Energy-Aware Compression and Consumption Algorithms for Efficient TinyML Model Using Aquila Optimization in Industrial IoT

With the proliferation of Industrial Internet of Things (IIoT) applications, the demand for energy efficient compression and clustering algorithms has escalated along with reliable routing algorithms. This paper proposes a novel approach: an energy aware compression based cluster and routing algorithms leveraging efficient tiny machine learning (TinyML) model using Aquila optimization for enhanced performance in IIoT environments. This approach aims to mitigate the challenges of energy and quality aware compression and energy consumption often encountered in IIoTs within underground coal mines. By dynamically forming clusters of IoT devices based on their proximity and data compression, our approach distributes the data traffic evenly across the network, thus preventing congestion and prolonging network lifetime. The integration of Aquila optimization further refines the routing decisions, optimizing the network’s overall performance. Through extensive simulations and comparative analysis, proposed method demonstrates superior performance metrics, including network lifetime by 62.3%, a 66% reduction in energy consumption and a 41% improvement in live nodes compared to existing methods. Additionally, our algorithms show 21% improvement in packet delivery ratio and increase network stability, making them particularly effective in challenging environments such as underground coal mines. This research contributes to the advancement of efficient routing techniques adapted for the demanding requirements of IIoT deployments, promising a more robust and sustainable IIoT.