Distributed State Estimation Over Time-Varying Graphs: Exploiting the Age-of-Information

We study the problem of designing a distributed observer for an LTI system over time-varying communication graph. The limited existing work on this topic imposes various restrictions either observation model or sequence graphs. In contrast, we propose single-time-scale that works under mild assumptions. Specifically, our only requires strong-connectivity to be preserved nonoverlapping, contiguous intervals are even allowed grow unbounded time. show suitable conditions bound growth such intervals, joint observability is sufficient track state any discrete-time exponentially fast, at desired rate. also develop variant algorithm provably robust worst-case adversarial attacks, provided graphs sufficiently connected key approach notion “freshness-index” keeps age-of-information being diffused across network. Such indices enable nodes reject stale estimates state, and, in turn, contribute stability error dynamics.