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 co
- PDF / 1,445,839 Bytes
- 15 Pages / 439.37 x 666.142 pts Page_size
- 97 Downloads / 282 Views
2
Deakin University, 221 Burwood Highway, Burwood, VIC, Australia {bgu,xdwang,y.qu,yong.xiang}@deakin.edu.au Swinburne University of Technology, John Street, Hawthorn, VIC, Australia [email protected]
Abstract. 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 -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 -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. Keywords: Differential privacy network
1
· Fog computing · Software defined
Introduction
With the rapid development of the fog computing [3], connectivity between fog nodes is becoming stronger and more ubiquitous. An increasing volume of entities are becoming intelligent and serving as fog nodes. They can perceive the surrounding environment, connect to the Internet, and receive commands remotely by their location information. The intelligence of these fog nodes is the result of data, analysis, and feedback from various systems or servers of different mobile devices. A large amount of the data leads to an increased possibility of privacy c Springer Nature Singapore Pte Ltd. 2020 S. Yu et al. (Eds.): SPDE 2020, CCIS 1268, pp. 176–190, 2020. https://doi.org/10.1007/978-981-15-9129-7_13
Location-Aware Privacy Preserving Scheme in SDN
177
issues such as location-aware privacy, sensitive context-aware privacy, etc. [7,8]. Therefore, it is necessary to protect the privacy between fog computing devices and users among the network infrastructure. According to the fog computing reference model [4], location-based services have been widely used [24], for example, vehicular systems, smart grid, and smart city. Traditionally, end devices are required to connect to the best available fog node in order to improve user experience in a convenient way [16]. Location-aware privacy issues occur when the transmission has been confirmed [25]. Adversaries can easily detect and attack fog nodes in an ap
Data Loading...
