The implementation of low-cost sensitive and selective gas sensors for monitoring fruit ripening and quality strongly depends on their long-term stability. Gas sensor drift undermines the long-term reliability of low-cost sensing platforms, particularly in precision agriculture. We present a real-time drift compensation framework based on a lightweight Temporal Convolutional Neural Network (TCNN) combined with a Hadamard spectral transform. The model operates causally on incoming sensor data, achieving a mean absolute error below 1 mV on long-term recordings (equivalent to <1 particle per million (ppm) gas concentration). Through quantization, we compress the model by over 70%, without sacrificing accuracy. Demonstrated on a combustion-type gas sensor system (dubbed GMOS) for ethylene monitoring, our approach enables continuous, drift-corrected operation without the need for recalibration or dependence on cloud-based services, offering a generalizable solution for embedded environmental sensing—in food transportation containers, cold storage facilities, de-greening rooms and directly in the field.

The need to achieve digital gas sensing technology, namely, a technology to sense and transmit gas-enabled digital media, has been recognized as highly challenging. This challenge has motivated the authors to focus on complementary metal oxide semiconductor silicon on insulator micro electro-mechanical system (CMOS-SOI-MEMS) technologies, and the result is a new pellistor-like sensor, dubbed GMOS, with integrated signal processing. In this study, we describe the performance of such sensors for the selective detection of mixtures of gases. The novel key ideas of this study are: (i) the use of the GMOS for gas sensing; (ii) applying the Kalman filter to improve the signal-to-noise ratio; (iii) adding artificial intelligence (AI) with tiny edge approach.

This is the fourth part of a study presenting a miniature, combustion-type gas sensor (dubbed GMOS) based on a novel thermal sensor (dubbed TMOS). The TMOS is a micromachined CMOS-SOI transistor, which acts as the sensing element and is integrated with a catalytic reaction plate, where ignition of the gas takes place. The GMOS measures the temperature change due to a combustion exothermic reaction. The controlling parameters of the sensor are the ignition temperature applied to the catalytic layer and the increased temperature of the hotplate due to the released power of the combustion reaction. The solid-state device applies electrical parameters, which are related to the thermal parameters. The heating is applied by Joule heating with a resistor underneath the catalytic layer while the signal is monitored by the change in voltage of the TMOS sensor. Voltage, like temperature, is an intensive parameter, and one always measures changes in such parameters relative to a reference point. The reference point for both parameters (temperature and voltage) is the blind sensor, without any catalytic layer and hence where no reaction takes place. The present paper focuses on the study of the effect of humidity upon performance. In real life, the sensors are exposed to environmental parameters, where humidity plays a significant role. Humidity is high in storage rooms of fruits and vegetables, in refrigerators, in silos, in fields as well as in homes and cars. This study is significant and innovative since it extends our understanding of the performance of the GMOS, as well as pellistor sensors in general, in the presence of humidity. The three main challenges in simulating the performance are (i) how to define the operating temperature based on the input parameters of the heater voltage in the presence of humidity; (ii) how to measure the dynamics of the temperature increase during cyclic operation at a given duty cycle; and (iii) how to model the correlation between the operating temperature and the sensing response in the presence of humidity. Due to the complexity of the 3D analysis of packaged GMOS, and the many aspects of humidity simultanoesuly affecting performane, advanced simulation software is applied, incorporating computational fluid dynamics (CFD). The simulation and experimental data of this study show that the GMOS sensor can operate in the presence of high humidity.

Embedded machine learning, TinyML, is a relatively new and fast-growing field of ML, enabling on-device sensor data analytics at low power requirements. This paper presents possible improvements to GMOS, a gas sensor, using TinyML technology. GMOS is a low-cost catalytic gas sensor, fabricated with the standard CMOS-SOI process, based on a suspended thermal transistor MOS (TMOS). Exothermic combustion reactions lead to temperature increases, which modify the suspended transistor’s (used as the sensing element) current-voltage characteristics. We were able to use GMOS measurements for gas classification (both for gas types, as well as concentration), resulting in high-proficiency gas detection at a low cost. Our preliminary results show great successes in the detection of ethanol and acetone gases. Moreover, we believe the method could be generalized to more gas types, concentrations, and gas mixes in future research.