An Efficient and Reliable Routing Method for Hybrid Mobile Ad Hoc Networks Using Deep Reinforcement Learning

Area of Science:

• Computer Science
• Wireless Communication
• Network Engineering

Background:

• Mobile smart devices with wireless communication are increasingly relied upon.
• Effective communication is crucial for locating and rescuing survivors during disasters.
• Restoring traditional communication infrastructure post-disaster is often time-consuming.

Purpose of the Study:

• To propose a method for extending the operational lifetime of Mobile Ad Hoc Networks (MANETs).
• To ensure continuous communication with the nearest base station (BS) during disaster scenarios.
• To facilitate rapid, temporary communication links for survivor assistance.

Main Methods:

• Implementing a load-balancing strategy across network nodes.
• Prioritizing energy conservation by avoiding the exhaustion of nodes with limited remaining energy.
• Maintaining stable communication pathways to the base station (BS).

Main Results:

• Significant improvement in the overall lifetime of the MANET.
• Comparable Packet Delivery Rate (PDR) to existing methods.
• Similar route generation time compared to existing methods.

Conclusions:

• The proposed load-balancing method effectively extends MANET lifetime for disaster response.
• The approach ensures reliable communication during emergencies while preserving network resources.
• This methodology offers a viable solution for rapid communication establishment in critical situations.