Reinforcement learning based Secure edge enabled multi task scheduling model for internet of everything applications

Area of Science:

• Computer Science
• Network Engineering
• Cybersecurity

Background:

• The Internet of Everything (IoE) generates vast network data, demanding distributed computing solutions.
• Wireless sensor networks are crucial for IoE data management and job scheduling.
• Security vulnerabilities and high energy consumption are significant challenges in IoE data scheduling.

Purpose of the Study:

• To propose the Secure Edge Enabled Multi-Task Scheduling (SEE-MTS) model for efficient IoE job allocation.
• To enhance IoE application efficiency and data management using edge computing.
• To address security concerns and optimize energy consumption in IoE environments.

Main Methods:

• Leveraging edge computing for efficient data processing and job allocation.
• Implementing a Multi-Task Scheduling (MTS) mechanism for optimized energy usage.
• Utilizing reinforcement learning techniques to minimize task completion time and data usage.
• Employing dynamic updates, multi-key search, data encryption, and result verification for security.

Main Results:

• The SEE-MTS model achieved 4 J energy utilization, 2s delay, and 4s reaction time.
• Demonstrated 89% energy efficiency and a 96% security level.
• Reduced computation time to 6s, improving overall efficiency and security.

Conclusions:

• The SEE-MTS model offers a significant improvement in efficiency and security for IoE data management.
• The model effectively reduces energy consumption, delay, reaction time, and processing time.
• Potential limitations for real-world implementation include the scale of devices and data volume.