Skip to Main Content
In this paper we consider the design of IEEE 802.15.4 wireless sensor networks (WSNs) where nodes belong to different priority classes. Each class is characterized by a specified average data transmission rate requirement, and the overall network forms a multi-level tree. We devise a framework for constructing TDMA schedules for solving the underlying rate differentiation problem using the GTS facility of the standard. Our framework defines a class of schedules, called flow balanced schedules, that are efficient in terms of the delay incurred by packets transmitted to the sink node, and the number of packets queued in each node. We identify two useful optimization aspects that help in constructing schedules with short cycle length. We then outline an algorithm, called GTS-TDMA, which integrates two algorithms that take advantage of the identified optimization aspects, and present simulation results that show the performance gains of the devised algorithm.