Hiding Contextual Information for Defending a Global Attacker

Abstract

As wireless sensor networks are mostly deployed in harsh and hostile environments, lack of human maintenance, it has become a severe challenge to offer security or privacy service, especially when facing a powerful attacker. For defending a global attacker, dummy message injection methods are employed by every node in WSN to enhance location privacy, which leads to significant overhead. Based on random walk model, we propose a novel scheme Chess-Board Alternation (CBA) which partition WSN into two sets, and each set operates active in an alternating fashion. We also prove CBA's outstanding performance, including delivery time, energy consumption, and privacy theoretically and experimentally. In comparison with existing approaches, CBA can not only achieve perfect privacy but also decrease energy cost by 50% and the transmission latency from the source to the sink by more than 40%.

Keywords

• Wireless sensor network
• global attacker
• contextual information
• location privacy
• random walk
• information privacy