Geçiçi araç ağları için çok sekmeli gruplama ve LTE bazlı heterojen mimariler

Vehicular Ad-Hoc Network (VANET) is a promising Intelligent Transportation System (ITS) technology that enables numerous applications such as safety message dissemination, dynamic route discovery, gaming and entertainment. First part of the thesis focuses on constructing stable clusters by determining the vehicles sharing similar mobility pattern to provide robust communication with minimum overhead in the presence of highly mobile vehicles. In this context, we propose VMaSC: Vehicular Multi-hop algorithm for Stable Clustering, a novel clustering technique based on choosing the node with the least mobility through multiple hops. Extensive simulation experiments performed using Network Simulator (ns-3) with the vehicle mobility input from the Simulation of Urban Mobility (SUMO) reveal that VMaSC increases cluster head duration by 25\% while decreasing the number of cluster head changes by 10%. Second part of the thesis considers the integration of IEEE Wireless Access in Vehicular Environments (WAVE) and 3GPP networks (LTE). WAVE operates based on ad-hoc mode with IEEE 802.11p protocol and enables vehicle-to-x (V2X) communication with vehicles and roadside infrastructures. LTE is a state-of-the art technology for mobile communication and provides a cellular infrastructure based solution. We propose an architecture combining these two technologies to achieve the high data rates of IEEE 802.11p-based VANETs and wide coverage of 3GPP (LTE) technology simultaneously. In this architecture, vehicles are clustered based on our approach VMaSC, and elected heads operate as dual-interface node with the functionality of IEEE 802.11p and LTE interface. By performing extensive simulation experiments in ns-3 with the vehicle mobility input from the Simulation of Urban Mobility (SUMO), multi-hop clustered VANET-LTE integrated architecture has been demonstrated to achieve over 90\% data packet delivery ratio with maximum delay below 1 second.