IEEE Internet of Things Journal · 2018 · 147 citations · 24 references
ComputingComputer ScienceCybersecurityComputer SecurityInternet Of ThingsNetwork SecurityDifferential PrivacyIot PrivacyIot SecuritySecure RoutingAttack Traffic ClassificationEncrypted Traffic AnalysisMalicious Traffic DetectionPrivacy-preserving Iot ApplicationsPrivacy-preserving Machine LearningSecurity In Smart HomesSmart Home PrivacyTraffic Analysis
The Internet of Things (IoT) becomes a novel paradigm as more and more devices are connected to the Internet, enabling several innovative applications such as smart home, industrial automation, and connected health. However, the cyber-attack to these applications is a big issue and countermeasures are in dire need to provide system security and user privacy. In this paper, we address the traffic analysis attack to smart homes, where adversaries intercept the Internet traffic from/to the smart home gateway and profile residents' behaviors through digital traces. Traditional cryptographic tools may not work well due to the effectiveness of adversaries' machine learning algorithms in classifying encrypted traffic, so here we propose a privacy-preserving traffic obfuscation framework to achieve the goal. To be specific, we leverage the smart community network of wirelessly connected smart homes and intentionally direct each smart home's traffic to another home gateway before entering the Internet. The design jointly considers the network energy consumption and the resource constraints in IoT devices, while achieving strong differential privacy guarantee so that adversaries cannot link any traffic flow to a specific smart home. Besides, we consider a hostile smart community network and develop secure multihop routing protocols to guarantee the source/destination unlinkability and satisfy each user's personalized privacy requirement. To evaluate the effectiveness of our framework in protecting privacy and reducing network energy consumption, extensive simulations are conducted and the results demonstrate that our design outperforms other differential privacy mechanism in preserving privacy and minimizing network utility cost.
24
The Internet of Things: A survey
Luigi Atzori, Antonio Iera, Giacomo Morabito · Computer Networks · 2010 · 15.3K citations
Epidemic routing for partially-connected ad hoc networks
Amin Vahdat · 2000 · 4.2K citations
Miguel E. Andrés, Nicolás E. Bordenabe, Konstantinos Chatzikokolakis et al. · 2013 · 1K citations · Full text
Computing, Cybersecurity, Geographic Information Systems +10
Reza Shokri, George Theodorakopoulos, Jean‐Yves Le Boudec et al. · 2011 · 701 citations · Full text