Transfer-Efficient Face Routing Using the Planar Graphs of Neighbors in High Density WSNs

Abstract

Face routing has been adopted in wireless sensor networks (WSNs) where topological changes occur frequently or maintaining full network information is difficult. For message forwarding in networks, a planar graph is used to prevent looping, and because long edges are removed by planarization and the resulting planar graph is composed of short edges, and messages are forwarded along multiple nodes connected by them even though they can be forwarded directly. To solve this, face routing using information on all nodes within 2-hop range was adopted to forward messages directly to the farthest node within radio range. However, as the density of the nodes increases, network performance plunges because message transfer nodes receive and process increased node information. To deal with this problem, we propose a new face routing using the planar graphs of neighboring nodes to improve transfer efficiency. It forwards a message directly to the farthest neighbor and reduces loads and processing time by distributing network graph construction and planarization to the neighbors. It also decreases the amount of location information to be transmitted by sending information on the planar graph nodes rather than on all neighboring nodes. Simulation results show that it significantly improves transfer efficiency.

Energy-Efficient and Reliable Face-Routing Scheme in Wireless Networks

“…Greedy mode, called greedy forwarding, is a scheme that can select the closest neighbor nodes to the destination in its one-hop range. As the previous studies of face-routing assumed a radio range shape as an ideal circle surface and dealt with each method [ 7 , 8 , 9 , 10 , 11 , 12 , 13 , 18 , 19 ], we also proceed based on their studies that assume a radio range as an ideal circle shape. After the current node calculates a circle based on the virtual line from this node to the destination node, it finds the closest candidate node to the destination node in the intersection area between this circle and its radio range.…”

“…There are various perimeter modes in geographic routing and, among them, face-routing is one of the well-known perimeter modes. Unlike greedy forwarding, face-routing searches for candidate nodes according to the determined searching direction by using a local graph generated within one hop, it can greatly improve a reliability of data transmission in geographic routing [ 7 , 8 , 9 , 10 , 11 , 12 , 13 , 18 , 19 ].…”

“…However, since face-routing uses the planar graph, it frequently results in unnecessary packet transmission in one-hop range [ 18 , 19 ]. As mentioned above, the planar graph consists of the shortest and adjacent links with neighbor nodes in one-hop range to avoid the routing problem.…”

“…To solve this problem, recent studies on face-routing have proposed to reduce the energy waste of nodes by searching for the farthest-neighbor node and transmitting data packets to the searched neighbor node [ 18 , 19 ]. Transfer-efficient face-routing (TEF) [ 18 ], one of the recent studies finds the farthest-neighbor node by collecting the information of neighbors in the two-hop range and then generating the planar graphs. Since each node generates its planar graph and transmits its information to the source node that starts face-routing, the source node can find the farthest-neighbor node in its one-hop range.…”

“…However, although TEF and EELFR [ 18 , 19 ] can reduce the number of data transmissions by transmitting the farthest-neighbor node, they may not be efficient in face-routing. To reduce overall data transmission from the source node to the destination node, all of studies search the farthest-neighbor node in one-hop range and transmit data packets to the farthest one.…”