Trust-based cooperative spectrum sensing against SSDF attacks in distributed cognitive radio networks

We propose and analyze a trust-based data fusion scheme against spectrum sensing data falsification attacks in a distributed cognitive radio network. Our trust-based data fusion scheme is based on mechanism design theory to motivate users to report authentic sensing data so as to improve the success rate. Further, we decouple erroneous sensing reports due to low sensing capabilities from false reports due to attacks, thus avoiding unnecessary punishments to benign users. We conduct a theoretical analysis validated with extensive simulation and identify optimal parameter settings under which our trust-based data fusion scheme outperforms existing non-trust based data fusion schemes.