Edge Based Priority-Aware Dynamic Resource Allocation for Internet of Things Networks.

Area of Science:

• Internet-of-Things (IoT)
• Wireless Communication Networks
• Resource Allocation Algorithms

Background:

• The proliferation of edge-based IoT services necessitates efficient communication networks like Low Power Wide Area Networks (LPWAN).
• LoRaWAN, a key LPWAN standard, faces challenges in dense environments including limited battery life, spectrum scarcity, and data collisions for smart End Devices (EDs).
• Intelligent and efficient service provisioning is crucial for optimizing LPWAN performance and reliability.

Purpose of the Study:

• To propose a novel Dynamic Reinforcement Learning Resource Allocation (DRLRA) approach for LPWAN.
• To enhance the performance of smart End Devices (EDs) in terms of energy consumption and communication reliability.
• To address the challenges of limited battery life, spectrum coverage, and data collisions in dense network environments.

Main Methods:

• Development of a Dynamic Reinforcement Learning Resource Allocation (DRLRA) model.
• Allocation of key resources including channel, Spreading Factor (SF), and Transmit Power (Tp) to End Devices (EDs).
• Extensive simulation and evaluation against existing algorithms like Adaptive Data Rate (ADR) and Adaptive Priority-aware Resource Allocation (APRA) using standard and advanced metrics.

Main Results:

• The DRLRA approach demonstrates improved performance in energy consumption and reliability for End Devices (EDs).
• Comparative analysis shows DRLRA outperforming ADR and APRA under various network conditions.
• Cross-validation ensures the robustness and unbiased nature of the obtained results.

Conclusions:

• The proposed DRLRA approach offers an effective solution for intelligent resource management in LPWAN environments.
• DRLRA significantly enhances the efficiency and reliability of smart End Devices (EDs) in dense IoT networks.
• This work provides a foundation for future research in optimizing LPWAN resource allocation for evolving IoT ecosystems.