Monitoring embedded flow networks using graph Fourier transform enabled sparse molecular relays

Many embedded networks are difficult to monitor, such as water distribution networks (WDNs). A key challenge is how to use minimum sparse sensors to measure contamination and transmit contamination data to a hub for system analysis. Existing approaches deploy sensors using multi-objective optimisation and transmit the data using ground penetrating waves or fixed-line access. Here, for the first time, we introduce a novel molecular communication relay system, which is able to transmit the data report to the hub via the water-flow of WDN itself, and avoids the complex ground penetrating techniques. A water flow data-driven Graph Fourier Transform (GFT) sampling method is designed to inform the invariant orthogonal locations for deploying the molecular relay sensors. Each sensor encodes information via a DNA molecule that enables the common hub to reconstruct the full contamination information. Numerical simulation validates the proposed system, providing a pathway to integrate MC into macro-scale Digital Twin platforms for infrastructure monitoring.