A geographically opportunistic routing protocol used in mobile wireless sensor networks

Abstract

Opportunistic routing protocols have been widely explored to improve the performance of multi-hop communication. However, most existing opportunistic routing protocols face several challenges such as low efficiency in mobile wireless sensor networks (WSNs), diverging, long latency etc. In this paper, a new geographically opportunistic routing protocol (GOR) is proposed to tackle these challenges. The essential idea of GOR is to base routing on static geographic information instead of mobile sensors. In GOR, the bounded sensor area is divided into unchangeable geographic grids at the initialization of a network. Each grid has its priority according to its distance to the static sink. When a source node is going to send a data packet to the sink, it follows the steps below: (1) This source node determines the forwarding path table and the starting grid whose distance is equal to the maximum one-hop transmission distance (the distance that ensures accepting power is larger than a specific threshold), and includes these information at the head of the packet. (2) Then this packet is broadcast. Nodes in the starting grid have the highest priority to forward the packet. If some of these nodes have received the packet, they compete to be the only forwarding node and send acknowledge signals (ACKs) to lower priority nodes by flooding meanwhile. Nodes in lower priority grids delay for different time according to their priorities and forward the packet if having not received ACKs from higher priority nodes. (3) The forwarding node only updates the starting grid of the next hop transmission at the head of the packet. The packet is then transmitted hop by hop until it reaches the sink. Simulation results are presented to demonstrate the effectiveness of GOR.