Location-Aware Privacy Preserving Scheme in SDN-Enabled Fog Computing

Fog computing, as a novel computing paradigm, aims at alleviating data loads of cloud computing and brings computing resources closer to end users. This is achieved through fog nodes such as access points, sensors, and fog servers. According to the fog computing location awareness capabilities, a large quantity of devices exists in the physical environment with a short cover range. This leads to location privacy exposure by the connection triggered. Adversaries can pry into more private data through the commodiously accessible location information. Although the existing privacy-preserving schemes can address some issues such as differential privacy, it cannot meet various privacy expectations in practice for fog computing variants. Motivated by this, we propose a location-aware dynamic dual \(\epsilon \)-differential privacy preservation scheme to provide the ultimate protection. We start by establishing the first scheme by clustering fog nodes with SDN-enabled fog computing. In addition, we customize \(\epsilon \)-differential privacy preservation scheme to tailor-made for the variant fog computing services. Furthermore, we employ a modified Laplacian mechanism to generate noise, with which we find the optimal trade-off. Extensive experimental results confirm the significance of the proposed model in terms of privacy protection level and data utility.