Grid-based directional minimum cost routing for massively dense wireless sensor networks

We propose the grid-based directional routing for massively dense wireless sensor networks to alleviate the complexity arising from the problem scale. The objective is to minimize the total routing cost, which only depends on positions. The grid-based directional routing consists of two stages: equally spaced grid points compute their routing directions and each node uses the routing direction of the closest grid point as guidance to determine its next forwarding node. This paper mainly describes two approaches for the first stage, based on Dijkstra's method and the fast marching method. Our simulation results reveal that the routing directions derived by the fast marching method work better for determining the minimum cost paths.