Blue and Green-Mode\nEnergy-Efficient Nanoparticle-Based\nChemiresistive Sensor Array Realized by Rapid Ensemble Learning

The rapid advancement of Internet of Things (IoT) necessitates\nthe development of optimized nanoparticle-based chemiresistive sensor\n(CRS) arrays that are energy-efficient, specific, and sensitive. This\nstudy introduces an optimization strategy that employs a rapid ensemble\nlearning-based model committee approach to achieve these goals. Utilizing\nmachine learning models such as Elastic Net Regression, Random Forests,\nand XGBoost, among others, the strategy identifies the most impactful\nsensors in a CRS array for accurate classification. A weighted voting\nmechanism is introduced to aggregate the models’ opinions in\nsensor selection, thereby setting up two distinct working modes, termed\n“Blue” and “Green”. The Blue mode operates\nwith all sensors for maximum detection capability, while the Green\nmode selectively activates only key sensors, significantly reducing\nenergy consumption without compromising detection accuracy. The strategy\nis validated through theoretical calculations and Monte Carlo simulations,\ndemonstrating its effectiveness and accuracy. The employed optimization\nstrategy elevates the detection capability of CRS arrays while also\npushing it closer to theoretical limits, promising significant implications\nfor the development of low-cost, easily fabricable next-generation\nIoT sensor terminals.