Grid Block Energy Based Data Gathering Algorithms for Wireless Sensor Networks

We propose Grid Block Energy based hierarchical Data Gathering (GBE-DG) algorithms for wireless sensor networks. We divide the whole sensor network into grid blocks of equal size. The energy level of a grid block is the sum of the energy levels of the sensor nodes located in it. The grid block that has the maximum energy level is called the leader grid block (LGB) and the sensor node that has the maximum energy level in the LGB is called the global cluster leader (GCL). Each grid block has a local cluster leader (LCL), which is the sensor node with the highest energy level within the grid block. The leaf nodes of the data gathering tree are the non-LCL nodes in each grid block, which either directly forward their data to the LCL (GBE-Cluster-DG tree) like in a LEACH cluster or by forming a chain of nodes involving the LCL (GBEChain-DG tree) as in PEGASIS. After receiving the aggregated data from the nodes in its grid block, an LCL node i forwards the data to the LCL node j (could be the GCL node) that is closer to i as well as to the GCL node. Simulation results show the GBE-Chain-DG trees to be relatively better than GBE-Cluster-DG trees and both these algorithms perform considerably better than the well-known LEACH and PEGASIS data gathering algorithms.