Araç oluşumlu ağlarda hareketlilik, kanal modelleme ve trafik yoğunluk tahmini

Vehicular Ad-Hoc Network (VANET) is a promising Intelligent Transportation System (ITS) technology that aims to improve road traffic conditions and safety of passengers. First part of our work deals with providing a realistic analysis of the VANET topology characteristics over time and space using various key metrics of interest. In this analysis, we integrate real-world road topology and real-time data extracted from the Freeway Performance Measurement System (PeMS) database into a microscopic mobility model to generate realistic traffic flows along the highway. Moreover, we use a more realistic, recently proposed, obstacle-based channel model and compare the performance of this sophisticated model to the most commonly used more simplistic channel models including the unit disc and log-normal shadowing models. Our investigation on the key metrics reveals that both log normal and unit disc models fail to provide realistic VANET topology characteristics. We therefore propose a matching mechanism to tune the parameters of the lognormal model according to the vehicle density and a correlation model to take into account the evolution of the link characteristics over time. The proposed method has been demonstrated to provide a good match with more sophisticated but computationally expensive and difficult to implement obstacle based model and validated over the real data of two different highways in California. Second part of our work deals with distributed algorithms for density estimation in VANETs. Vehicle density is an important system metric used in monitoring road traffic conditions. Most of the existing methods for vehicular density estimation either use infrastructure, or use local neighbor information to estimate global vehicle density. These techniques however suffer from low reliability and limited coverage as well as high deployment and maintenance cost. We adapted and implemented three fully distributed algorithms for density estimation, inspired by the mechanisms proposed for system size estimation in peer-to-peer networks. Results show that system size estimation technique can be used for density estimation in VANETs. Moreover, we proposed a completely distributed algorithm CluSampling which has been specifically tailored for VANETs. The extensive simulations of these algorithms at different vehicle traffic densities and area sizes for both highways and urban areas reveal that CluSampling is robust to changes in the network and it provides high accuracy in least convergence time and introduces less overhead on the network and the initiator node.