Abstract:
In this work, we devise a framework for incorporating the application dependence into the lifetime measurement process of the wireless sensor networks, thereafter via extensive experiments, demonstrate the signi cance of the lifetime metric itself in the quanti cation process for a variety of application scenarios including both scalar and video based wireless sensor networks. We show that the lifetime metrics that ignore application dependence fail in solving the network lifetime quanti cation problem in WSNs. Our proposed framework, weighted cumulative operational time (WCOT), combines two distinct mechanisms for realistic and application context aware network lifetime evaluation. Firstly, by introducing the utility function it enables the users of the network to inscribe their own application level requirements in a formal setting. This clari es the inherent subjectivity due to the application dependence involved in the WSN network lifetime quanti cation problem by transforming it into a form that renders further computation possible. The utility function denotes the total cumulative utility (usefulness) o ered by the collaboration of the sensor nodes. Secondly, instead of o ering a single cut-off threshold value for de ning the point after which the network is assumed to be nonfunctional, WCOT framework makes use of the gradual change in the utility of the network and record how the network evolves over time in terms of functionality o ered by keeping the weighted sum of the operational time. Unlike lifetime metrics that focus on a single threshold value, WCOT is able to di erentiate network performances that di er in how the network evolves till the utility drops to zero.
