Enhancing Topology Control Algorithms in Wireless Sensor Network using Non-Isotropic Radio Models

In reality, radio range of any wireless communication device does not emulate an isotropic behavior or as modeled by theory. An isotropic antenna is a hypothetical lossless antenna having equal energy being radiated or transmitted in all directions. Along with wireless sensor networks, any wireless network suffered propagation complications including multipath propagation, shadowing and path loss to name a few which are vital factors to consider in a wireless environment. Radio irregularity is ubiquitous in every wireless sensor network that it should be constituted in every protocol or algorithm design. As per some other studies, radio propagation can be evaluated by statistical means therefore, making it an easier and general way to characterize radio irregularity without being dependent to any specific environment. Furthermore, it could be very flexible in varying network conditions like mobility. In this paper, we develop both linear and non-linear models to capture radio irregularity using regression analysis on distance and received signal strength. These models are assessed by implementing a simple middleware to be added to topology control algorithms.