EE-SAMS: An adaptive, SNN based energy-efficient data aggregation framework for agrovoltaic monitoring systems

Abstract

The emerging trend of agrovoltaic farming monitoring system aims to the reduce energy depletion in battery constrained wireless sensor network. The proposed EE-SAMS model introduces an energy-efficient approach to reduce sensor node energy depletion caused by redundant data transmissions in agrovoltaic farming systems. At the node level, redundancy is minimized using a Euclidean distance-based threshold, forwarding only non-redundant data to the aggregator. At the aggregator level, feature extraction and data classification are conducted through Conv1D and MaxPooling layers, with classification powered by a modified Spiking Neural Network (SNN) using the Adaptive Exponential Integrate-and-Fire (AdEx) model, achieving a high classification accuracy of 99.27%. Selective forwarding further enhances energy efficiency by transmitting only prioritized, non-redundant data to the base station. The performance of EE-SAMS is compared with the MLEM, MLELMAKF, SOF-SVM, and MRMR-KNN models in terms of efficiency and accuracy. The proposed model outperforming the other models and proven that the EE-SAMS is highly suitable for sustainable agrovoltaic monitoring.

Keywords

• Agrovoltaic system
• Data aggregation
• Spike neural network
• Redundancy removal
• Modified AdEx model
• Data classification