Block Outlier Methods for Malicious User Detection in Cooperative Spectrum Sensing
Abstract
Block outlier detection methods, based on Tietjen-Moore (TM) and Shapiro-Wilk (SW) tests, are proposed to detect and suppress spectrum sensing data falsification (SSDF) attacks by malicious users in cooperative spectrum sensing. First, we consider basic and statistical SSDF attacks, where the malicious users attack independently. Then we propose a new SSDF attack, which involves cooperation among malicious users by masking. In practice, the number of malicious users is unknown. Thus, it is necessary to estimate the number of malicious users, which is found using clustering and largest gap method. However, we show using Monte Carlo simulations that, these methods fail to estimate the exact number of malicious users when they cooperate. To overcome this, we propose a modified largest gap method.
Cite
@article{arxiv.1402.4109,
title = {Block Outlier Methods for Malicious User Detection in Cooperative Spectrum Sensing},
author = {Sanket S. Kalamkar and Praveen Kumar Singh and Adrish Banerjee},
journal= {arXiv preprint arXiv:1402.4109},
year = {2016}
}
Comments
Accepted in Proceedings of 79th IEEE Vehicular Technology Conference-Spring (VTC-Spring), May 2014, Seoul, South Korea