Malicious User Detection in Spectrum Sensing for WRAN Using Different Outliers Detection Techniques

TL;DR

This paper investigates various outlier detection techniques to identify malicious secondary users in spectrum sensing for cognitive radio networks, considering realistic environmental conditions like fading and noise.

Contribution

It evaluates and compares different outlier detection methods to determine the most effective approach for malicious user detection in practical spectrum sensing scenarios.

Findings

Certain outlier detection techniques outperform others under fading conditions

The study identifies the most suitable method for real-time malicious user detection

Environmental factors like noise significantly impact detection accuracy

Abstract

In cognitive radio it is of prime importance that the presence of Primary Users (PU) is detected correctly at each of the time. In order to do so the help from all present Secondary Users (SU) is taken and such a taken is known as co-operative spectrum sensing. Ideally it is assumed that all the secondary users give the correct result to the control center. But there are certain conditions under which the secondary users deliberately forward wrong result to the control center so as to degrade the performance of the cognitive network. In this paper we study the different techniques for detecting the malicious users or outliers. We take into consideration practical environmental condition such that the received signal of the secondary users is made to undergo fading and noise is also introduced in the signal. We further go on to examine each of the outlier detector techniques and find out the most suitable at various instants.