Adapting statistical multi-hop wireless broadcast protocol decision thresholds using rate control

Abstract

Multi-hop wireless broadcast is an important component of Vehicular Ad-hoc Network (VA NET) data dissemination applications. Protocols that address multi-hop broadcast must perform well across a broad spectrum of network conditions while being resilient to the rapidly changing network topology of moving vehicles. We propose the Rate-Adaptive Broadcast (RAB) protocol to address this problem. RAB is tailored specifically for information dissemination in VANETs. It is based on the simple distance-method scheme, requires no knowledge of network topology, and yet is adaptive to network conditions. By assuming a VA NET dissemination application with fixed periodic updates, RAB is able to use a novel decision threshold control algorithm based on the rate of incoming messages. It tracks both the rate of received new messages and duplicate messages. If the new message rate dips below its long-run average, the decision threshold is adjusted to improve message propagation. While the new message rate is stable, RAB adjusts the decision threshold to keep the duplicate message rate in an efficient range. Thus RAB jointly optimizes the broadcast message delivery rate and the bandwidth consumption. RAB is evaluated in a realistic vehicular networking simulation using JiST/SWANS and unlike similar protocols that require no overhead messaging, is shown to achieve a high reachability level across a wide range of networking scenarios while maintaining efficient bandwidth consumption.