In this paper, we study the geographic collaborative forwarding (GCF) scheme, a variant of opportunistic routing, which exploits the broadcast nature and spatial diversity of the wireless medium to improve the packet delivery efficiency. Our goal is to fully understand the principles, the gains, and the tradeoffs of the node collaboration and its associated cost, thus provide insightful analysis and guidance to the design of more efficient routing/forwarding protocols. We first identify the upper bound of the expected packet advancement (EPA) that GCF can achieve and prove the concavity of the maximum EPA. With energy efficiency as a major concern, we propose a new metric, EPA per unit energy consumption, which balances the packet advancement, reliability and energy consumption. By leveraging the proved properties, we then propose an efficient algorithm which selects a feasible candidate set that maximizes this local metric. We validate our analysis results by simulations, and justify the effectiveness of the new metric by comparing the performance of GCF with those of the existing geographic and opportunistic routing schemes.
展开▼
