Artificial Neural Network Based Hybrid Spectrum Sensing Scheme for Cognitive Radio

Vyas, Maunil R, Patel, DK and Lopez-Benitez, M ORCID: 0000-0003-0526-6687 (2017) Artificial Neural Network Based Hybrid Spectrum Sensing Scheme for Cognitive Radio. In: 2017 IEEE 28th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC), 2017-10-8 - 2017-10-13, Montreal, Quebec, Canada.

Text PIMRC_2017.pdf - Author Accepted Manuscript Download (1MB)

Abstract

Spectrum sensing is a key aspect of Cognitive Radio (CR). The main requirement in CR systems is the ability to sense the primary signal accurately and rapidly. In this paper, a novel hybrid spectrum sensing scheme in CR is proposed which considers the hypothesis problem as a binary classification problem. The proposed scheme is a combination of classical energy detection, Likelihood Ratio Test statistic (LRS-G2) and Artificial Neural Network (ANN). The scheme utilises energy from energy detection and Zhang test statistic from LRS-G2 as features to train the ANN while ANN provides the adaptive learning and stable performance to the scheme. The performance of proposed sensing scheme is evaluated on several real-world primary signals of various radio technologies and it has been found out that for all those radio technologies the proposed scheme outperforms the classical energy detection and the improved energy detection.

Item Type:	Conference or Workshop Item (Unspecified)
Uncontrolled Keywords:	Cognitive Radio, Spectrum Sensing, Energy Detection, Likelihood Ratio Statistic, Artificial Neural Network
Depositing User:	Symplectic Admin
Date Deposited:	02 Oct 2017 08:51
Last Modified:	19 Jan 2023 06:53
DOI:	10.1109/PIMRC.2017.8292449
Related URLs:	Author Publisher
URI:	https://livrepository.liverpool.ac.uk/id/eprint/3009724