Search

The design and analysis of gas detection chips directly affect their detection efficiency and applicability. Detection devices are currently restricted by detection principles, facing drawbacks like intricate structural design, limited applicability, and low detection efficiency. We have designed a complete set of design and analysis scheme for a peptide gas detection chip. First, we selected specific and high-affinity peptide combinations from existing peptide-gas affinity datasets. Then, the peptide chip's arrangement was grouped according to the variations in peptides' affinity towards different gases. Peptides were arranged based on their affinity levels within each group, striking a balance between discrimination and flexibility in the design of the chip. Finally, we evaluated the analysis methods by generating simulated data based on a reference affinity matrix constructed from actual data. Due to the preprocessing role of chip design on affinity data, all methods can effectively accomplish gas classification. In gas concentration prediction tasks, our method reduced mean square error to 0.41, significantly outperforming other methods. This gas detection scheme shortens the development cycle of chip design and analysis methods, fully utilizing the specificity of peptides, enhancing gas analysis effectiveness, and demonstrating the agile development of gas detection chips.

Metal oxide sensor is widely used in many research fields, including E-nose for gas detection due to their tunable sensitivity, space efficiency and low cost. One of the most popular open data sets in electronic nose research contains data on various gases sampled using a MOx sensor in a wind tunnel over 16 months. A recent study published in 2022 by Nik Dennler has reported the discovery of the drift effect of a public dataset due to incorrect experimental design. they reported that the order of gas collection was not randomized and further discovered that a select set of gases were collected over a particular period. This paper expands the previous paper, by analyzing the drift effect with low signal, zero-offset subtracted signal’s mean, and standard deviation value by location and time, and examining it with TSNE, a dimensional reduction method. In addition, the accuracy by time and location was analyzed by applying it to various Deep Learning methods. According to the results, we confirmed that gas information is already classified before the gas leaks in terms of temporal and spatial domain. Therefore, the classification accuracy overestimates the actual accuracy that can be obtained due to the drift effect. Based on the results of this study, it is necessary to thoroughly verify the temporal and spatial validity of the gas dataset when using the publicly available gas dataset to develop gas detection algorithms.

The sensitivity and selectivity profiles of gas sensors are always changed by sensor drifting, sensor aging, and the surroundings (e.g., temperature and humidity changes), which lead to a serious decline in gas recognition accuracy or even invalidation. To address this issue, the practical solution is to retrain the network to maintain performance, leveraging its rapid, incremental online learning capacity. In this paper, we develop a bio-inspired spiking neural network (SNN) to recognize nine types of flammable and toxic gases, which supports few-shot class-incremental learning, and can be retrained quickly with a new gas at a low accuracy cost. Compared with gas recognition approaches such as support vector machine (SVM), k-nearest neighbor (KNN), principal component analysis (PCA) +SVM, PCA+KNN, and artificial neural network (ANN), our network achieves the highest accuracy of 98.75% in five-fold cross-validation for identifying nine types of gases, each with five different concentrations. In particular, the proposed network has a 5.09% higher accuracy than that of other gas recognition algorithms, which validates its robustness and effectiveness for real-life fire scenarios. [ABSTRACT FROM AUTHOR]

With the development of the field of e-nose research, the potential for application is increasing in various fields, such as leak measurement, environmental monitoring, and virtual reality. In this study, we characterize electronic nose data as structured data and investigate and analyze the learning efficiency and accuracy of deep learning models that use unstructured data. For this purpose, we use the MOX sensor dataset collected in a wind tunnel, which is one of the most popular public datasets in electronic nose research. Additionally, a gas detection platform was constructed using commercial sensors and embedded boards, and experimental data were collected in a hood environment such as used in chemical experiments. We investigated the accuracy and learning efficiency of deep learning models such as deep learning networks, convolutional neural networks, and long short-term memory, as well as boosting models, which are robust models on structured data, using both public and specially collected datasets. The results showed that the boosting models had a faster and more robust performance than deep learning series models.

In recent years, the application of Deep Neural Networks to gas recognition has been developing. The classification performance of the Deep Neural Network depends on the efficient representation of the input data samples. Therefore, a variety of filtering methods are firstly adopted to smooth filter the gas sensing response data, which can remove redundant information and greatly improve the performance of the classifier. Additionally, the optimization experiment of the Savitzky–Golay filtering algorithm is carried out. After that, we used the Gramian Angular Summation Field (GASF) method to encode the gas sensing response data into two-dimensional sensing images. In addition, data augmentation technology is used to reduce the impact of small sample numbers on the classifier and improve the robustness and generalization ability of the model. Then, combined with fine-tuning of the GoogLeNet neural network, which owns the ability to automatically learn the characteristics of deep samples, the classification of four gases has finally been realized: methane, ethanol, ethylene, and carbon monoxide. Through setting a variety of different comparison experiments, it is known that the Savitzky–Golay smooth filtering pretreatment method effectively improves the recognition accuracy of the classifier, and the gas recognition network adopted is superior to the fine-tuned ResNet50, Alex-Net, and ResNet34 networks in both accuracy and sample processing times. Finally, the highest recognition accuracy of the classification results of our proposed route is 99.9%, which is better than other similar work. [ABSTRACT FROM AUTHOR]

Gas recognition by electronic noses under mixed interference is a challenging problem. We propose correlation analysis for robust gas recognition by calculating the similarity of signals between target gases and mixtures. The gas sensing datasets were clustered according to the values of correlation coefficients with the target gases. The correlation analysis outperformed neural networks and other clustering algorithms on robust gas recognition under mixed interference. The correlation analysis maintained 100% accuracy even with a response change of about 40% up to an interference ratio of 13%. The excellent performance of correlation analysis can be ascribed to its powerful capacity for measuring the similarity between signals via relative variation. Correlation analysis is suggested to be a robust clustering algorithm for gas recognition. [ABSTRACT FROM AUTHOR]

With the rising industrialization, monitoring of chemical gases such as carbon dioxide, methane, and sulfur dioxide has become critical due to their profound impact on the global environment, human health, and industrial development. Recently, the Uncooled Snapshot Infrared Spectrometer (USIRS) has been employed for rapid, long-distance identification and concentration sensing of various gases. However, these uncooled instruments face limitations in gas detection accuracy due to weak signals and a low signal-to-noise ratio. To address these issues, we introduce the Cross-talk Transformer, which leverages the attention mechanism to learn deep correlations between the target gas and the gas spectral library. Furthermore, we have developed a comprehensive radiation transfer model for USIRS to generate substantial data, thereby adequately training the Cross-talk Transformer. Upon validation, our method achieves a recognition accuracy of 98.63% on experimental data for 11 types of chemical gases. In terms of concentration prediction, our approach attains a mean error of 330 ppm for chemical gases with concentrations up to 30,000 ppm. These results highlight the high accuracy and robustness of our method in monitoring gas types and concentrations, demonstrating its potential for industrial monitoring and other applications. • A physical model generating data with various gas performances for deep learning. • Gas identification and concentration prediction by a cross-talk transformer framework. • A new dataset validating precise gas detection performance by cross-talk transformer. [ABSTRACT FROM AUTHOR]

Gas recognition is essential in an electronic nose (E-nose) system, which is responsible for recognizing multivariate responses obtained by gas sensors in various applications. Over the past decades, classical gas recognition approaches such as principal component analysis (PCA) have been widely applied in E-nose systems. In recent years, artificial neural network (ANN) has revolutionized the field of E-nose, especially spiking neural network (SNN). In this paper, we investigate recent gas recognition methods for E-nose, and compare and analyze them in terms of algorithms and hardware implementations. We find each classical gas recognition method has a relatively fixed framework and a few parameters, which makes it easy to be designed and perform well with limited gas samples, but weak in multi-gas recognition under noise. While ANN-based methods obtain better recognition accuracy with flexible architectures and lots of parameters. However, some ANNs are too complex to be implemented in portable E-nose systems, such as deep convolutional neural networks (CNNs). In contrast, SNN-based gas recognition methods achieve satisfying accuracy and recognize more types of gases, and could be implemented with energy-efficient hardware, which makes them a promising candidate in multi-gas identification. [ABSTRACT FROM AUTHOR]

The sensitivity and selectivity profiles of gas sensors are always changed by sensor drifting, sensor aging, and the surroundings (e.g., temperature and humidity changes), which lead to a serious decline in gas recognition accuracy or even invalidation. To address this issue, the practical solution is to retrain the network to maintain performance, leveraging its rapid, incremental online learning capacity. In this paper, we develop a bio-inspired spiking neural network (SNN) to recognize nine types of flammable and toxic gases, which supports few-shot class-incremental learning, and can be retrained quickly with a new gas at a low accuracy cost. Compared with gas recognition approaches such as support vector machine (SVM), k-nearest neighbor (KNN), principal component analysis (PCA) +SVM, PCA+KNN, and artificial neural network (ANN), our network achieves the highest accuracy of 98.75% in five-fold cross-validation for identifying nine types of gases, each with five different concentrations. In particular, the proposed network has a 5.09% higher accuracy than that of other gas recognition algorithms, which validates its robustness and effectiveness for real-life fire scenarios.

In recent years, the application of Deep Neural Networks to gas recognition has been developing. The classification performance of the Deep Neural Network depends on the efficient representation of the input data samples. Therefore, a variety of filtering methods are firstly adopted to smooth filter the gas sensing response data, which can remove redundant information and greatly improve the performance of the classifier. Additionally, the optimization experiment of the Savitzky–Golay filtering algorithm is carried out. After that, we used the Gramian Angular Summation Field (GASF) method to encode the gas sensing response data into two-dimensional sensing images. In addition, data augmentation technology is used to reduce the impact of small sample numbers on the classifier and improve the robustness and generalization ability of the model. Then, combined with fine-tuning of the GoogLeNet neural network, which owns the ability to automatically learn the characteristics of deep samples, the classification of four gases has finally been realized: methane, ethanol, ethylene, and carbon monoxide. Through setting a variety of different comparison experiments, it is known that the Savitzky–Golay smooth filtering pretreatment method effectively improves the recognition accuracy of the classifier, and the gas recognition network adopted is superior to the fine-tuned ResNet50, Alex-Net, and ResNet34 networks in both accuracy and sample processing times. Finally, the highest recognition accuracy of the classification results of our proposed route is 99.9%, which is better than other similar work.

The Surface-Enhanced Raman Scattering (SERS) technique is utilized to fabricate sensors for gas detection due to its rapid detection speed and high sensitivity. However, gases with similar molecular structures are difficult to directly discriminate using SERS gas sensors because there are characteristic peak overlaps in the Raman spectra. Here, we proposed a multiple SERS gas sensor matrix via a spin-coating functional polymer to enhance the gas recognition capability. Poly (acrylic acid) (PAA), Poly (methyl methacrylate) (PMMA) and Polydimethylsiloxane (PDMS) were employed to fabricate the polymer film. The high design flexibility of the two-layer film was realized by the layer-by-layer method with 2 one-layer films. The SERS gas sensor coated by different polymer films showed a distinct affinity to target gases. The principle component analysis (PCA) algorithm was used for the further clustering of gas molecules. Three target gases, phenethyl alcohol, acetophenone and anethole, were perfectly discriminated, as the characteristic variables in the response matrix constructed by the combination of gas responses obtained 3 one-layer and 3 two-layer film-coated sensors. This research provides a new SERS sensing approach for recognizing gases with similar molecular structures.

With the development of the field of e-nose research, the potential for application is increasing in various fields, such as leak measurement, environmental monitoring, and virtual reality. In this study, we characterize electronic nose data as structured data and investigate and analyze the learning efficiency and accuracy of deep learning models that use unstructured data. For this purpose, we use the MOX sensor dataset collected in a wind tunnel, which is one of the most popular public datasets in electronic nose research. Additionally, a gas detection platform was constructed using commercial sensors and embedded boards, and experimental data were collected in a hood environment such as used in chemical experiments. We investigated the accuracy and learning efficiency of deep learning models such as deep learning networks, convolutional neural networks, and long short-term memory, as well as boosting models, which are robust models on structured data, using both public and specially collected datasets. The results showed that the boosting models had a faster and more robust performance than deep learning series models.

Because of the low accuracy of the current machine olfactory algorithms in detecting two mixed gases, this study proposes a hybrid gas detection algorithm based on an extreme random tree to greatly improve the classification accuracy and time efficiency. The method mainly uses the dynamic time warping algorithm (DTW) to perform data pre-processing and then extracts the gas characteristics from gas signals at different concentrations by applying a principal component analysis (PCA). Finally, the model is established by using a new extreme random tree algorithm to achieve the target gas classification. The sample data collected by the experiment was verified by comparison experiments with the proposed algorithm. The analysis results show that the proposed DTW algorithm improves the gas classification accuracy by 26.87%. Compared with the random forest algorithm, extreme gradient boosting (XGBoost) algorithm and gradient boosting decision tree (GBDT) algorithm, the accuracy rate increased by 4.53%, 5.11% and 8.10%, respectively, reaching 99.28%. In terms of the time efficiency of the algorithms, the actual runtime of the extreme random tree algorithm is 66.85%, 90.27%, and 81.61% lower than that of the random forest algorithm, XGBoost algorithm, and GBDT algorithm, respectively, reaching 103.2568 s. [ABSTRACT FROM AUTHOR]

Electronic nose (EN) is widely used to identify and detect gases. However, ENs are inevitably subject to sensor drift and the intrusion of unknown gases during prolonged exposure to the environment. A combined model of sensor drift compensation and open-set gas recognition, class anchoring and contrasting semantic alignment (CACSA) was proposed and tested on two commonly used EN drift datasets. Firstly, the model learns an embedding subspace that is discriminative, where mapped domains are semantically aligned and maximally separated. In addition, the method requires a very small number of labeled target training samples, even if one per class can be effective. Then, our model learns a fixed anchor for each known class. Known classes form tight clusters around the centers of class dependencies anchored in the feature space. CACSA can be degenerated into a closed-set drift compensation model without sacrificing accuracy and not require the additional target domain sample labeling and repeated training for open-set recognition. According to the experimental result, our model showed better drift compensation performance, and achieved the highest open-set recognition accuracy. • A combined model is proposed to realize anti-drift and open-set gas recognition. • Domain adaptation is adopted to solve sensor drift under open-set recognition. • Our model shows excellent performance on two E-nose datasets with drift. [ABSTRACT FROM AUTHOR]

Drift is an important issue that impairs the reliability of sensors, especially in gas sensors. The conventional method usually adopts the reference gas to compensate for the drift. However, its classification accuracy is not high. We propose a supervised learning algorithm that is based on multi-classifier integration for drift compensation in this paper, which incorporates drift compensation into the classification process, motivated by the fact that the goal of drift compensation is to improve the classification performance. In our method, with the obtained characteristics of sensors and the advantage of Support Vector Machine (SVM) in few-shot classification, the improved Long Shot Term Memory (LSTM) is integrated to build the multi-class classifier model. We tested the proposed approach on the publicly available time series dataset that was collected over three years by the metal-oxide gas sensors. The results clearly indicate the superiority of multiple classifier approach, which achieves higher classification accuracy as compared with different approaches during testing period with an ensemble of classifiers in the presence of sensor drift over time.

Because of the low accuracy of the current machine olfactory algorithms in detecting two mixed gases, this study proposes a hybrid gas detection algorithm based on an extreme random tree to greatly improve the classification accuracy and time efficiency. The method mainly uses the dynamic time warping algorithm (DTW) to perform data pre-processing and then extracts the gas characteristics from gas signals at different concentrations by applying a principal component analysis (PCA). Finally, the model is established by using a new extreme random tree algorithm to achieve the target gas classification. The sample data collected by the experiment was verified by comparison experiments with the proposed algorithm. The analysis results show that the proposed DTW algorithm improves the gas classification accuracy by 26.87%. Compared with the random forest algorithm, extreme gradient boosting (XGBoost) algorithm and gradient boosting decision tree (GBDT) algorithm, the accuracy rate increased by 4.53%, 5.11% and 8.10%, respectively, reaching 99.28%. In terms of the time efficiency of the algorithms, the actual runtime of the extreme random tree algorithm is 66.85%, 90.27%, and 81.61% lower than that of the random forest algorithm, XGBoost algorithm, and GBDT algorithm, respectively, reaching 103.2568 s.

This work deals with the problem of volatile chemical classification with an electronic nose (e-nose), and particularly focuses on the case where the e-nose is not collecting samples in a stationary fashion but is carried by a moving platform (mobile robot, car, bike, etc.). We bring to light that, under these specific circumstances, substantial changes in the transient response of the gas sensors arise (something that has not been considered until now). We experimentally demonstrate that these changes in the sensor's response have an important impact on the classification accuracy if not properly considered, resulting in a decrease of up to 30% in some configurations. We back our conclusions with an extensive experimental evaluation consisting of a mobile robot inspecting a long indoor corridor with two chemical volatiles sources (ethanol and acetone) more than 240 times, at four different motion speeds. The paper also reveals the relevance of training the classifiers with data collected in motion, and proposes different training schemes suitable to this problem. [ABSTRACT FROM AUTHOR]

Metal oxide (MOx) gas sensors are a popular choice for many applications, due to their tunable sensitivity, space efficiency and low cost. Publicly available sensor datasets are particularly valuable for the research community as they accelerate the development and evaluation of novel algorithms for gas sensor data analysis. A dataset published in 2013 by Vergara and colleagues contains recordings from MOx gas sensor arrays in a wind tunnel. It has since become a standard benchmark in the field. Here we report a latent property of this dataset that limits its suitability for gas classification studies. Measurement timestamps show that gases were recorded in separate, temporally clustered batches. Sensor baseline response before gas exposure were strongly correlated with the recording batch, to the extent that baseline response was largely sufficient to infer the gas used in a given trial. Zero-offset baseline compensation did not resolve the issue, since residual short-term drift still contained enough information for gas/trial identification using a machine learning classifier. A subset of the data recorded within a short period of time was minimally affected by drift and suitable for gas classification benchmarking after offset-compensation, but with much reduced classification performance compared to the full dataset. We found 18 publications where this dataset was used without precautions against the circumstances we describe, thus potentially overestimating the accuracy of gas classification algorithms. These observations highlight potential pitfalls in using previously recorded gas sensor data, which may have distorted widely reported results. • Limitations found in popular MOx gas sensor dataset due to unaccounted sensor drift. • Baselines recordings before gas exposure are sufficient for accurate gas classification. • Previously reported gas classification results may be systematically overestimated. • We give recommendations to avoid such problems in the future. [ABSTRACT FROM AUTHOR]

Electronic nose (E-Nose) has been widely used in detection and classification of gases. The learning models of traditional E-Noses are generally limited in closed-set environment: the training and test samples share the same label spaces. However, a more challenging and realistic scenario of E-Noses is open-set learning, where the test samples contains classes unseen during the model training. This study investigated the possibility of open-set learning models for the recognition and classification of gases based on a public electronic nose datasets. The dataset includes the response of a 72-channels MOS sensor array on 10 gaseous substances. The original data was preprocessed by two methods: one is to manually extract features from the response curve of each sample, and the other is to down-sample the original sample into a matrix. Then multilayer perceptron (MLP) and convolution neural network (CNN) were used to extract the feature vectors of the data processed by the two processing methods respectively. The performance of four different open-set recognition models, including softmax threshold (ST), OpenMax, extreme value machine (EVM) and class anchor clustering (CAC), was compared based on the feature vectors obtained from two neural networks. To understand the effect of sensor drift on the models, we also validated the models on a commonly used sensor drift dataset. The results demonstrated that for the open-set detection task, the CNN-based CAC (CAC-CNN) outperformed the other methods. For the closed-set recognition task, the CNN-based classification model achieved higher accuracy. On sensor drift dataset, the performance of open-set recognition models has decreased a lot, and it seems that drift has a large negative impact on the open-set gas recognition. • Open-set gas recognition method is proposed to classify the known and reject the unknown gases with high accuracy. • The adopted convolution neural network (CNN) can achieve better feature extraction and closed-set recognition. • The open-set recognition algorithm class anchor clustering (CAC) has the best open-set gas recognition performance. [ABSTRACT FROM AUTHOR]

Abstract: Data processing methods commonly used in conjunction with the array- and quasi-array-based gas identification systems generally include a dimension reduction followed by categorization using a classifier. Here, we have applied an ensemble of classifiers, directly to the high dimensional feature vectors and fused their verdicts by majority voting. The quasi-array investigated is a metal oxide sensor temperature-modulated with different rectangular heating voltage pulses. The experimental database was developed by recording the temporal responses obtained at different conditions to methanol, ethanol and 1-butanol vapors. Features related to each response were extracted by wavelet transform. The classification rates achieved with traditional methods were compared to that obtained using an ensemble of classifiers. The classification rate was improved by majority voting among the classifiers, each trained on different feature subsets, for the classification verdict. [Copyright &y& Elsevier]

Abstract: The design and high-level modeling of a multi-channel integrated circuit for infrared gas recognition suitable for low-power consumption are presented. The thermal compensation and independent programmability methods are used in all the main channel stages: sensor biasing, high-pass filtering and pre-amplification, blind cancellation, lock-in demodulation and A/D conversion. CMOS transistor subthreshold operation and circuit reuse techniques are widely applied in order to obtain a low-power and compact channel realization. Experimental results are presented for a 340×880μm channel integrated in 0.35μm CMOS technology reporting good agreement with electrical and higher level simulations. [Copyright &y& Elsevier]

There are increasing demands for gas sensors in various fields such as food, cosmetics, and safety applications, and a method is required which can recognize the existence of a gas quickly and with high accuracy, as well as a method to measure the gas concentration quantitatively. In the system proposed in this paper, the gas is recognized and its concentration is estimated by using a quartz crystal resonator sensor. The sensor can detect various gases, but may require a long time, depending on the type of gas, due to its operating principle. In the proposed method, the response of the sensor is monitored in time division, considering the time from the start of the response to detection, so that the gas is detected in a short time. In addition, a procedure is included in which the certainty of the results of recognition is examined, so that the gas type can be determined with stability. The effectiveness of the proposed system is shown by applying the method to common organic solvent gases. © 2000 Scripta Technica, Syst Comp Jpn, 31(6); 45–55, 2000 [ABSTRACT FROM AUTHOR]

The Surface-Enhanced Raman Scattering (SERS) technique is utilized to fabricate sensors for gas detection due to its rapid detection speed and high sensitivity. However, gases with similar molecular structures are difficult to directly discriminate using SERS gas sensors because there are characteristic peak overlaps in the Raman spectra. Here, we proposed a multiple SERS gas sensor matrix via a spin-coating functional polymer to enhance the gas recognition capability. Poly (acrylic acid) (PAA), Poly (methyl methacrylate) (PMMA) and Polydimethylsiloxane (PDMS) were employed to fabricate the polymer film. The high design flexibility of the two-layer film was realized by the layer-by-layer method with 2 one-layer films. The SERS gas sensor coated by different polymer films showed a distinct affinity to target gases. The principle component analysis (PCA) algorithm was used for the further clustering of gas molecules. Three target gases, phenethyl alcohol, acetophenone and anethole, were perfectly discriminated, as the characteristic variables in the response matrix constructed by the combination of gas responses obtained 3 one-layer and 3 two-layer film-coated sensors. This research provides a new SERS sensing approach for recognizing gases with similar molecular structures. [ABSTRACT FROM AUTHOR]

Robot olfaction takes inspiration from animals for locating a gas source in the environment, such as a gas leak to fix or an explosive to neutralize. In these cases, gas sources emit Volatile Organic Compounds (VOCs) which can be measured with an electronic nose. This instrument can detect a broad variety of VOCs, so the same device can then be used for many different applications. In a realistic environment, several VOCs of interest can be present at the same time and mix. This creates difficulties for gas recognition, and in the literature, the problem is often ignored. In this article, we deal both with gas recognition of a large number of VOCs and gas unmixing. For that, we use a recently developed optoelectronic nose which uses peptides as sensing materials and Surface Plasmon Resonance imaging as transduction method. We present two different setups. The first setup studies the recognition of 24 gas sources of 12 VOCs disseminated in the environment. The second setup studies various realistic scenarios in which mixtures occur, due to gas sources being spatially close. We propose a real-time dictionary-based algorithm for dealing both with gas recognition and gas unmixing. We succeed in obtaining a score of 73% for the gas recognition task, meaning that 73% of the 24 gas sources have been well identified over several runs. For the unmixing issue, we correctly identify the VOCs composing the mixtures. However, we also show that this performance is strongly related to the VOCs used in the dictionary. • A new SPR-based electronic nose which boards peptides as sensors is used in a robotics application. • We propose a real-time baseline drift correction and a real-time dictionary-based algorithm for gas recognition and unmixing. • We describe two open sampling systems we use to evaluate the recognition of 24 gas sources of 12 VOCs and the unmixing of binary and ternary mixtures. • Our algorithms successfully identify 73% of the gas sources in the gas recognition task. • Using the unmixing algorithm, we can map gas sources despite significant mixing, provided that the pure compounds are resolved well-enough by the instrument. [ABSTRACT FROM AUTHOR]

Drift is an important issue that impairs the reliability of sensors, especially in gas sensors. The conventional method usually adopts the reference gas to compensate for the drift. However, its classification accuracy is not high. We propose a supervised learning algorithm that is based on multi-classifier integration for drift compensation in this paper, which incorporates drift compensation into the classification process, motivated by the fact that the goal of drift compensation is to improve the classification performance. In our method, with the obtained characteristics of sensors and the advantage of Support Vector Machine (SVM) in few-shot classification, the improved Long Shot Term Memory (LSTM) is integrated to build the multi-class classifier model. We tested the proposed approach on the publicly available time series dataset that was collected over three years by the metal-oxide gas sensors. The results clearly indicate the superiority of multiple classifier approach, which achieves higher classification accuracy as compared with different approaches during testing period with an ensemble of classifiers in the presence of sensor drift over time. [ABSTRACT FROM AUTHOR]

In this work, we present a high-performance smart electronic nose (E-nose) system consisting of a multiplexed tin oxide (SnO 2 ) nanotube sensor array, read-out circuit, wireless data transmission unit, mobile phone receiver, and data processing application (App). Using the designed nanotube sensor device structure in conjunction with multiple electrode materials, high-sensitivity gas detection and discrimination have been achieved at room temperature, enabling a 1000 times reduction of the sensor's power consumption as compared to a conventional device using thin film SnO 2 . The experimental results demonstrate that the developed E-nose can identify indoor target gases using a simple vector-matching gas recognition algorithm. In addition, the fabricated E-nose has achieved state-of-the-art sensitivity for H 2 and benzene detection at room temperature with metal oxide sensors. Such a smart E-nose system can address the imperative needs for distributed environmental monitoring in smart homes, smart buildings, and smart cities.