Exploiting Constructive Interference for Scalable Flooding in Wireless Networks

Constructive interference-based flooding (CIBF) is a latency-optimal flooding protocol, which can realize millisecond network flooding latency and submicrosecond time synchronization accuracy, require no network state information, and be adapted to topology changes. However, constructive interference (CI) has a precondition to function, i.e., the maximum temporal displacement Δ of concurrent packet transmissions should be less than a given hardware constrained threshold (e.g., 0.5 μs, for the IEEE 802.15.4 radio). In this paper, we derive the closed-form packet reception ratio (PRR) formula for CIBF and theoretically disclose that CIBF suffers the scalability problem. The packet reception performance of intermediate nodes degrades significantly as the density or the size of the network increases. We analytically show that CIBF has a PRR lower bound (94.5%) in the grid topology. Based on this observation, we propose the spine constructive interference-based flooding (SCIF) protocol for an arbitrary uniformly distributed topology. Extensive simulations show that SCIF floods the entire network much more reliably than the state-of- the-art Glossy protocol does in high-density or large-scale networks. We further explain the root cause of CI with waveform analysis, which is mainly examined in simulations and experiments.