Market Model for Resource Allocation in Emerging Sensor Networks with Reinforcement Learning

Area of Science:

• Computer Science
• Network Engineering
• Artificial Intelligence

Background:

• Emerging sensor networks (ESNs) are integral to the Internet of Things (IoT), connecting numerous intelligent devices.
• Efficient resource allocation is critical in ESNs, particularly under limited resource conditions.

Purpose of the Study:

• To address resource allocation challenges in ESNs by modeling them as multi-agent environments.
• To develop and verify efficient resource allocation strategies using market models and reinforcement learning.

Main Methods:

• Utilized agent-based modeling (ABM) to represent ESNs as multi-agent systems.
• Employed market models to handle resource allocation problems based on user patterns.
• Applied reinforcement learning (RL) to estimate user patterns and validate market model outcomes.

Main Results:

• Demonstrated the efficiency of the proposed resource allocation methods through experimental results.
• The developed methods proved effective in guiding topology management for ESNs.

Conclusions:

• The integration of ABM, market models, and RL offers an effective solution for resource allocation in ESNs.
• The proposed approach enhances efficiency and provides a framework for intelligent network topology management.