TrickleTree: A Gossiping Approach To Fast And Collision Free Staggered Scheduling

In recent years, data gathering has received significant attention as an application of Wireless Sensor Networks (WSNs). Staggered data tree based protocols have been shown to be successful in reducing energy consumption in data gathering scenarios. An important part of staggered protocols is the process of schedule construction. In order to minimize energy consumption, this process must be fast. In this paper, we present TrickleTree, a fast distributed protocol for establishing staggered and collision free communication schedule. TrickleTree has three functions: to establish routes, i.e., construct a data gathering tree, to establish a staggered communication schedule, i.e, assign time slots to links, and to disseminate the maximal tree depth in the network. To minimize network setup time, TrickleTree combines neighborhood discovery and schedule construction into one step. To ensure that good neighbors are discovered before a node joins the network, TrickleTree uses a rating mechanism. Collisions during node association are reduced by using association slots. To increase the message delivery rate with small message overhead, TrickleTree uses adaptive gossiping. We provide a formal analysis of the protocol properties i.e., collision free scheduling and termination. The behavior of the proposed approach is evaluated in simulation. The results show up to 90% in a reduction in schedule setup time and a 50% reduction of duty cycle compared to a flooding approach.