Reliable multipath wireless sensor network routing protocol scheme for network lifetime maximization

Abstract

Wireless sensor networks (WSNs) consist of tiny sensor nodes with sensing, computation and wireless communication capabilities. As sensor nodes are resource constrained, the usage of energy and memory must be done wisely to increase the lifetime of nodes. It is also necessary to deliver data reliably to make any application more useful to the end user. In this paper we propose a protocol using cluster based tree topology for network lifetime maximization. The whole network is divided into several clusters. The clusters contain high to low varied density of sensors. Each cluster contains a cluster head which is a child node of some other cluster head. The number of child nodes is given by `n/2i' where `n' is the number of children nodes of the parent node and `i' represents a particular generation/level of the tree. In this way the entire network is arranged in a cluster tree topology with the root node as the sink. Few sensors in the cluster, randomly selected by the sink, sense the environmental parameters and share the information among the neighboring nodes. Thereafter, each node aggregate the parameter values received from other sensors, till these converge to the estimated true value. The cluster heads transmit the data packets of their own clusters, and those obtained from their corresponding child nodes. If the cluster heads find significant traffic in the routes, they dynamically select alternative routes expecting less degree of congestion, to forward the data packets to the higher parent nodes. Thus, this protocol establishes multipath routing scheme in the networks. Energy of the higher parent nodes in the tree structure get depleted faster due to the upstream nature of the traffic. Hence, the density of sensor nodes in the higher parent nodes is made high to maximize the network lifetime. Reliability of this protocol is more, as, a large number of sensor nodes are sensing the parameters of the local area and the consensus decision of the cluster members is sent to the sink.