Deep learning-based energy prediction and tangent search remora optimization-based secure multi-path data...

Area of Science:

• Computer Science
• Network Engineering
• Artificial Intelligence

Background:

• Wireless Sensor Networks (WSNs) are crucial for data collection in inaccessible areas.
• Efficient data transmission to the Base Station (BS) is vital for WSN functionality.
• Multipath routing protocols offer enhanced reliability for WSN communication.

Purpose of the Study:

• To develop an optimal multipath routing strategy for WSNs.
• To minimize energy consumption while meeting end-to-end delay requirements.
• To reduce the error rate in data transmission.

Main Methods:

• An effective multipath routing protocol integrating energy prediction and hybrid optimization.
• Deep Q-Network (DQN) for accurate energy prediction.
• Tangent Search Remora Optimization (TSRO) algorithm for routing execution and optimization.
• A comprehensive fitness function considering residual energy, distance, throughput, reliability, trust, predicted energy, Link Life Time (LLT), delay, and traffic intensity.

Main Results:

• Achieved superior energy efficiency with 0.402 J consumption.
• Maximized throughput at 25.056 Mbps.
• Enhanced network trust to 84.975.
• Minimized distance to 29.964 m and delay to 0.750 ms.

Conclusions:

• The proposed TSRO-based multipath routing effectively optimizes WSN performance.
• The integration of DQN and TSRO offers a robust solution for energy-efficient and reliable WSN communication.
• The protocol demonstrates significant improvements in key performance metrics, making it suitable for practical WSN deployments.