Evaluating delay and energy in sensor networks with sporadic and correlated traffic

Wireless sensor networks are usually comprised of a set of sensors that are equipped with a wireless transmitter/ receiver and of a set of data sinks where the information gathered by sensors should be treated. The delay, from the time when events are sensed to the time when the information arrives at sinks, has to satisfy some specific requirements. These requirements depend on the underlying application targeted by the network, and typically consist in ensuring that the delay is less than a pre-defined threshold with high probability. A common issue in sensor networks is then to design distributed scheduling and routing algorithms so as to maximize the network lifetime while guaranteeing the delay requirements. This optimization problem has to account for the particular nature of traffic in sensor networks: the traffic there is (i) sporadic, in the sense that sensors do not have always packets to transmit, (ii) correlated, meaning that the sensors belonging to a common neighborhood may generate traffic simultaneously (e.g., a network sensing temperatures typically exhibits correlated traffic). In this work, we aim at providing analytical and simulation tools able to quantify the packet delays in sensor networks with sporadic and correlated traffic, for various types of distributed scheduling protocols. These tools may then be used to design protocols maximizing the lifetime of delay-constrained networks.