Balancing Localization Accuracy and Location Privacy in Mobile Cooperative Localization

Location privacy leakage in cooperative localization is receiving increasing attention. Geo-indistinguishability (GI) protects location privacy by adding random noise to an actual location, which inevitably degrades localization accuracy. This work considers a cooperative localization scenario with mobile nodes, where the cooperative nodes (CNs) share their locations under GI. We investigate the tradeoff between target localization accuracy and CNs’ location privacy. Different from the static scenario, the temporal correlations among the target states and CN locations in the mobile scenario are considered in evaluating the localization accuracy and location privacy. From the aspect of the target, we derive the posterior Cramér-Rao lower bound (PCRLB) about the target state and provide the theoretical localization accuracy. From the aspect of the CNs, their location privacy is evaluated by the squared position error bound (SPEB) inferred by adversaries. To balance the target localization accuracy and CNs’ location privacy, we propose a game-based incentive mechanism by formulating a bilevel joint optimization problem, which meets the objectives of the CNs and the target. First, we derive the closed-form solution for CNs’ privacy preservation strategies by maximizing their utilities. Second, with the privacy preservation information from CNs, the target decides its payment allocation strategy by minimizing its localization error under a finite budget. Extensive numerical results validate our theoretical analysis and the effectiveness of the proposed incentive mechanism in terms of localization accuracy and CNs’ utilities.