Mitigating Constant Jamming in Cognitive Radio Networks using Hybrid FEC Code Victor Balogun Computer Science Department University of Idaho Moscow, ID, U.S.A. Email:balo8072@vandals.uidaho.edul Axel Krings Computer Science Department University of Idaho Moscow, ID, U.S.A. Email:krings@uidaho.edu Abstract—The task of detecting and mitigating Cognitive Radios (CRs) operating as constant jammers in a Cognitive Radio network (CRN) can be very daunting. The CR constant jammers prey on the adaptable functionalities of CRN so as to cause serious denial of service to the users of the network. In addition, these jammers are capable of introducing value faults in pathological cases as a result of being able to manipulate transmitted data. In previous research, we have investigated the performance of CRNs operating in the presence of jamming attacks that are capable of introducing value faults. The results obtained show that CR constant jammers are very effective in their operations and they are capable of bringing down the entire CRN when their jamming rate is just about 30%. In this paper, we show that none of the existing anti-jamming solutions are able to mitigate CR jammers that are capable of manipulating communicated data. As a result, we propose a hybrid forward error correction (FEC) code that incorporates data integrity checking into an efficient forward error correction mechanism. The approach uses data redundancy to remove the need for retransmission of lost or detected manipulated data caused by jamming attacks by exploring the recovery block component of the proposed solution. We present the algorithm for our proposed solution and evaluate its performance through simulation. The result of our analysis using suitable performance metrics shows that the solution is very efficient and robust against the different rates of jamming perpetrated by constant jammers. I. I NTRODUCTION AND BACKGROUND The Cognitive Radio Network (CRN) has been adopted as a technology that will alleviate the spectrum shortage problem [1]. The technology allows unlicensed users to share the unused part of the licensed bands of spectrum with the incumbents when present (Underlay approach) or during the incumbents’ OFF period (Overlay approach). The benefits of CRNs are broad and range from providing low cost Internet connectivity to providing easy access to government-delivered services to rural and underserved regions. Accurate detection of the presence of the incumbents by CRs is a crucial issue as the CRN standard does not tolerate any level of interference to the incumbent’s network. Channel impairments like deep shadowing [16] and multi- path fading [14] are some of the barriers to adequate sensing of the spectrum for the presence of an incumbent by a CR. Therefore, a CR needs the assistance of other CRs operating in its neighborhood to be able to accurately sense the spectrum. This is called cooperative spectrum sensing (CSS). Different CSS architectures were proposed in [2] and [3]. Despite the expected success and potentials of CSS, the presence of jam- mers operating as CRs in a CSS Cognitive Radio network is of great security concern. This is because the CR jammers have reconfigurable features that make them capable of introducing value faults along the transmission channels. The legitimate CR nodes themselves are easy prey to such attackers as they could be manipulated by their reconfigurable features so as to cause serious DoS to the entire network. In order to represent CRN as a network that is capable of exhibiting different types of faults including value faults, the research in [3] presents different jamming scenarios in a CRN based on a hybrid fault model that identifies the possibility of different fault types including omissive and transmissive value faults. The impact of these jammers have been investigated in [4] and the effect of CR jamming, especially the case of constant jammers on the average throughput of the network is shown to be huge. Therefore, the need to design a strategy that is capable of mitigating this category of CR jammers is of great importance. To the best of our knowledge, none of the existing anti- jamming mechanism is capable of effectively mitigating CR jammers that are able to manipulate transmitted data in CRNs. Based on the hybrid fault model classification in [3] and the perceived impacts of jamming on this model as quantified in [4], the main contribution of this paper is a new hybrid forward error correction (FEC) code that is capable of mitigating jamming attacks under this classification. Furthermore, we present the algorithmic design of the proposed solution, its implementation in Network Simulator 2 (NS-2) and the anal- ysis of its performance measurement using suitable metrics. 1) Anti-jamming Strategies for CR Networks : Several authors [5], [6] have investigated jamming in Cognitive Radio Networks. Some of these authors have proposed the use of Spread Spectrum Frequency Hopping (FH) and Direct Se- quence Spread Spectrum (DSSS) as a solution to alleviate jamming attacks in wireless networks. The popular approach is to spread the signal over a larger bandwidth, thereby making it costly for jammers to hinder an on-going transmission. The combination of Spread Spectrum and Orthogonal Frequency Division Multiplexing [7] was also considered as an efficient means of mitigating jamming attacks in wireless networks. In CRNs based on the Spread Spectrum approach [7], the avail- able spectrum is divided into several pieces of non-overlapping channels in which only a small portion of the channels is used for transmission at a time. The malicious jammer either jams a large number of the channels with negligible jamming