Loading...

Presenting an Energy-efficient Routing Mechanism in VANETs Using Clustering

Javadzadeh No, Negar | 2025

0 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 58343 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Movaghar Rahimabadi, Ali; Safaei, Bardia
  7. Abstract:
  8. Vehicular Ad-Hoc Networks enable communication between vehicles and roadside units over wireless communication channels. These networks have garnered significant attention in industrial and research domains, particularly in areas such as intelligent transportation systems, traffic management, and safety alert dissemination. However, they face numerous computational and communication challenges, including limited bandwidth, high interference rates, latency, and routing inefficiencies. These issues are further exacerbated by the dynamic nature and rapidly changing topology of the network, resulting in unstable communication links. Typically, mobile networks rely on batteries with limited capacity as their primary energy source, and increased consumption directly impacts system efficiency. In many Internet of Things applications, a significant portion of energy consumption arises from inter-node communications; thus, any optimization in this area can have a direct impact on battery lifetime. One fundamental approach to addressing this issue is the use of energy-efficient routing algorithms. Traditional routing techniques involve the participation of a majority of nodes in the routing process, which leads to unnecessary energy consumption. In contrast, clustering-based methods can reduce energy usage by limiting the number of active nodes involved in routing. In recent years, the adoption of Software-Defined Networking in the IoT domain has seen significant growth as a means of centralized management, improving quality of service, and enhancing energy efficiency. Nevertheless, the potential of SDN has not been widely exploited in VANET environments. urthermore, due to the rapid changes in network topology, any delay in the configuration process and its dissemination to network nodes may lead to frequent reconfigurations, reduce the effectiveness of real-time decisions, and increase management overhead. To address this challenge, predicting the future positions of nodes can enable proactive decision-making for cluster management. Based on the aforementioned considerations, this study proposes a novel framework that leverages the advantages of a software-defined architecture within the dynamic context of VANETs. The proposed framework employs a clustering mechanism and Gated Recurrent Unit for predicting node positions over upcoming time intervals. This approach contributes effectively to enhancing communication performance, increasing stability, and reducing energy consumption within the network. Evaluation results demonstrate that the proposed solution achieves a 20% reduction in energy consumption, while also yielding significant improvements in key performance metrics, including an increase in effective throughput of up to 73%, and a reduction in communication overhead of up to 63%
  9. Keywords:
  10. Mobile Ad Hoc Network ; Vehicular Ad Hoc Network (VANET) ; Energy Consumption ; Routing ; Software Defined Networking (SDN) ; Clustering ; Gated Recurrent Unit (GRU)