Adaptive LINE-P: An Adaptive Linear Energy Prediction Model for Wireless Sensor Network Nodes

Area of Science:

• Wireless Sensor Networks (WSNs)
• Energy Harvesting
• Power Management

Background:

• Existing energy prediction models for WSNs often use fixed weighting parameters, limiting their effectiveness with variable energy sources like solar harvesters.
• Effective power management in WSNs is crucial for node longevity and network reliability.

Purpose of the Study:

• To propose an adaptive energy prediction model, Adaptive LINE-P, for WSNs that overcomes the limitations of fixed weighting parameters.
• To develop a profile compression method to decrease memory requirements for WSN energy management.

Main Methods:

• Developed the Adaptive LINE-P model, which calculates adaptive weighting parameters using stored energy profiles.
• Implemented a profile compression technique to optimize memory usage.
• Validated the model using real-world solar and wind energy profile data.

Main Results:

• The Adaptive LINE-P model demonstrated 90-94% accuracy in energy prediction.
• The profile compression method achieved a 50% reduction in memory overhead compared to current state-of-the-art models.

Conclusions:

• The proposed Adaptive LINE-P model offers a significant improvement in energy prediction accuracy for WSNs with variable energy sources.
• The integrated profile compression method enhances the practicality of WSN energy management by reducing memory footprint.