MoN18: Eighteenth Mathematics of Networks meeting
Arta Cika (Oxford) Quantifying link stability in mobile ad hoc networks using Hidden Markov Models
In mobile ad hoc networks, the existence of a link between two nodes depends on their distance, which is governed by the mobility model, and Rayleigh fading affecting the connection between them. In a typical multipath propagation environment, the channel response, in the equivalent complex baseband, is often described by a zero-mean complex Gaussian random process. It follows that the received signal envelope shows a Rayleigh distribution. Node mobility is modeled using an Ornstein-Uhlenbeck (OU) process. This mean-reverted process is particularly suitable for describing the movement of a group of elements having the same destination suffering from random scatters around the projected trajectory, as it can be the case for a group of Unmanned Aerial Vehicles (UAV) moving toward the same target. The evolution of the link between any two nodes can be modeled as a Hidden Markov Process, which can effectively predict the presence (or absence) of a connection according to its signal-to-noise ratio (SNR). This model is used to formulate a stability metric based on the notion of entropy rate, taking full advantage of the correlation between the link current and future state. The motivations behind this work arise from the intrinsic location uncertainty of MANETs and the ability of the entropy rate to capture this randomness. The final goal of my research is the design of link state prediction-based routing algorithms using the entropy rate and the aforementioned model. An efficient routing protocol is one which adapts to the frequently changing network topology. Therefore, the purpose is to exploit the concept of routing by predicting the states of the links and estimating the route stability using the entropy rate.
Contact:
Keith Briggs
(
)
or
Richard G. Clegg (richard@richardclegg.org)