Non-equilibrium statistical behaviors within common MANET routing protocols

Mobile adhoc networks (MANETs) have been widely proposed as solutions for delivering peered wireless services (i.e., for disaster recovery scenarios, etc). The quality of service (QoS) and quality of experience (QoE) delivered by such MANETs depends largely on the performance of their underlying routing protocols. Early research has begun to show that networklevel MANET behaviors tend not to follow standard equilibrium system assumptions (i.e., QoS measures not exhibit the statistical stationarity and ergodicity required for Birkhoff´s Ergodic theorem to apply). Hence, the standard approach of reporting MANET results in terms of averages computed across ensembles of Monte Carlo simulation runs is insufficient since ergodicity does not hold. This work extends such prior works by showing that non-equilibrium behaviors continues to hold even when much larger sets of Monte Carlo runs are analyzed (i.e., sets of 100 runs as opposed to the prior 15 runs) providing further credence to the claim that these MANETs behave more generally as non-equilibrium systems that has previously been supposed within the prior literature.