A COOPERATIVE SPECTRUM SENSING SCHEME IN WIRELESS COGNITIVE SENSOR NETWORK

摘要

Cognitive radio technology has experienced a fast development since its basic concept was raised by Joseph Mitola. And wireless cognitive sensor network is a new application field of Cognitive Radio based on the traditional wireless sensor network. The senor node is assigned the ability of spectrum sensing, data analysis and parameter adjusting and can function as a cognitive node. On the one side, the cognitive sensor node can sense the spectrum environment, detect the primary users and take use of the vacant spectrum. On the other side, the new functions of sensor node arouse some energy and security concerns. Malicious sensor node which reports false information about the spectrum may easily exhaust the nodes energy and break down the whole sensor network. To deal with the problems, some researches need to be done to improve the spectrum cognitive accuracy. In this paper, we investigate schemes to identify the malicious users for cooperative sensing system. Then a novel method is proposed to reduce the interference from false data and indentify the malicious nodes with the concept of bi-weight, outlier factor and penalty factor. The performance of the proposed scheme is studied using simulations.