The significance of intelligent sensor systems has grown across diverse sectors, including healthcare, environmental surveillance, industrial automation, and security. Photoacoustic gas sensors are a promising type of optical gas sensor due to their high sensitivity, enhanced frequency selectivity, and fast response time. However, they have limitations such as dependence on a high-power light source, a requirement for a high-quality acoustic signal detector, and sensitivity to environmental factors, affecting their accuracy and reliability. Machine learning has great potential in the analysis and interpretation of sensor data as it can identify complex patterns and make accurate predictions based on the available data. We propose a novel approach that utilizes wavelet analysis and neural networks with enhanced architectures to improve the accuracy and sensitivity of photoacoustic gas sensors. Our proposed approach was experimentally tested for methane concentration measurements, showcasing its potential to significantly advance the field of gas detection and analysis, providing more accurate and reliable results.

In this study, we address the challenge of estimating the resonance frequency of a photoacoustic detector (PAD) gas cell under varying temperature conditions, which is crucial for improving the accuracy of gas concentration measurements. We introduce a novel approach that uses a long short-term memory network and a self-attention mechanism to model resonance frequency shifts based on temperature data. To investigate the impact of the gas mixture temperature on the resonance frequency, we modified the PAD to include an internal temperature sensor. Our experiments involved multiple heating and cooling cycles with varying methane concentrations, resulting in a comprehensive dataset of temperature and resonance frequency measurements. The proposed models were trained and validated on this dataset, and the results demonstrate real-time prediction capabilities with a mean absolute error of less than 1 Hz for frequency shifts exceeding 30 Hz over four-hour periods. This approach allows continuous, real-time tracking of the resonance frequency without interrupting the laser operation, significantly enhancing gas concentration measurements and contributing to the long-term stabilization of the sensor. The results suggest that the proposed approach is effective in managing temperature-induced frequency shifts, making it a valuable tool for improving the accuracy and stability of gas sensors in practical applications.

Distributed acoustic sensing (DAS) systems enable real-time monitoring of physical events across extended areas using optical fiber that detects vibrations through changes in backscattered light patterns. In perimeter security applications, these systems must accurately distinguish between legitimate activities and potential security threats by analyzing complex spatio-temporal data patterns. However, the high dimensionality and noise content of raw DAS data presents significant challenges for effective feature extraction and event classification, particularly when computational efficiency is required for real-time deployment. Traditional approaches or current machine learning methods often struggle with the balance between information preservation and computational complexity. This study addresses the critical need for efficient and accurate feature extraction methods that can identify informative signal components while maintaining real-time processing capabilities in DAS-based security systems. Here we show that wavelet packet decomposition (WPD) combined with a cascaded machine learning approach achieves 98% classification accuracy while reducing computational load through intelligent channel selection and preliminary filtering. Our modified peak signal-to-noise ratio metric successfully identifies the most informative frequency bands, which we validate through comprehensive neural network experiments across all possible WPD channels. The integration of principal component analysis with logistic regression as a preprocessing filter eliminates a substantial portion of non-target events while maintaining high recall level, significantly improving upon methods that processed all available data. These findings establish WPD as a powerful preprocessing technique for distributed sensing applications, with immediate applications in critical infrastructure protection. The demonstrated gains in computational efficiency and accuracy improvements suggest broad applicability to other pattern recognition challenges in large-scale sensor networks, seismic monitoring, and structural health monitoring systems, where real-time processing of high-dimensional acoustic data is essential.

The global market for infrastructure security systems based on distributed acoustic sensors is rapidly expanding, driven by the need for timely detection and prevention of potential threats. However, deploying these systems is challenging due to the high costs associated with dataset creation. Additionally, advanced signal processing algorithms are necessary for accurately determining the location and nature of detected events. In this paper, we present an enhanced approach based on semi-supervised learning for developing event classification models tailored for real-time and continuous perimeter monitoring of infrastructure facilities. The proposed method leverages a hybrid architecture combining an autoencoder and a classifier to enhance the accuracy and efficiency of event classification. The autoencoder extracts essential features from raw data using unlabeled data, improving the model’s ability to learn meaningful representations. The classifier, trained on labeled data, recognizes and classifies specific events based on these features. The integrated loss function incorporates elements from both the autoencoder and the classifier, guiding the autoencoder to extract features relevant for accurate event classification. Validation using real-world datasets demonstrates that the proposed method achieves recognition performance comparable to the baseline model, while requiring less labeled data and employing a simpler architecture. These results offer practical insights for reducing deployment costs, enhancing system performance, and increasing throughput for new deployments.