Modeling cooperative, selfish and malicious behaviors for Trajectory Privacy Preservation using Bayesian game theory

As new mobile Wireless Sensor Networks (mWSNs) for location-aware applications are emerging, trajectory privacy invasion is becoming an indispensable issue. Many promising techniques are under development. Considering the decentralized network architecture, most of Trajectory Privacy Preservation (TPP) techniques rely on the cooperation from peer nodes, cluster headers, or a third party. However, only a few works have addressed the issue of selfish behaviors in such cooperation required techniques. Nevertheless, the problem of facing selfish and compromised nodes in the noncooperative and hostile environment is rarely touched. In this paper, we apply Bayesian game theory to model cooperative, selfish and malicious behaviors of autonomous mobile nodes in decentralized mWSNs. We formulate and analyze the TPP game among peer nodes in both strategic and dynamic forms. The equilibrium strategies for users to evaluate the degree of trust in participating in in-network TPP activities are provided and analyzed in theoretical and simulation results.