A Channel Borrowing Approach for Cluster-based Hierarchical Wireless Sensor Networks

In Wireless Sensor Networks (WSNs), energy-efficient routing is required to conserve the scarce resources of these networks. Various energy-efficient routing protocols have been proposed in this context. Among them, cluster-based hierarchical routing protocols are the most efficient ones. In these protocols, each sensor node, i.e., a member node, continuously captures the data from environment, stores in its buffer and waits for its allocated timeslot to transmit the data to a designated cluster head. The buffer of each node is subject to one or more pre-defined threshold values. If the buffer occupancy level exceeds the threshold value, the node starts to drop the captured data. A member node may continuously capture the data and may have to wait longer than expected to transmit it to the base station. In this context, the data generated by the underlying application may suffer excessive delay and may no longer be useful for further analysis. To address this issue, we propose a channel borrowing scheme for the member nodes within each cluster. Each member node monitors its buffer and if the occupancy level exceeds a threshold value, a channel borrowing request is initiated. Each node maintains a table of potential cluster heads. A neighboring cluster head is selected for channel borrowing, based on its signal strength. If a neighboring cluster head has a spare channel, it is allocated to the member node. The request for borrowing a channel must be acknowledged within a pre-defined threshold waiting time. The experimental results show that our propose scheme reduces packet loss, energy consumption, end-to-end delay and enhances throughput and network lifetime.