Efficient receiver-based flooding in mobile ad hoc networks

Flooding is one of the most important operations in mobile ad hoc networks. There are two strategies for existing flooding algorithms to make the forwarding decision. The first strategy is sender-based, where a forwarding node is in charge of determining its next-hop forwarding nodes. Existing sender-based algorithms can achieve linear computation overhead for each node, but this is at the expense of the transmission redundancy. The second strategy is receiver-based, where a node makes the forwarding decision of itself when it receives the flooding message. Receiver-based flooding algorithms perform well on reducing redundancy of transmissions, but each node requires at least O(n2)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$O(n^2)$$\end{document} computations, where n is its number of neighbors. In this paper, we show that a receiver-based algorithm can achieve both good performance and low computation overhead at the same time. We first introduce an efficient O(nlogn) algorithm, RBF. Second, two extensions of RBF, RBF-E1 and RBF-E2, were proposed. RBF-E1 has outstanding performance (1) in 3D wireless networks, and (2) when the defer time is short. RBF-E2 further reduces the computation overhead for each node to O(n). The proposed algorithms require only one-hop neighbor location information to make the forwarding decision. Extensive simulations are carried out to evaluate the performance of our algorithms.