Performance Evaluation of a Cluster-Based Service Discovery Protocol for Heterogeneous Wireless Sensor Networks

Abstract—This paper evaluates the performance in terms of resource consumption of a service discovery protocol proposed for heterogeneous Wireless Sensor Networks (WSNs). The protocol is based on a clustering structure, which facilitates the construction of a distributed directory. Nodes with higher capabilities are chosen to be clusterheads. We measure the energy and memory costs depending on the capabilities and the roles of the nodes in the clustering structure, both during the maintenance of the structure and discovery of services. We show that the service discovery protocol succeeds in alleviating the resource-lean sensor nodes of heavy tasks, while delegating more consuming duties to the more powerful nodes.