Abstract
This study investigates the graph theoretic implications of applying collaborative rate-adaptive storage and data management to enhance the reliability of connectivity-challenged sensor networks employing dynamic heterogeneity setups. The topologies of sensor networks are synthesized through a spatially-dependent probabilistic random graph model. Around this model a degree-density-connectivity-clustering graph theoretic framework is developed and utilized to study the performance of a recently introduced sensing-and-storage management mechanism by investigating storage outage, weighted storage outage and storage traffic patterns. Through both analytical formulations and simulation results the proposed mechanism shows promising potential to improve the network's reliability in different initial and emergent conditions. The paper also shows how the overall network's quality of observation is affected by upgrading the network's density, radio ranges or clustering characteristics.