Research On Delay In Low-Duty-Cycle Wireless Sensor Networks

The development of computer networks and wireless communication technologies have prompt the popularization of wireless sensor network applications such as forest monitoring and military surveillance. Most of these applications require short data delivery latency, i.e., the communication time between the source and the sink needs to be as short as possible. However, a majority of the applications introduced duty cycle mechanism to save energy and this results in sleep latency.To shorten the data delivery latency under a low power consuming mechanism, we introduced two strategies for delay control in wireless sensor networks. Firstly, we proposed a multi-pipeline feature based multi-schedule forwarding (MSF) strategy which includes two optimization algorithms. In this strategy, every node has more than one parent node and is able to assign different active times to all of them. To further reduce the time interval between data sensing and data receiving, we also take packet generation time into consideration while optimizing packet delivery latency. By assigning pre-calculated wake-up time slots to the nodes, the end-to-end average delay and maximum delay could be minimized through coordinating multiple paths with little costs. Secondly, we present a dynamic duty-cycle control strategy (D2CS) strategy based on opportunistic routing. In our design, nodes in network dynamically re-calculate their wake-up duration based on current packet latency. With D2CS, network delay can be controlled on demand and the overall duty-cycle could be reduced.Finally, we compared the MSF strategy and D2CS strategy with traditional ones separately through experiments. The results show that with MSF strategy, both average and maximum end-to-end delay could be kept at a low level with very few overhead. The MSF performs much better than traditional multi-pipeline technology under different situations of link quality and network traffic. Specially, when the network traffic is heavy, the average delay can be reduced by20%. And D2CS can reduce significant energy cost under different link qualities and network traffic conditions comparing to the traditional opportunistic routing strategy.