Fuzzy inference rule based task offloading model (FI-RBTOM) for edge computing

Area of Science:

• Computer Science
• Artificial Intelligence
• Distributed Systems

Background:

• Edge computing aims to minimize latency and enhance service quality.
• Challenges include high user mobility, dynamic IoT environments, and limited device resources.
• Task offloading is a critical issue in edge computing.

Purpose of the Study:

• To propose an efficient rule-based task-offloading model (FI-RBTOM).
• To optimize task placement decisions (edge, cloud, or local processing).
• To address resource constraints and dynamic environments in edge computing.

Main Methods:

• Developed an efficient rule-based task-offloading model (FI-RBTOM).
• Input parameters include bandwidth, CPU utilization, task length, and task size.
• Simulations conducted using MATLAB with fuzzy logic.

Main Results:

• The FI-RBTOM model demonstrated effective task offloading decisions.
• Achieved an overall error rate of 0.39875 using MATLAB simulations.
• Model trained with 75% data and tested with 25% data.

Conclusions:

• The proposed FI-RBTOM model offers an efficient solution for task offloading in edge computing.
• The model effectively manages task placement considering various environmental and resource parameters.
• Results indicate the model's potential to improve edge computing service quality and reduce delays.