Distributed Real Time TDMA Scheduling Algorithm for Tree Topology WSNs

Abstract

In this paper, we address the problem of developing TDMA scheduling algorithm for tree topology WSNs. The data transmissions are organized into periodic data flows that may have opposite directions since they are carrying sensor and actuator values for feedback control. It is required to determine a periodic and collision-free allocation of the time-slots to the sensor nodes such that the end-to-end deadline of each data flow, as given in time units, is satisfied. The objective is to maximize the lifetime of the network by maximizing the time when the nodes are in the sleep mode. However, the longer the time at which the nodes stay in the sleep mode, the harder is to meet the timeliness requirements of the data flows. To solve the TDMA scheduling problem, we have found an elegant approach to express the end-to-end deadline as an integer number of the length of the schedule period. Moreover, since the distributed algorithms, in compassion with the centralized algorithms, well-suit the scarce resources of the WSNs, we focus on the distributed methods that allow each node in the network to come up with its allocated time-slots in the schedule. The proposed algorithm is based on the graph theory algorithms, namely the distributed shortest path and the distributed topological ordering. Furthermore, it falls into the category of the exact algorithms for tree topology with single-collision domain and in the category of the heuristic algorithms for multiple-collision domains tree topology.