Secure Proactive Model Predictive Control Based MAC Protocol for Cognitive Radio Networks

Mgr. Milan Patnaik

Master's thesis

Abstract:

Cognitive Radio (CR) is a recent network paradigm that allows Secondary Users (SUs), such as, wireless devices, to intelligently access unlicensed radio spectrum that are not allocated to them. The SUs carry out transmission without interfering with the transmission of licensed users/Primary Users (PUs) who are allocated certain dedicated portions of the radio spectrum. With the proliferation of CRs, new security threats have also come into picture. Malicious users can exploit several vulnerabilities of CR technology causing heavy per- formance degradation by Denial of Service (DoS) attacks. One such DoS attack related to the cognitive capability, where a malicious node mimics a PU, is called the Primary User Emulation Attack (PUEA). Another attack that involves transmission of false observations re- lated to spectrum sensing by a SU, is called the Spectrum Sensing Data Falsification (SSDF) attack. Many such DoS attacks exploit the spectrum sensing process in a Cognitive Radio Network (CRN). Therefore, spectrum sensing and dynamic spectrum allocation problems have been the primary concern in the field of CRs. To mitigate these challenges of spectrum sensing and dynamic spectrum allocation, feedback and learning based algorithms have been proposed in recent past. These algorithms enable the CRN to adapt itself dynam- ically to varied network changes. In this thesis, we investigate one such algorithm called the Proactive Model Predictive Control (MPC) based MAC (ProMAC). Firstly, we simulate ProMAC with an external DoS attack detection mechanism that shows significant immunity to PUEA and SSDF attacks with an insignificant reduction in performance. Secondly, to improve this performance, we propose the best time interval to invoke an external DoS attack detection mechanism for immunity against PUEA and SSDF attacks. Finally, we propose a Secure MPC based MAC protocol (S-ProMAC) that without any external DoS detection mechanism shows better performance than Pro- MAC, an external DoS attack detection mechanism aided ProMAC and a recently proposed CR MAC protocol called the Preemptive Op- portunistic MAC (PO-MAC) when compared with respect to their channel utilization, backoff rate and sensing delay.

Keywords

Cognitive Radio Network, Denial of Service, Model Predictive Con- trol, Primary User Emulation, Spectrum Sensing Data Falsification.

Language used: English

Date on which the thesis was submitted / produced: 14. 12. 2017

Thesis defence

• Date of defence: 1. 2. 2018
• Supervisor: prof. RNDr. Václav Matyáš, M.Sc.
• Reader: Ing. Mgr. et Mgr. Zdeněk Říha, Ph.D.

Masaryk University

Faculty of Informatics

Master programme / field:
Informatics / Information Technology Security (eng.)