Multi-Party Dynamic State Estimation That Preserves Data and Model Privacy

In this paper we focus on the dynamic state estimation which harnesses a vast amount of sensing data harvested by multiple parties and recognize that in many applications, to improve collaborations between parties, procedure must be designed with awareness protecting participants’ model privacy, where latter refers privacy key parameters observation models. We develop paradigm for scenario concerns are involved. Multiple monitor physical process deploying their own sensor networks update estimate according average all calculated cloud server security module. The taps additively homomorphic encryption enables module jointly fuse parties’ while preserving privacy. Meanwhile, collaboratively stable (or optimal) fusion rule without divulging sensitive information. For proposed filtering paradigm, analyze stabilization optimality. First, stabilize multi-party estimator two design methods proposed. special scenarios, directly gains matrix norm relaxation. general after transforming original problem into convex semi-definite programming problem, derive suitable based alternating direction method multipliers (ADMM). Second, an optimal collaborative gain guarantees is provided, results asymptotic minimum mean square error (MMSE) estimation. Finally, numerical examples presented illustrate our theoretical findings.