Cross-Layer Protocol Design and Performance Evaluation of LoRa Ad Hoc Networks for Heterogeneous Traffic

Area of Science:

• Wireless Communication Networks
• Internet of Things (IoT)
• Network Protocols

Background:

• LoRa networks face coverage blind spots and collision bottlenecks in dense, 3D environments.
• Existing single-star architectures limit network capacity and reliability.
• Heterogeneous traffic demands require advanced Quality of Service (QoS) guarantees.

Purpose of the Study:

• To propose a distributed cross-layer protocol framework for LoRa ad hoc networks.
• To overcome limitations of traditional LoRa network architectures.
• To support heterogeneous traffic with guaranteed QoS and improved network performance.

Main Methods:

• Developed a 3D penetration loss model at the physical layer.
• Designed a distributed relay deployment algorithm using hybrid simulated annealing.
• Implemented a non-preemptive priority access mechanism with differentiated backoff windows at the MAC layer.
• Proposed the CAM-AODV routing algorithm integrating multiple performance metrics.

Main Results:

• Achieved blind-spot-free connectivity in complex 3D spaces.
• Reduced average end-to-end delay by 19.46% compared to traditional AODV in large-scale scenarios.
• Improved packet delivery ratio (PDR) by 16.32% and reduced system delay by 13.66% under high-concurrent loads.
• Enhanced Energy Balancing Index (EBI) by over 10% in evaluation scenarios.

Conclusions:

• The proposed cross-layer framework effectively breaks traditional network capacity bottlenecks.
• The solution provides an efficient joint optimization for high-capacity, wide-coverage, and long-lifespan IoT networks.
• Demonstrated significant improvements in delay, PDR, and energy balancing for LoRa ad hoc networks.