Microwave MIMO E-Nose for Wireless Communication and Selective Detection of VOC Mixtures with Concentration Estimation

Microwave MIMO E-Nose for Wireless Communication and Selective Detection of VOC Mixtures with Concentration Estimation

WoS Article

We present the first dual-functional microwave electronic nose (E-nose) that enables wireless communication, VOC mixture detection, and reliable concentration estimation, designed for seamless integration with wireless sensor networks. The proposed E-nose features multiple-input multiple-output (MIMO) antenna system functionalized with molecularly imprinted polymer (MIP) and multiwalled carbon nanotube-based sensing materials for the selective detection of individual or mixed volatile organic compounds (VOCs). We addressed several novel challenges such as managing cross-reactivity under electromagnetic interference with wideband decoupling, employing a dual-branch neural network (NN) with feature prioritization and transducer behavior insights, and optimizing sensor placement for spatial isolation in a compact design. The developed E-nose demonstrated high sensitivity and selectivity for VOC mixtures at room temperature, with detection limits below safety thresholds, ensuring practical applicability. The optimized NN model achieved high predictive accuracy (R 2 = 0.982 to 0.991), delivering near-perfect concentration estimations with negligible errors for pure (0.35%) and mixtures of 2-4 VOCs (2.0 to 3.5%). The proposed framework, customizable for detecting diverse VOCs and toxic gases, enables scalable indoor and outdoor air-quality monitoring. Its seamless dual functionality ensures uninterrupted wireless communication services during gas sensing, establishing a new paradigm in advanced sensor technology with microwave MIMO E-Nose systems.

We present the first dual-functional microwave electronic nose (E-nose) that enables wireless communication, VOC mixture detection, and reliable concentration estimation, designed for seamless integration with wireless sensor networks. The proposed E-nose features multiple-input multiple-output (MIMO) antenna system functionalized with molecularly imprinted polymer (MIP) and multiwalled carbon nanotube-based sensing materials for the selective detection of individual or mixed volatile organic compounds (VOCs). We addressed several novel challenges such as managing cross-reactivity under electromagnetic interference with wideband decoupling, employing a dual-branch neural network (NN) with feature prioritization and transducer behavior insights, and optimizing sensor placement for spatial isolation in a compact design. The developed E-nose demonstrated high sensitivity and selectivity for VOC mixtures at room temperature, with detection limits below safety thresholds, ensuring practical applicability. The optimized NN model achieved high predictive accuracy (R 2 = 0.982 to 0.991), delivering near-perfect concentration estimations with negligible errors for pure (0.35%) and mixtures of 2-4 VOCs (2.0 to 3.5%). The proposed framework, customizable for detecting diverse VOCs and toxic gases, enables scalable indoor and outdoor air-quality monitoring. Its seamless dual functionality ensures uninterrupted wireless communication services during gas sensing, establishing a new paradigm in advanced sensor technology with microwave MIMO E-Nose systems.

MIMO, E-nose, MIP, CNT, VOC, molecularly imprinted polymer, carbon nanotubes, antenna sensor, microwavesensor, dual-functional, gas sensor, selective, volatile organic compounds, mixture sensor, ethanol, isopropanol, butanol, wireless sensor network, neural network, machinelearning, decoupling and matching network

MIMO, E-nose, MIP, CNT, VOC, molecularly imprinted polymer, carbon nanotubes, antenna sensor, microwavesensor, dual-functional, gas sensor, selective, volatile organic compounds, mixture sensor, ethanol, isopropanol, butanol, wireless sensor network, neural network, machinelearning, decoupling and matching network

HASAN, M.; ALEV, O.; ŠKRABÁNEK, P.; SOUKUPOVA, G.; HASSOUNA, F.; CHEFFENA, M.

25.08.2025

ACS Sensors

10

9

United States of America

6446

6463

18

https://pubs.acs.org/doi/full/10.1021/acssensors.5c00243