Spass: Spectrum Sensing as a Service via Smart Contracts

Mobile network operators can expand their capacity by aggregating their licensed spectrum with the spectrum discovered opportunistically, i.e., spatiotemporally unused spectrum by other primary users. For an accurate identification of the spectral opportunities, the mobile network has to deploy multiple sensors or it can offload this task to nearby nodes with sensing capabilities, so called helpers. Unfortunately, incentives are limited for helpers to perform energy-wasteful spectrum sensing. Instead, we envision spectrum sensing as a service (Spass) in which a smart contract running on a blockchain (BC) describes the required sensing service parameters and the contracted helpers receive payments only if they perform sensing accurately as agreed in the contract. In this paper, we first introduce Spass and derive a closed formula defining the profitability of a Spass-based business as a function of the spectral efficiency, cost of helpers, and cost of the service. Moreover, we propose two-threshold based voting (TTBV) algorithm to ensure that the fraudulent helpers are excluded from Spass. Via numerical analysis, we show that TTBV causes almost zero false alarms and can exclude malicious users from the contract after only a few iterations. Finally, we develop a running prototype of Spass on Ethereum BC and share the related source code on a publicly-available repository.