“When Reputation Is Currency” Scalable Blockchain to Enforce Spectrum Policies

Abstract

Distributed enforcement of spectrum policies requires fusion of sensing results from a set of spatially scattered sensors to detect and localize unknown transmitters with the highest possible accuracy. Central to this problem is the lack of trust and reputation of the participating sensors and the challenge in scaling to larger geographical areas, which often leads to delayed, biased, and localized inference. We propose a distributed fusion algorithm leveraging the core principles of Blockchain networks to create an immutable ledger of historical sensing activity. Unlike other approaches, we integrate a trust evaluation method to determine the veracity of the sensor reports in order to calculate the reputation that reflects the perceived value of a sensorfs measurement in the information fusion process. Along with a computation-friendly Proof-of-Work method and a lowoverhead consensus algorithm, the proposed Blockchain provides provenance for the sensorsf past activities that can be verified as they move over a large geographical region. Evaluation using an integrated Blockchain and trust assessment simulator shows that using reputation-based fusion, spectrum policy infractions are detected with a consistent accuracy of >90% and are localized within a region of radius.35m even when 80% of the sensor reports are maliciously altered, in <10sec on average