AVOCA Algorithm Enhances Efficiency in Internet of Vehicles

By Dr Silpaja Chandrasekar, PhDReviewed by Susha Cheriyedath, M.Sc.Jan 19 2024

In a paper published in the journal PLOS ONE, researchers investigated the integration of the automotive industry and the Internet of Things (IoT) through the Internet of Vehicles (IoV) to address the challenges in vehicular ad-hoc networks (VANETs). They proposed the African vulture optimization-based clustering algorithm (AVOCA), outperforming existing algorithms by generating fewer clusters and optimizing network parameters.

Related Work

Previous work has focused on mitigating the growing challenges of accidents and traffic congestion in sustainable cities by enhancing intelligent transportation systems (ITS) within the IoT framework. The evolution of VANETs from mobile ad hoc networks (MANETs) has emerged as a critical component in shaping the future of ITS.

In the IoV, where each internet-connected vehicle contributes to intelligent city functionalities, VANETs play a pivotal role. Past research efforts have leveraged technologies like vehicle-to-infrastructure (V2I) and vehicle-to-vehicle (V2V). However, challenges like high mobility, intermittent connectivity, and sparse vehicle distribution in VANETs have spurred investigations into clustering techniques like mobility-based clustering (MOBIC) to optimize network stability.

African Vulture-Inspired VANET Optimization

In the proposed cluster optimization for VANETs using the AVOCA, the biological life of African vultures serves as inspiration. Native to the African continent, these vultures exhibit unique features, including baldness for hygiene during feeding and keen vision. The vultures' migratory behavior, cultural significance, and long lifespan contribute to their classification based on agility. They often engage in rotational flight over long distances in search of food, and conflicts can arise when multiple vultures converge on a single food source.

The AVOCA algorithm is developed based on fundamental vulture-inspired concepts. In the first phase, the algorithm determines the best vulture in the population, utilizing non-dominated random solution vectors. The second phase computes vultures' starvation rates, considering their hunger levels to transition between the exploration and exploitation phases. The third phase involves exploration, where vultures search for food using different tactics. Researchers proposed a comprehensive approach where the fourth phase marks the transition into the exploitation stage, further segmented into two distinctive phases determined by the vultures' starvation rates.

The algorithm maintains an equilibrium phase to balance exploration and exploitation, preventing premature convergence and ensuring diversity. The determination of the computational complexity of AVOCA involves actively assessing the initialization, fitness evaluation, and vulture updating processes. Initialization complexity is O(N) for N vultures. In contrast, the update mechanism complexity is O(N x (T + TD)), where T is the maximum number of iterations and D is the problem dimension. Researchers actively compare AVOCA's computational complexity with that of state-of-the-art algorithms. These algorithmic phases and complexities collectively contribute to the effectiveness of AVOCA in optimizing clusters within VANETs.

VANET Cluster Efficiency Analysis Summary

The analysis evaluates the influence of network nodes and grid sizes on clustering efficiency within the context of VANETs. The experiments maintain consistency in the setup, progressively increasing the grid size from 3 x 3 km² to 4 x 4 km². The schematic representation reveals that the AVOCA algorithm consistently generates fewer clusters than the clustering approach for mobile ad hoc networks (CAMONET), spatially aware mobile ad hoc networks (SAMNET), improved whale optimization algorithm (I-WOA), and hybrid harmony optimization in clustering networks (HHOCNET), showcasing its efficiency across diverse grid sizes.

This efficiency is particularly evident when examining the direct proportional relationship between the number of clusters and network nodes. The optimal cluster generation aligns with the minimum network nodes, demonstrating AVOCA's ability to adapt to varying network scales.

Researchers have delved deeper into this observation, scrutinizing the implications across a broader 4 x 4 km² grid size. The results reaffirm AVOCA's superiority in cluster generation efficiency, consistently producing fewer clusters across different node counts. The percentages highlight the substantial reduction in clusters generated by AVOCA compared to other algorithms. It emphasizes the algorithm's scalability and effectiveness in managing larger network areas with diverse node populations.

The discussion then delves into the relationship between network grid size and cluster efficiency. The inverse proportional connection underscores the critical role of grid size in influencing clustering efficiency. Larger grid sizes are associated with more vehicles within a single grid cell, potentially leading to larger clusters. However, the drawback of increased communication overhead challenges scalability and overall efficiency.

In contrast, smaller grid sizes result in fewer vehicles per cell, fostering smaller, more dynamic clusters. This approach minimizes communication overhead and enhances adaptability to changing network conditions. The emphasis on selecting an optimal grid size underscores the strategic decision-making required for achieving an efficient balance between cluster formation, inter-cluster communication, and adaptability within the dynamic VANET environment.

Conclusion

To sum up, VANETs face scalability, routing, and security challenges. The proposed AVOCA addresses these challenges by optimizing node clustering considering transmission range, node count, and grid size. AVOCA outperforms CAMONET, SAMNET, i-WOA, and HHO, generating 40-45% fewer clusters across diverse scenarios.

Results show direct proportionality between grid size and network nodes with the number of clusters, while transmission range exhibits an inverse relationship. AVOCA ensures optimal cluster generation, stability, load optimization, and improved network utilization. Further enhancements can consider factors like routing protocols and security in real-world scenarios.