New strategies in IoT-enabled Urban vehicular Ad Hoc Networks for smart cities

Abstract

Urban Vehicular Ad Hoc Networks (UVANET) is a subclass of Mobile Ad Hoc Network (MANET) that mainly consists of moving and/or stationary vehicles, connected through spe cialized wireless protocols. This paradigm allows vehicles to exchange messages for safety and non-safety applications through vehicle-to-vehicle, and vehicle-to roadside (V2R) communica tions. Due to the nature of urban environments and vehicular entities, it is challenging to process and demonstrate vehicular related contextual information in urban scenarios. In ad dition, from the literature we find that the combined cost of procurement, deployment and maintenance of RSUs in urban areas is too high. In this thesis, new IoT-based strategies have been developed to overcome potential challenges in Urban Vehicular Ad Hoc Networks (UVANETs) to support smart city conception. The strategies developed communicate important road and vehicular contextual information. Road conditions, weather parameters, the extent of road air pollution and contextual information in public vehicles. Baseline algorithms have been implemented for the vehicular IoT contextual data processing and demonstration. Appropriate allocation of vehicular communication infras tructure resources are studied. Schemes have been conducted that optimally deploy roadside units in urban areas for vehicular communications. Optimization, traffic generator and vehic ular network communication simulation tools have been used to model the optimal placement of infrastructure, traffic generation and simulating the urban vehicular communications. A specific percentage (%) of parked vehicles is exploited as IoT gateways for a specific percentage of their parking duration. 10% percentage of the parked traffic if exploited as IoT gateways for 10% of their parking sites. The studies compare on-street parked vehicles as roadside IoT gateways with main road intersection deployment, and free vehicular communication. Furthermore, a travel matrix roadside units deployment scheme is studied based on a classical delta mechanism, a procedural algorithm is implemented for the optimal placement of RSUs for vehicular communications. A comparison is carried out with road segmentation RSUs deploy ment, hot-spot road junctions deployment and free inter-vehicular communication. Network v performance evaluation is carried out using a network simulator (𝑁𝑆3). According to the re sults, the performance of the networks using introduced models have shown better results in terms of overall data throughput, success delivery rate, end-to-end delay and jitter. For the same traffic densities and the same environment, the main road intersections, road segmen tation and travel matrix RSUs deployment require 13, 18, 10 respectively. Interestingly, the studied method yields better communication results while it needs less number of RSUs