Novel approaches to cooperation schemes in mobile networks
MetadataShow full item record
The advent of next-generation technologies such as short range ad hoc communication devices, sensors, and autonomous robotic agents has introduced a new type of wireless mobile ad hoc network. The dynamic nature of such a network, inherent from its node mobility, introduces major challenges to efficient data communications and information sharing. Even though there have been a considerable number of studies on the impact of random or deterministic node mobility on existing network protocols, few works have exploited controllable or semi-deterministic node mobility for the design of network protocols as a means to tackle those challenges. In this work, we investigate how node mobility (controllable or semi-deterministic) can be considered when designing network protocols to address problems for which traditional approaches cannot be used. First, we study the design of cooperative protocols for mobile wireless sensor networks that exploit the controllable mobility of sensor nodes that are power-constrained and self-propelled. We consider two different but related problems in the remote sensing operation, namely search and survey and monitoring problems, where deploying stationary sensor nodes becomes infeasible. The objective of the first problem is to perform a fast but coarse-grained survey of a very large area, targeting time-sensitive search and survey applications (e.g., surveillance, search and rescue operations, or seismic monitoring). For this problem, we propose a synchronization-based survey (SBS) scheme, in which mobile sensor nodes form an intermittently connected network (ICN). The objective of the second problem is to monitor closely the phenomena of interest which may migrate over time across a more targeted area. In order to address this problem, we propose a fully distributed algorithm, called Causataxis, inspired by biological systems. In the second part of this work, we extend our discussion to design a network architecture for an ITS (Intelligent Transportation System) exploiting vehicles' semideterministic mobility to provide a high-quality data transfer service to users. The proposed ITS architecture utilizes the mobility profile of vehicles as well as a wireless mesh network, called an ad hoc distribution network (ADN), consisting of one or more server stations (SSs) and access points (APs) using ad hoc channels. The objective is, given the mobility information from the vehicles, to design routing protocols and data transmission scheduling algorithms which maximize the amount of requested data that can be delivered to the requesting vehicles during their journey using the ADN. We also prove that the optimization problem, as well as its key variations, is NP-hard.