Location Privacy Preservation for Optimal Sensing in Cognitive Radio Networks

Graduation date: 2015Access restricted to the OSU Community, at author's request, from June 17, 2015 - June 17, 2016Cognitive Radio Networks (CRNs) enable opportunistic access to the licensed channel resources by allowing unlicensed users to exploit vacant channel opportunities. One effective technique through which unlicensed users, often referred to as Secondary Users (SUs), acquire whether a channel is vacant is cooperative spectrum sensing. Despite its effectiveness in enabling CRN access, cooperative sensing suffers from location privacy threats, merely because the sensing reports that need to be exchanged among the SUs to perform the sensing task are highly correlated to the SUs' locations.\ud In this thesis, we propose three private sensing protocols. The first scheme, Location Privacy for Optimal Sensing (LPOS) preserves the location privacy of SUs while achieving optimal sensing performance through voting-based sensing. In addition, LPOS is the only alternative among existing CRN location privacy preserving schemes (to the best of our knowledge) that ensures high privacy, achieves fault tolerance, and is robust against the highly dynamic and wireless nature of CRNs. We provide also a second variant of LPOS, that we call REP-LPOS which incorporates a reputation mechanism and uses Elliptic Curve El Gamal with Pollard lambda method to boost the decryption. The third scheme is called Public Register Private Sensing (PRPS) which is the most efficient scheme but offers lower privacy than LPOS and REP-LPOS