Analytic study of Delay/Fault-Tolerant Mobile Sensor Networks (DFT-MSN's)

The Delay/Fault-Tolerant Mobile Sensor Network (DFT-MSN) has been proposed recently for pervasive information gathering. A DFT-MSN consists of a number of wearable sensor nodes and high-end sink nodes, forming a loosely connected mobile sensor network. In this paper we introduce a generic queuing analytic model for DFT-MSN, where the inputs are the data delivery scheme employed and the nodal mobility pattern, while the outputs are the queuing characteristics of the network. Based on our analysis of the message arrival and service processes, we find that each individual sensor can be modeled as an M/M/I/K queue, and the whole network can be treated as a network of queues. Following Jackson network theory, major queuing characteristics of the network can thus be obtained. We also exemplify the generic analytic model with several representative data delivery schemes (including Direct Transmission, ZebraNet, and Replication-based Data Delivery) and nodal mobility patterns (such as uniform and power-law distributions). To validate our analytic model, we have carried out extensive simulations and observed a good match between analytic and simulation results.