Energy-Efficient Routing Protocol in Vehicular Opportunistic Networks: A Dynamic Cluster-based Routing Using Deep Reinforcement Learning

Abstract: Opportunistic Networks (OppNets) employ the Store-Carry-Forward (SCF) paradigm to maintain communication during intermittent connectivity. However, routing performance suffers due to dynamic topology changes, unpredictable contact patterns, and resource constraints including limited energy and buffer capacity. These challenges compromise delivery reliability, increase latency, and reduce node longevity in highly dynamic environments. This paper proposes Cluster-based Routing using Deep Reinforcement Learning (CR-DRL), an adaptive routing approach that integrates an Actor-Critic learning framework with a heuristic function. CR-DRL enables real-time optimal relay selection and dynamic cluster overlap adjustment to maintain connectivity while minimizing redundant transmissions and enhancing routing efficiency. Simulation results demonstrate significant improvements over state-of-the-art baselines. CR-DRL extends node lifetimes by up to 21%, overall energy use is reduced by 17%, and nodes remain active for 15% longer. Communication performance also improves, with up to 10% higher delivery ratio, 28.5% lower delay, 7% higher throughput, and data requiring 30% fewer transmission steps across the network.

• The paper introduces CR-DRL, integrating an Actor-Critic framework with heuristic functions to optimize real-time relay selection in vehicular opportunistic networks.
• The paper demonstrates that CR-DRL extends node lifetimes by up to 21%, reduces energy use by 17%, and improves delivery ratio by 10% over existing protocols.
• The paper shows how adaptive cluster overlap adjustments and DRL techniques significantly enhance routing efficiency and reduce transmission latency.

Dynamic Cluster-Based Routing Using Deep Reinforcement Learning in Vehicular Opportunistic Networks

This essay provides a detailed examination of the paper "Energy-Efficient Routing Protocol in Vehicular Opportunistic Networks: A Dynamic Cluster-based Routing Using Deep Reinforcement Learning" (2511.19026). It explores the integration of machine learning techniques, specifically deep reinforcement learning (DRL), within the domain of vehicular opportunistic networks (OppNets), aiming to enhance energy efficiency and routing performance.

Introduction to OppNets and Routing Challenges

Vehicular Opportunistic Networks (OppNets) employ a Store-Carry-Forward (SCF) paradigm to maintain communication despite intermittent connectivity. They face inherent challenges due to dynamic topology changes, unpredictable contact patterns, and resource constraints like limited energy and buffer capacity, which compromise delivery reliability, increase latency, and reduce node longevity. This paper proposes a novel solution through Cluster-based Routing using Deep Reinforcement Learning (CR-DRL), leveraging adaptive strategies to optimize relay selection and cluster dynamics.

Cluster-Based Routing Protocol (CR-DRL)

The CR-DRL approach integrates an Actor-Critic (AC) learning framework with a heuristic function to achieve real-time optimal relay selection and dynamic cluster overlap adjustment. By incorporating DRL, CR-DRL enhances routing efficiency by minimizing redundant transmissions and improving energy consumption. The algorithm evaluates candidate nodes based on encounter rate, remaining energy, and buffer size, selecting the most efficient nodes as cluster heads. This selection is refined using a heuristic function that dynamically adjusts cluster overlap based on node density and distribution, ensuring optimal connectivity and minimized transmission delays.

Numerical Results and Performance Evaluation

Simulation results demonstrate CR-DRL's significant improvements over state-of-the-art baselines. Specifically, CR-DRL extends node lifetimes by up to 21%, reduces overall energy use by 17%, and keeps nodes active for 15% longer. Additionally, the communication performance is enhanced, achieving up to 10% higher delivery ratio, 28.5% lower delay, 7% higher throughput, and requiring 30% fewer transmission steps across the network. These metrics underscore CR-DRL's capacity to provide efficient, resilient routing in highly dynamic vehicular environments.

Implications for Future Developments

The introduction of DRL into cluster-based routing protocols opens pathways for further advancements in adaptive vehicular network management. Practical implications include improved battery sustainability, extended operational time, and reduced latency, enhancing the applicability of OppNets in smart city infrastructures. Theoretically, integrating advanced AI models like DRL within communication frameworks paves the way for scalable, intelligent network solutions adaptable to the unpredictable nature of vehicular networks.

Future research could explore integrating generative AI to simulate realistic vehicular behavior patterns, and federated learning for edge computing deployment in 6G networks. These developments can further optimize routing strategies and scalability, making CR-DRL a foundational model for next-generation intelligent transportation systems.

Conclusion

The "Energy-Efficient Routing Protocol in Vehicular Opportunistic Networks" paper successfully demonstrates how deep reinforcement learning can improve routing efficiency and energy consumption in dynamic network environments. By addressing core challenges with AI-driven adaptive strategies, CR-DRL offers a robust solution for enhancing vehicular communication protocols, setting a precedent for integrating machine learning in network management systems. As AI technologies evolve, CR-DRL provides a promising direction for scalable, real-time solutions tailored to future smart city infrastructures.