C44. Secure cooperative blindly-optimized compressive spectrum sensing for cognitive radio

In this paper, we propose a reduced complexity secure cooperative cognitive radio spectrum sensing algorithm for wideband systems. The proposed scheme aims to reduce the receiver hardware complexity by minimizing the required sampling rates. The proposed scheme is immune to spectrum sensing data falsification attacks. Each cognitive radio node samples the analog wideband signal from the primary users using an analog-to-information converter and sends the compressed signal to the fusion center that makes a global decision on the spectrum availability. Using the proposed authentication algorithm, the fusion center selects a limited number of trustful nodes and jointly reconstructs an estimate of the wideband signal spectrum. The proposed sensing algorithm blindly optimizes the compressive sensing sparsifying bases without prior knowledge about the sampled spectrum characteristics. Comparing the performance with previous sensing techniques shows that our proposed scheme has a better security and receiver operating characteristics at the same sampling rates.