Stability of consensus-based distributed estimation under denial of service

This paper aims to study the stability properties of consensus-based distributed state estimation (DSE) in the presence of denial of service (DoS) attacks. Specifically, we adopt a model which describes DoS in terms of its average frequency and duration. We focus on a family of DSE algorithms enjoying stability properties in the ideal case of no attacks, and prove that such properties are preserved under DoS provided that appropriate conditions are satisfied, with specific emphasis on the relation between transmission rate on one hand, and bounds on DoS frequency and duration on the other. Numerical simulation tests are shown concerning a target tracking case study.