A Privacy Preserving Mobile Crowdsensing Architecture for a Smart Farming Application

Abstract

Smart Farming refers to the act of utilizing modern information and sensor technology in conventional industrial farming. An important plant parameter that can be estimated by sensor technology in the context of Smart Farming is the leaf area index (LAI) which is a key variable used to model processes such as photosynthesis and evapotranspiration. Nowadays, leveraging the enhanced sensor peripherals of current devices and their computing capabilities, smartphone applications present a fast and economical alternative to estimate the LAI compared to traditional methods. This paper exemplarily extends such an application, namely Smart fLAIr, with features of Mobile Crowdsensing (MCS) in order to create a system for a crowd-sensed LAI enabling an increased spatio-temporal resolution of LAI estimations. Besides the system design, this paper conducts a threat analysis for user privacy in the application-specific scenario which can be transferred to general Smart Farming scenarios. As a consequence, a perturbation based privacy mechanism is developed and applied in conjunction with a Trusted Third Party (TTP) architecture to ensure user privacy. Subsequently, its impact is demonstrated. Moreover, the energy consumption of the extended Smart fLAIr application is evaluated showing negligible additional costs of the proposed MCS extension.