A comparison of the proposed classifier with another machine learning classifier, namely, support vector machine (SVM), is presented, clearly showing that the ensemble classifier outperforms SVM. The results of the simulation also prove the robustness and superior efficiency of the detector proposed in this paper in comparison with a cyclostationary detector without machine learning as well as the classical energy detector.

Spectrum Sensing for Cognitive Radio: Recent Advances and Future Challenge

“…Local sensing corresponds to the case where a single node senses the spectrum and makes its own decision. In this context, the work of [ 95 ] seeks to discriminate between and hypotheses by being trained with the extracted cyclic features of PU’s signal in low SNR conditions. This ensemble classifier is based on decision trees and using AdaBoost algorithm [ 96 ].…”

Sixth Generation (6G) Cognitive Radio Network (CRN) Application, Requirements, Security Issues, and Key Challenges

“…Combining domain information into the DL architectures embraces huge potential to rush the training process and model junction though attractive the model effectiveness. 3 Wireless Communications and Mobile Computing consumption for WSN is vital since under protocols nodes are working for many years extension of a lifetime [29].…”

AI Meets CRNs: A Prospective Review on the Application of Deep Architectures in Spectrum Management

“…DL methods, through the use of DNNs, are able to automatically extract high-level features through the layers of different depths from data that have complex structures and inner correlations. The importance of feature extraction and pattern recognition can be amplified in the context of spectrum sensing for PU signal classification, where the pattern of signals generated by heterogeneous sources can easily be recognized [143]. Since some signals are often noisy and usually exhibit some non-trivial spatial/temporal patterns, labeling them using feature engineering would require outstanding human effort and skills [144], DL would prove beneficial for this task.…”