A Novel Deep Learning-Based Cooperative Communication Channel Model for Wireless Underground Sensor Networks.

Area of Science:

• Computer Science
• Electrical Engineering
• Network Engineering

Background:

• Wireless Underground Sensor Networks (WUSNs) are crucial for monitoring subterranean environments.
• Existing WUSN communication models face challenges in hostile underground locations, impacting reliability and energy efficiency.

Purpose of the Study:

• To propose a novel Deep Learning (DL)-based cooperative communication channel model for WUSNs.
• To enhance accurate and reliable monitoring in challenging underground conditions.
• To optimize energy performance and Quality of Service (QoS) through cluster-based cooperative transmission.

Main Methods:

• Developed a DL-based cooperative communication channel model for WUSNs.
• Implemented cluster-based cooperative models utilizing relay nodes and inter-cluster transmission.
• Evaluated performance using Energy Cooperative Channel Allocation Routing (ECCAR), Energy Hierarchical Optimistic Routing (EHOR), Non-Cooperative, and Dynamic Energy Routing (DER) methods.
• Assessed Quality of Service (QoS) parameters including transmission time, throughput, packet loss, and efficiency.

Main Results:

• The proposed model achieved superior energy utilization (89.71%) compared to existing methods.
• Demonstrated significant improvements in Packet Delivery ratio (78.2%) and data packet throughput (83.5%).
• Showcased enhanced performance with reduced Average Packet Delay (82.3%), Average Network overhead (77.4%), and average system packet loss (91%).

Conclusions:

• The proposed DL-based cooperative communication model offers a significant advancement for WUSNs.
• The model provides a reliable and energy-efficient solution for underground monitoring.
• This research contributes to the development of more robust and effective WUSN systems.