Study of Byzantine Attackers and Countermeasures in Spectrum Sensing

Abstract

Cognitive radio is viewed as a novel approach for improving the utilization of the spectrum as it utilizes the under-utilized spectrum bands of the primary users. Cognitive radio is aware of its environment and can modify its transmission configuration accordingly. Spectrum sensing is the most important component of a cognitive radio network. Various types of security threats like incumbent emulation attacks and Byzantine attacks are present in the cognitive radio environment. In this thesis, we mainly concentrate on Byzantine attacks. We consider an energy detection scheme which is used to detect primary users, and the decision fusion rule to obtain a decision regarding the primary user at the fusion center. Byzantine attacker strategies and reputation based schemes to counter the attackers have been proposed. Analysis of the model is performed to assess the game between attackers and the fusion center with the help of simulations.Conclusions:Byzantine attackers and the fusion center are aware of strategies of each other. The aim of the attackers is to cause maximum damage to the system and the aim of the fusion center is to eliminate the attackers from the system and improve the system performance. Based on the performance metrics chosen, when attackers are being eliminated from the system, they tend to decrease the rate of isolation by attacking with lower probabilities such that they cannot be identified and can cause more damage to the system. At the same time, the fusion center wants to eliminate more attackers, so it tries to change the reputation threshold and time windows accordingly. The proposed algorithm is verified by conducting experiments with the help of simulations in Matlab.