Location-Based Lightweight Security Scheme for Wireless Communications in ROAR Architecture

Abstract

Dynamic spectrum access is considered to be an emerging technology for enhancing RF spectrum efficiency where secondary unlicensed users access primacy licensed RF bands in an opportunistic manner for wireless communications. In dynamic spectrum access, RF spectrum geolocation database-driven approach has relieved the wireless devices from uncertainties caused by spectrum sensing at low signal levels. However, because of the openness in dynamic spectrum access environment, malicious users could mislead the overall communication resulting in no wireless services to legitimate secondary users. For instance, when secondary user queries the RF spectrum database, it sends a query along with its geolocation which could be intercepted and hijacked by the malicious users and block the secondary users from using available RF spectrum. To address this problem, we have developed a lightweight security technique where hash of a geolocation of the secondary user is used as a part of an encryption key to provide security in the Real-time Opportunistic Spectrum Access in Cloud-assisted Cognitive Radio Networks (ROAR) architecture. The proposed location-based lightweight security scheme leverages the secondary user's location (latitude, longitude and altitude) information to encrypt the data being transmitted to protect the system from possible malicious attacks. This approach has two stages: first, share the unique hash value generated using location information (i.e., longitude, latitude and altitude) secretly with the help of a pre-shared string with the ROAR's geolocation spectrum server. Second, the secondary users use their hash values obtained by using location information to encrypt the message being transmitted to their corresponding receivers. A key feature of this approach is that if a third party (malicious user) tries to send data to the geolocation spectrum database for querying the RF spectrum database, malicious actions will be detected and malicious users would not get the service for dynamic spectrum access. We evaluate the proposed approach using Monte Carlo simulations and experiments using software defined radios in ROAR architecture.