Age of Information Minimization in Multicarrier-Based Wireless Powered Sensor Networks

Area of Science:

• Computer Science
• Electrical Engineering
• Network Engineering

Background:

• Wireless Powered Sensor Networks (WPSNs) face challenges in timely information delivery.
• Optimizing data transmission and energy transfer is crucial for network efficiency.

Purpose of the Study:

• To minimize the long-term average weighted sum of the Age of Information (WAoI) in WPSNs.
• To develop an efficient algorithm for balancing data transmission and energy transfer.

Main Methods:

• Formulated the problem as a multi-stage stochastic optimization program.
• Applied Lyapunov optimization to decompose the problem into per-time-block deterministic subproblems.
• Utilized model-free deep reinforcement learning (DRL) to solve these subproblems.

Main Results:

• The proposed DRL-based algorithm significantly reduced WAoI compared to existing methods (DQN, greedy algorithms).
• The algorithm effectively mitigated excessive instantaneous Age of Information (AoI) for individual sensors.
• Demonstrated superior performance in timely information delivery within WPSNs.

Conclusions:

• The novel Lyapunov optimization and DRL approach provides an effective solution for minimizing WAoI in WPSNs.
• This method offers a practical strategy for enhancing real-time monitoring in wireless sensor networks.
• The findings contribute to advancements in efficient data management for resource-constrained networks.