Estimating Maximum Traffic Volume in Wireless Sensor Networks Using Fluid Dynamics Principles

Studying the behavior of Wireless Sensor Networks (WSNs) is a complex task, since the effects of significant network parameters are frequently unpredictable. This, along with the fact that in most network deployments, wireless sensor nodes are densely and randomly deployed, renders the individual study of the behavior of each sensor node impractical. In this work, we attempt to analyze, model, and estimate the maximum volume of traffic than can be carried out from the sources to the sink(s) of a WSN, without the use of any congestion control algorithms. To perform our analysis we employ a macroscopic fluid dynamic model. Using this model and three fundamental traffic variables, packet density, packet flow, and spatial packet rate, we calculate the limits of the network flows, in terms of capacity, in the absence of congestion control. Calculating these limits helps us prove a relation between incoming and outgoing flow in the bottleneck nodes that can specify the optimal point at which the network should operate without the need of congestion control algorithms.