Search

Recent advances in emotion recognition through Artificial Intelligence (AI) have demonstrated potential applications in various fields (e.g., healthcare, advertising, and driving technology), with electroencephalogram (EEG)-based approaches demonstrating superior accuracy compared to facial or vocal methods due to their resistance to intentional manipulation. This study presents a novel approach to enhance EEG-based emotion estimation accuracy by emphasizing temporal features and efficient parameter space exploration. We propose a model combining Long Short-Term Memory (LSTM) with an attention mechanism to highlight temporal features in EEG data while optimizing LSTM parameters through Particle Swarm Optimization (PSO). The attention mechanism assigned weights to LSTM hidden states, and PSO dynamically optimizes the vital parameters, including units, batch size, and dropout rate. Using the DEAP and SEED datasets, which serve as benchmark datasets for emotion estimation research using EEG, we evaluate the model's performance. For the DEAP dataset, we conduct a four-class classification of combinations of high and low valence and arousal states. We perform a three-class classification of negative, neutral, and positive emotions for the SEED dataset. The proposed model achieves an accuracy of 0.9409 on the DEAP dataset, surpassing the previous state-of-the-art accuracy of 0.9100 reported by Lin et al. The model attains an accuracy of 0.9732 on the SEED dataset, recording one of the highest accuracies among the related research. These results demonstrate that integrating the attention mechanism with PSO significantly improves the accuracy of EEG-based emotion estimation, contributing to the advancement of emotion recognition technology. [ABSTRACT FROM AUTHOR]

Satellite SAR (synthetic aperture radar) imagery offers global coverage and all-weather recording capabilities, making it valuable for applications like remote sensing and maritime surveillance. However, its use in machine learning-based automatic target classification faces challenges, including the limited availability of SAR target training samples and the inherent constraints of SAR images, which provide less detailed features compared to natural images. These issues hinder the effective training of convolutional neural networks (CNNs) and complicate the transfer learning process due to the distinct imaging mechanisms of SAR and natural images. To address these challenges, we propose a shallow CNN architecture specifically designed to optimize performance on SAR datasets. Evaluations were performed on three datasets: FUSAR-Ship, OpenSARShip, and MSTAR. While the FUSAR-Ship and OpenSARShip datasets present difficulties due to their limited and imbalanced class distributions, MSTAR serves as a benchmark with balanced classes. To compare and optimize the proposed shallow architecture, we examine various properties of CNN components, such as the filter numbers and sizes in the convolution layers, to reduce redundancy, improve discrimination capability, and decrease network size and learning time. In the second phase of this paper, we combine the CNN with Long short-term memory (LSTM) networks to enhance SAR image classification. Comparative experiments with six state-of-the-art CNN architectures (VGG16, ResNet50, Xception, DenseNet121, EfficientNetB0, and MobileNetV2) demonstrate the superiority of the proposed approach, achieving competitive accuracy while significantly reducing training times and network complexity. This study underscores the potential of customized architectures to address SAR-specific challenges and enhance the efficiency of target classification. [ABSTRACT FROM AUTHOR]

Intrusion Detection Systems (IDSs) are a crucial component of modern corporate firewalls. The ability of IDS to identify malicious traffic is a powerful tool to prevent potential attacks and keep a corporate network secure. In this context, Machine Learning (ML)-based methods have proven to be very effective for attack identification. However, traditional approaches are not always applicable in a real-time environment as they do not integrate concrete traffic management after a malicious packet pattern has been identified. In this paper, a novel combined approach to both identify and discard potential malicious traffic in a real-time fashion is proposed. In detail, a Long Short-Term Memory (LSTM) supervised artificial neural network model is provided in which consecutive packet groups are considered as they flow through the corporate network. Moreover, the whole IDS architecture is modeled by a Petri Net (PN) that either blocks or allows packet flow throughout the network based on the LSTM model output. The novel hybrid approach combining LSTM with Petri Nets achieves a 99.71% detection accuracy—a notable improvement over traditional LSTM-only methods, which averaged around 97%. The LSTM–Petri Net approach is an innovative solution combining machine learning with formal network modeling for enhanced threat detection, offering improved accuracy and real-time adaptability to meet the rapid security needs of virtual environments and CPS. Moreover, the approach emphasizes the innovative role of the Intrusion Detection System (IDS) and Intrusion Prevention System (IPS) as a form of "virtual sensing technology" applied to advanced network security. An extensive case study with promising results is provided by training the model with the popular IDS 2018 dataset. [ABSTRACT FROM AUTHOR]

Sensor networks generate vast amounts of data in real-time, which challenges existing predictive maintenance frameworks due to high latency, energy consumption, and bandwidth requirements. This research addresses these limitations by proposing an edge-cloud hybrid framework, leveraging edge devices for immediate anomaly detection and cloud servers for in-depth failure prediction. A K-Nearest Neighbors (KNNs) model is deployed on edge devices to detect anomalies in real-time, reducing the need for continuous data transfer to the cloud. Meanwhile, a Long Short-Term Memory (LSTM) model in the cloud analyzes time-series data for predictive failure analysis, enhancing maintenance scheduling and operational efficiency. The framework's dynamic workload management algorithm optimizes task distribution between edge and cloud resources, balancing latency, bandwidth usage, and energy consumption. Experimental results show that the hybrid approach achieves a 35% reduction in latency, a 28% decrease in energy consumption, and a 60% reduction in bandwidth usage compared to cloud-only solutions. This framework offers a scalable, efficient solution for real-time predictive maintenance, making it highly applicable to resource-constrained, data-intensive environments. [ABSTRACT FROM AUTHOR]

Given the urgent challenges posed by global climate change and the ongoing energy crisis, fuel cell electric vehicles (FCEVs) have emerged as a promising solution. Incorporating sophisticated energy management strategies (EMSs) into FCEVs can significantly enhance the efficiency of the complex powertrain under diverse driving conditions. In this paper, a dual-model predictive control energy management strategy based on long short-term memory (LSTM)-based driving condition recognition is proposed to enhance the economic performance of FCEVs and robustness across diverse driving conditions. Firstly, to improve the generalization capability and adaptability of the LSTM model and to enhance the accuracy of driving condition recognition, wavelet transform (WT) is introduced into both the offline training and online application of LSTM. Secondly, to enhance the real-time performance and control effectiveness of the EMS, model predictive control (MPC) and explicit model predictive control (eMPC) are established based on a unified optimization objective and constraints. Thirdly, a dual MPC switching logic is developed using the information of driving condition prediction, ensuring the coordination of dual MPCs in practical applications and enhancing their adaptability to various conditions. Finally, an evaluation of the simulations demonstrates that the proposed dual-model predictive control energy management strategy based on wavelet transform LSTM driving condition recognition (WTL-DMPC EMS) can improve economic performance. Compared with other baselines, the energy-saving capability is remarkable, showcasing its promising performance. [ABSTRACT FROM AUTHOR]

Autism spectrum disorder (ASD) is a brain disorder causing issues among many young children. For children suffering from ASD, their learning ability is typically slower when compared to normal children. Therefore, many technologies aiming to teach ASD children with optimized learning approaches have emerged. With this motivation, this study presents a smart autism spectrum disorder learning system based on remote edge healthcare clinics and the Internet of Medical Things, the objective of which is to offer an online education and healthcare environment for autistic children. Concave and convex optimization constraints, such as accuracy, learning score, total processing time with deadline, and resource failure, are considered in the proposed system, with a focus on different autism education learning applications (e.g., speaking, reading, writing, and listening), while respecting the system's quality of service (QoS) requirements. All of the autism applications are executed on smartwatches, mobile devices, and edge healthcare nodes during their training and analysis in the system. This study presents the smartwatch autism spectrum data learning scheme (SM-ASDS), which consists of different offloading approaches, training analyses, and schemes. The SM-ASDS algorithm methodology includes partitioning offloading and deep convolutional neural network (DCNN)- and adaptive long short-term memory (ALSTM)-based schemes, which are used to train autism-related data on different nodes. The simulation results show that SM-ASDS improved the learning score by 30%, accuracy by 98%, and minimized the total processing time by 33%, when compared to baseline methods. Overall, this study presents an education learning system based on smartwatches for autistic patients, which facilitates educational training for autistic patients based on the use of artificial intelligence techniques. [ABSTRACT FROM AUTHOR]

During roadway excavation, the presence of roof deterioration zones, such as layered spaces and weak interlayers, significantly affects the stability of the surrounding rock. To achieve timely and effective support for roadways, it is essential to utilize drilling measurement signals obtained during the construction of anchorage holes for the identification and prediction of these deterioration zones. This study systematically investigates the response characteristics of thrust, torque, and Y-direction vibration signals to different combinations of rock layers through theoretical analysis, laboratory experiments, ABAQUS dynamic numerical simulations, and field measurements. The results indicate that these drilling parameters effectively characterize variations in rock structure and strength, with distinct signal features observed particularly in roof deterioration zones. Based on these findings, this paper proposes a deep learning algorithm that employs Long Short-Term Memory (LSTM) recurrent neural networks for classification prediction, along with a random forest algorithm for regression prediction, aimed at the intelligent identification and prediction of roof deterioration zones. The algorithm demonstrates outstanding performance in both laboratory experiments and field tests, achieving a 100% recognition rate for layered spaces and a 96.6% accuracy for identifying deterioration zones, with high accuracy at lower values of mechanical specific energy (MSE). The proposed method provides significant insights for real-time monitoring and control of roof deterioration zones, enhancing the safety and stability of roadway excavations, and serves as a valuable reference for future research and practical applications. [ABSTRACT FROM AUTHOR]

Inductive proximity sensors are relevant in position-sensing applications in many industries but, in order to be used in harsh industrial environments, they need to be immune to electromagnetic interference (EMI). The use of conventional filters to mitigate these perturbations often compromises signal bandwidth, ranging from 100 Hz to 1.6 kHz. We have exploited recent advances in the field of artificial intelligence (AI) to study the ability of neural networks (NNs) to automatically filter out EMI features. This study offers an analysis and comparison of possible NN models (a 1D convolutional NN, a recurrent NN, and a hybrid convolutional and recurrent approach) for denoising EMI-perturbed signals and proposes a final model, which is based on gated recurrent unit (GRU) layers. This network is compressed and optimised to meet memory requirements, so that in future developments it could be implemented in application-specific integrated circuits (ASICs) for inductive sensors. The final RNN manages to reduce noise by 70% (MSEred) while occupying 2 KB of memory. [ABSTRACT FROM AUTHOR]

A large number of sensors are typically installed in industrial plants to collect real-time operational data. These sensors monitor data with time series correlation and spatial correlation over time. In previous studies, GNN has built many successful models to deal with time series data, but most of these models have fixed perspectives and struggle to capture the dynamic correlations in time and space simultaneously. Therefore, this paper constructs a multi-scale dynamic graph neural network (MSDG) for anomaly detection in industrial sensor data. First, a multi-scale sliding window mechanism is proposed to input different scale sensor data into the corresponding network. Then, a dynamic graph neural network is constructed to capture the spatial–temporal dependencies of multivariate sensor data. Finally, the model comprehensively considers the extracted features for sequence reconstruction and utilizes the reconstruction errors for anomaly detection. Experiments have been conducted on three real public datasets, and the results show that the proposed method outperforms the mainstream methods. [ABSTRACT FROM AUTHOR]

To enhance the classification accuracy of lower limb movements, a fusion recognition model integrating a surface electromyography (sEMG)-based convolutional neural network, transformer encoder, and long short-term memory network (CNN-Transformer-LSTM, CNN-TL) was proposed in this study. By combining these advanced techniques, significant improvements in movement classification were achieved. Firstly, sEMG data were collected from 20 subjects as they performed four distinct gait movements: walking upstairs, walking downstairs, walking on a level surface, and squatting. Subsequently, the gathered sEMG data underwent preprocessing, with features extracted from both the time domain and frequency domain. These features were then used as inputs for the machine learning recognition model. Finally, based on the preprocessed sEMG data, the CNN-TL lower limb action recognition model was constructed. The performance of CNN-TL was then compared with that of the CNN, LSTM, and SVM models. The results demonstrated that the accuracy of the CNN-TL model in lower limb action recognition was 3.76%, 5.92%, and 14.92% higher than that of the CNN-LSTM, CNN, and SVM models, respectively, thereby proving its superior classification performance. An effective scheme for improving lower limb motor function in rehabilitation and assistance devices was thus provided. [ABSTRACT FROM AUTHOR]

With the increasing level of industrial informatization, massive industrial data require real-time and high-fidelity wireless transmission. Although some industrial wireless network protocols have been designed over the last few decades, most of them have limited coverage and narrow bandwidth. They cannot always ensure the certainty of information transmission, making it especially difficult to meet the requirements of low latency in industrial manufacturing fields. The 5G technology is characterized by a high transmission rate and low latency; therefore, it has good prospects in industrial applications. To apply 5G technology to factory environments with low latency requirements for data transmission, in this study, we analyze the statistical performance of the round-trip time (RTT) in a 5G-R15 communication system. The results indicate that the average value of 5G RTT is about 11 ms, which is less than the 25 ms of WIA-FA. We then consider 5G RTT data as a group of time series, utilizing the augmented Dickey–Fuller (ADF) test method to analyze the stability of the RTT data. We conclude that the RTT data are non-stationary. Therefore, firstly, the original 5G RTT series are subjected to first-order differencing to obtain differential sequences with stronger stationarity. Then, a time series analysis-based variational mode decomposition–long short-term memory (VMD-LSTM) method is proposed to separately predict each differential sequence. Finally, the predicted results are subjected to inverse difference to obtain the predicted value of 5G RTT, and a predictive error of 4.481% indicates that the method performs better than LSTM and other methods. The prediction results could be used to evaluate network performance based on business requirements, reduce the impact of instruction packet loss, and improve the robustness of control algorithms. The proposed early warning accuracy metrics for control issues can also be used to indicate when to retrain the model and to indicate the setting of the control cycle. The field of industrial control, especially in the manufacturing industry, which requires low latency, will benefit from this analysis. It should be noted that the above analysis and prediction methods are also applicable to the R16 and R17 versions. [ABSTRACT FROM AUTHOR]

The accurate prediction of gas emissions has important guiding significance for the prevention and control of gas disasters in order to further improve the prediction accuracy of gas emissions in the mining face. According to the absolute gas emission monitoring data of the 1417 working face in a coal mine in Shaanxi Province, a GA-VMD-SSA-LSTM gas emission prediction model (GVSL) based on genetic algorithm (GA)-optimized variational mode decomposition (VMD) and sparrow search algorithm (SSA)-optimized long short-term memory (LSTM) is proposed. Firstly, a VMD evaluation standard for evaluating the amount of decomposition loss is proposed. Under this standard, the GA is used to find the optimal parameters of the VMD. Then, the SSA is used to optimize the key parameters of the LSTM to establish a GVSL prediction model. The model predicts each component and finally superimposes the prediction results for each component to obtain the final gas emission result. The results show that the accuracy of the evaluation indexes of the GVSL model and VMD-LSTM model, as well as the SSA-LSTM model and Gaussian process regression (GPR) model, are compared and analyzed horizontally and vertically under three scenarios with prediction sets of 121,94 and 57 groups. The GVSL model has the best prediction effect, and its fitting degree R2 values are 0.95, 0.96, and 0.99, which confirms the effectiveness of the proposed GVSL model for the time series prediction of gas emission in the mining face. [ABSTRACT FROM AUTHOR]

From June to October, 2022, we recorded the weight, the internal temperature, and the hive entrance video traffic of ten managed honey bee (Apis mellifera) colonies at a research apiary of the Carl Hayden Bee Research Center in Tucson, AZ, USA. The weight and temperature were recorded every five minutes around the clock. The 30 s videos were recorded every five minutes daily from 7:00 to 20:55. We curated the collected data into a dataset of 758,703 records (280,760–weight; 322,570–temperature; 155,373–video). A principal objective of Part I of our investigation was to use the curated dataset to investigate the discrete univariate time series forecasting of hive weight, in-hive temperature, and hive entrance traffic with shallow artificial, convolutional, and long short-term memory networks and to compare their predictive performance with traditional autoregressive integrated moving average models. We trained and tested all models with a 70/30 train/test split. We varied the intake and the predicted horizon of each model from 6 to 24 hourly means. Each artificial, convolutional, and long short-term memory network was trained for 500 epochs. We evaluated 24,840 trained models on the test data with the mean squared error. The autoregressive integrated moving average models performed on par with their machine learning counterparts, and all model types were able to predict falling, rising, and unchanging trends over all predicted horizons. We made the curated dataset public for replication. [ABSTRACT FROM AUTHOR]

In this article, a deep neural network (DNN)-based underwater acoustic (UA) communication receiver is proposed. Conventional orthogonal frequency-division multiplexing (OFDM) receivers perform channel estimation using linear interpolation. However, due to the significant delay spread in multipath UA channels, the frequency response often exhibits strong non-linearity between pilot subcarriers. Since the channel delay profile is generally unknown, this non-linearity cannot be modeled precisely. A neural network (NN)-based receiver effectively tackles this challenge by learning and compensating for the non-linearity through NN training. The performance of the DNN-based UA communication receiver was tested recently in river trials in Western Australia. The results obtained from the trials prove that the DNN-based receiver performs better than the conventional least-squares (LS) estimator-based receiver. This paper suggests that UA communication using DNN receivers holds great potential for revolutionizing underwater communication systems, enabling higher data rates, improved reliability, and enhanced adaptability to changing underwater conditions. [ABSTRACT FROM AUTHOR]

Detection of abnormal situations in mobile systems not only provides predictions about risky situations but also has the potential to increase energy efficiency. In this study, two real-world drives of a battery electric vehicle and unsupervised hybrid anomaly detection approaches were developed. The anomaly detection performances of hybrid models created with the combination of Long Short-Term Memory (LSTM)-Autoencoder, the Local Outlier Factor (LOF), and the Mahalanobis distance were evaluated with the silhouette score, Davies–Bouldin index, and Calinski–Harabasz index, and the potential energy recovery rates were also determined. Two driving datasets were evaluated in terms of chaotic aspects using the Lyapunov exponent, Kolmogorov–Sinai entropy, and fractal dimension metrics. The developed hybrid models are superior to the sub-methods in anomaly detection. Hybrid Model-2 had 2.92% more successful results in anomaly detection compared to Hybrid Model-1. In terms of potential energy saving, Hybrid Model-1 provided 31.26% superiority, while Hybrid Model-2 provided 31.48%. It was also observed that there is a close relationship between anomaly and chaoticity. In the literature where cyber security and visual sources dominate in anomaly detection, a strategy was developed that provides energy efficiency-based anomaly detection and chaotic analysis from data obtained without additional sensor data. [ABSTRACT FROM AUTHOR]

New applications such as augmented reality/virtual reality (AR/VR), Internet-of-Things (IOT), autonomous mobile robot (AMR) services, etc., require high reliability and high accuracy real-time positioning and tracking of persons and devices in indoor areas. Among the different visible-light-positioning (VLP) schemes, such as proximity, time-of-arrival (TOA), time-difference-of-arrival (TDOA), angle-of-arrival (AOA), and received-signal-strength (RSS), the RSS scheme is relatively easy to implement. Among these VLP methods, the RSS method is simple and efficient. As the received optical power has an inverse relationship with the distance between the LED transmitter (Tx) and the photodiode (PD) receiver (Rx), position information can be estimated by studying the received optical power from different Txs. In this work, we propose and experimentally demonstrate a real-time VLP system utilizing long short-term memory neural network (LSTM-NN) with principal component analysis (PCA) to mitigate high positioning error, particularly at the positioning unit cell boundaries. Experimental results show that in a positioning unit cell of 100 × 100 × 250 cm3, the average positioning error is 5.912 cm when using LSTM-NN only. By utilizing the PCA, we can observe that the positioning accuracy can be significantly enhanced to 1.806 cm, particularly at the unit cell boundaries and cell corners, showing a positioning error reduction of 69.45%. In the cumulative distribution function (CDF) measurements, when using only the LSTM-NN model, the positioning error of 95% of the experimental data is >15 cm; while using the LSTM-NN with PCA model, the error is reduced to <5 cm. In addition, we also experimentally demonstrate that the proposed real-time VLP system can also be used to predict the direction and the trajectory of the moving Rx. [ABSTRACT FROM AUTHOR]

Gait disorders in neurological diseases are frequently associated with spasticity. Intramuscular injection of Botulinum Toxin Type A (BTX-A) can be used to treat spasticity. Providing optimal treatment with the highest possible benefit–risk ratio is a crucial consideration. This paper presents a novel approach for predicting knee and ankle kinematics after BTX-A treatment based on pre-treatment kinematics and treatment information. The proposed method is based on a Bidirectional Long Short-Term Memory (Bi-LSTM) deep learning architecture. Our study's objective is to investigate this approach's effectiveness in accurately predicting the kinematics of each phase of the gait cycle separately after BTX-A treatment. Two deep learning models are designed to incorporate categorical medical treatment data corresponding to the injected muscles: (1) within the hidden layers of the Bi-LSTM network, (2) through a gating mechanism. Since several muscles can be injected during the same session, the proposed architectures aim to model the interactions between the different treatment combinations. In this study, we conduct a comparative analysis of our prediction results with the current state of the art. The best results are obtained with the incorporation of the gating mechanism. The average prediction root mean squared error is 2.99° ( R 2 = 0.85) and 2.21° ( R 2 = 0.84) for the knee and the ankle kinematics, respectively. Our findings indicate that our approach outperforms the existing methods, yielding a significantly improved prediction accuracy. [ABSTRACT FROM AUTHOR]

Thailand's hospitals face overcrowding, particularly with non-communicable disease (NCD) patients, due to a doctor shortage and an aging population. Most literature showed implementation merely on web or mobile application to teleconsult with physicians. Instead, in this work, we developed and implemented a telemedicine health kiosk system embedded with non-invasive biosensors and time-series predictors to improve NCD indicators over an eight-month period. Two cohorts were randomly selected: a control group with usual care and a telemedicine-using group. The telemedicine-using group showed significant improvements in average fasting blood glucose (148 to 130 mg/dL) and systolic blood pressure (152 to 138 mmHg). Data mining with the Apriori algorithm revealed correlations between diseases, occupations, and environmental factors, informing public health policies. Communication between kiosks and servers used LoRa, 5G, and IEEE802.11, which are selected based on the distance and signal availability. The results support telemedicine kiosks as effective for NCD management, significantly improving key NCD indicators, average blood glucose, and blood pressure. [ABSTRACT FROM AUTHOR]

Huge waves caused by typhoons often induce severe disasters along coastal areas, making the effective prediction of typhoon-induced waves a crucial research issue for researchers. In recent years, the development of the Internet of Underwater Things (IoUT) has rapidly increased the prediction of oceanic environmental disasters. Past studies have utilized meteorological data and feedforward neural networks (e.g., BPNN) with static network structures to establish short lead time (e.g., 1 h) typhoon wave prediction models for the coast of Taiwan. However, sufficient lead time for prediction remains essential for preparedness, early warning, and response to minimize the loss of lives and properties during typhoons. The aim of this research is to construct a novel long lead time typhoon-induced wave prediction model using Long Short-Term Memory (LSTM), which incorporates a dynamic network structure. LSTM can capture long-term information through its recurrent structure and selectively retain necessary signals using memory gates. Compared to earlier studies, this method extends the prediction lead time and significantly improves the learning and generalization capability, thereby enhancing prediction accuracy markedly. [ABSTRACT FROM AUTHOR]

With the escalation in the size and complexity of modern Denial of Service attacks, there is a need for research in the context of Machine Learning (ML) used in attack execution and defense against such attacks. This paper investigates the potential use of ML in generating behavioral telemetry data using Long Short-Term Memory network and spoofing requests for the analyzed traffic to look legitimate. For this research, a custom testing environment was built that listens for mouse and keyboard events and analyzes them accordingly. While the economic feasibility of this attack currently limits its immediate threat, advancements in technology could make it more cost-effective for attackers in the future. Therefore, proactive development of countermeasures remains essential to mitigate potential risks and stay ahead of evolving attack methods. [ABSTRACT FROM AUTHOR]

Deep learning models have significantly improved object detection, which is essential for visual sensing. However, their increasing complexity results in higher latency and resource consumption, making real-time object detection challenging. In order to address the challenge, we propose a new lightweight filtering method called L-filter to predict empty video frames that include no object of interest (e.g., vehicles) with high accuracy via hybrid time series analysis. L-filter drops those frames deemed empty and conducts object detection for nonempty frames only, significantly enhancing the frame processing rate and scalability of real-time object detection. Our evaluation demonstrates that L-filter improves the frame processing rate by 31–47% for a single traffic video stream compared to three standalone state-of-the-art object detection models without L-filter. Additionally, L-filter significantly enhances scalability; it can process up to six concurrent video streams in one commodity GPU, supporting over 57 fps per stream, by working alongside the fastest object detection model among the three models. [ABSTRACT FROM AUTHOR]

This study aims to enhance diagnostic capabilities for optimising the performance of the anaerobic sewage treatment lagoon at Melbourne Water's Western Treatment Plant (WTP) through a novel machine learning (ML)-based monitoring strategy. This strategy employs ML to make accurate probabilistic predictions of biogas performance by leveraging diverse real-life operational and inspection sensor and other measurement data for asset management, decision making, and structural health monitoring (SHM). The paper commences with data analysis and preprocessing of complex irregular datasets to facilitate efficient learning in an artificial neural network. Subsequently, a Bayesian mixture density neural network model incorporating an attention-based mechanism in bidirectional long short-term memory (BiLSTM) was developed. This probabilistic approach uses a distribution output layer based on the Gaussian mixture model and Monte Carlo (MC) dropout technique in estimating data and model uncertainties, respectively. Furthermore, systematic hyperparameter optimisation revealed that the optimised model achieved a negative log-likelihood (NLL) of 0.074, significantly outperforming other configurations. It achieved an accuracy approximately 9 times greater than the average model performance (NLL = 0.753) and 22 times greater than the worst performing model (NLL = 1.677). Key factors influencing the model's accuracy, such as the input window size and the number of hidden units in the BiLSTM layer, were identified, while the number of neurons in the fully connected layer was found to have no significant impact on accuracy. Moreover, model calibration using the expected calibration error was performed to correct the model's predictive uncertainty. The findings suggest that the inherent data significantly contribute to the overall uncertainty of the model, highlighting the need for more high-quality data to enhance learning. This study lays the groundwork for applying ML in transforming high-value assets into intelligent structures and has broader implications for ML in asset management, SHM applications, and renewable energy sectors. [ABSTRACT FROM AUTHOR]

As an indispensable component of coal-fired power plants, boilers play a crucial role in converting water into high-pressure steam. The oxygen content in the flue gas is a crucial indicator, which indicates the state of combustion within the boiler. The oxygen content not only affects the thermal efficiency of the boiler and the energy utilization of the generator unit, but also has adverse impacts on the environment. Therefore, accurate measurement of the flue gas's oxygen content is of paramount importance in enhancing the energy utilization efficiency of coal-fired power plants and reducing the emissions of waste gas and pollutants. This study proposes a prediction model for the oxygen content in the flue gas that combines the whale optimization algorithm (WOA) and long short-term memory (LSTM) networks. Among them, the whale optimization algorithm (WOA) was used to optimize the learning rate, the number of hidden layers, and the regularization coefficients of the long short-term memory (LSTM). The data used in this study were obtained from a 350 MW power generation unit in a coal-fired power plant to validate the practicality and effectiveness of the proposed hybrid model. The simulation results demonstrated that the whale optimization algorithm–long short-term memory (WOA-LSTM) model achieved an MAE of 0.16493, an RMSE of 0.12712, an MAPE of 2.2254%, and an R 2 value of 0.98664. The whale optimization algorithm–long short-term memory (WOA-LSTM) model demonstrated enhancements in accuracy compared with the least squares support vector machine (LSSVM), long short-term memory (LSTM), particle swarm optimization–least squares support vector machine (PSO-LSSVM), and particle swarm optimization–long short-term memory (PSO-LSTM), with improvements of 4.93%, 4.03%, 1.35%, and 0.49%, respectively. These results indicated that the proposed soft sensor model exhibited more accurate performance, which can meet practical requirements of coal-fired power plants. [ABSTRACT FROM AUTHOR]

The critical challenge of estimating the Remaining Useful Life (RUL) of MoSi2 heating elements utilized in pusher kiln processes is to enhance operational efficiency and minimize downtime in industrial applications. MoSi2 heating elements are integral components in high-temperature environments, playing a pivotal role in achieving optimal thermal performance. However, prolonged exposure to extreme conditions leads to degradation, necessitating precise RUL predictions for proactive maintenance strategies. Since insufficient failure experience deals with Predictive Maintenance (PdM) in real-life scenarios, a Generative Adversarial Network (GAN) generates specific training data as failure experiences. The Remaining Useful Life (RUL) is the duration of the equipment's operation before repair or replacement, often measured in days, miles, or cycles. Machine learning models are trained using historical data encompassing various operational scenarios and degradation patterns. The RUL prediction model is determined through training, hyperparameter tuning, and comparisons based on the machine-learning model, such as Long Short-Term Memory (LSTM) or Support Vector Regression (SVR). As a result, SVR reflects the actual resistance variation, achieving the R-Square (R2) of 0.634, better than LSTM. From a safety perspective, SVR offers high prediction accuracy and sufficient time to schedule maintenance plans. [ABSTRACT FROM AUTHOR]

Technological progress has led to significant advancements in Earth observation and satellite systems. However, some services associated with remote sensing face issues related to timeliness and relevance, which affect the application of remote sensing resources in various fields and disciplines. The challenge now is to help end-users make precise decisions and recommendations for relevant resources that meet the demands of their specific domains from the vast array of remote sensing resources available. In this study, we propose a remote sensing resource service recommendation model that incorporates a time-aware dual LSTM neural network with similarity graph learning. We further use the stream push technology to enhance the model. We first construct interaction history behavior sequences based on users' resource search history. Then, we establish a category similarity relationship graph structure based on the cosine similarity matrix between remote sensing resource categories. Next, we use LSTM to represent historical sequences and Graph Convolutional Networks (GCN) to represent graph structures. We construct similarity relationship sequences by combining historical sequences to explore exact similarity relationships using LSTM. We embed user IDs to model users' unique characteristics. By implementing three modeling approaches, we can achieve precise recommendations for remote sensing services. Finally, we conduct experiments to evaluate our methods using three datasets, and the experimental results show that our method outperforms the state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

In ubiquitous healthcare systems, energy expenditure estimation based on wearable sensors such as inertial measurement units (IMUs) is important for monitoring the intensity of physical activity. Although several studies have reported data-driven methods to estimate energy expenditure during activities of daily living using wearable sensor signals, few have evaluated the performance while walking at various speeds and inclines. In this study, we present a hybrid model comprising a convolutional neural network (CNN) and long short-term memory (LSTM) to estimate the steady-state energy expenditure under various walking conditions based solely on IMU data. To implement and evaluate the model, we performed level/inclined walking and level running experiments on a treadmill. With regard to the model inputs, the performance of the proposed model based on fixed-size sequential data was compared with that of a method based on stride-segmented data under different conditions in terms of the sensor location, input sequence format, and neural network model. Based on the experimental results, the following conclusions were drawn: (i) the CNN–LSTM model using a two-second sequence from the IMU attached to the lower body yielded optimal performance, and (ii) although the stride-segmented data-based method showed superior performance, the performance difference between the two methods was not significant; therefore, the proposed model based on fixed-size sequential data may be considered more practical as it does not require heel-strike detection. [ABSTRACT FROM AUTHOR]

The energy consumption of a building is significantly influenced by the habits of its occupants. These habits not only pertain to occupancy states, such as presence or absence, but also extend to more detailed aspects of occupant behavior. To accurately capture this information, it is essential to use tools that can monitor occupant habits without altering them. Invasive methods such as body sensors or cameras could potentially disrupt the natural habits of the occupants. In our study, we primarily focus on occupancy states as a representation of occupant habits. We have created a model based on artificial neural networks (ANNs) to ascertain the occupancy state of a building using environmental data such as CO2 concentration and noise level. These data are collected through non-intrusive sensors. Our approach involves rule-based a priori labeling and the use of a long short-term memory (LSTM) network for predictive purposes. The model is designed to predict four distinct states in a residential building. Although we lack data on actual occupancy states, the model has shown promising results with an overall prediction accuracy ranging between 78% and 92%. [ABSTRACT FROM AUTHOR]

The accurate prediction of joint torque is required in various applications. Some traditional methods, such as the inverse dynamics model and the electromyography (EMG)-driven neuromusculoskeletal (NMS) model, depend on ground reaction force (GRF) measurements and involve complex optimization solution processes, respectively. Recently, machine learning methods have been popularly used to predict joint torque with surface electromyography (sEMG) signals and kinematic information as inputs. This study aims to predict lower limb joint torque in the sagittal plane during walking, using a long short-term memory (LSTM) model and Gaussian process regression (GPR) model, respectively, with seven characteristics extracted from the sEMG signals of five muscles and three joint angles as inputs. The majority of the normalized root mean squared error (NRMSE) values in both models are below 15 % , most Pearson correlation coefficient (R) values exceed 0.85 , and most decisive factor ( R 2 ) values surpass 0.75 . These results indicate that the joint prediction of torque is feasible using machine learning methods with sEMG signals and joint angles as inputs. [ABSTRACT FROM AUTHOR]

Mobile traffic prediction enables the efficient utilization of network resources and enhances user experience. In this paper, we propose a state transition graph-based spatial–temporal attention network (STG-STAN) for cell-level mobile traffic prediction, which is designed to exploit the underlying spatial–temporal dynamic information hidden in the historical mobile traffic data. Specifically, we first identify the semantic context information over different segments of the historical data by constructing the state transition graphs, which may reveal different patterns of random fluctuation. Then, based on the state transition graphs, a spatial attention extraction module using graph convolutional networks (GCNs) is designed to aggregate the spatial information of different nodes in the state transition graph. Moreover, a temporal extraction module is employed to capture the dynamic evolution and temporal correlation of the state transition graphs over time. Such a spatial–temporal attention network can be further integrated with a parallel long short-term memory (LSTM) module to improve the accuracy of mobile traffic prediction. Extensive experiments demonstrate that the STG-STAN can better exploit the spatial–temporal information hidden in the state transition graphs, achieving superior performance compared with several baselines. [ABSTRACT FROM AUTHOR]

In wireless communication, multiple signals are utilized to receive and send information in the form of signals simultaneously. These signals consume little power and are usually inexpensive, with a high data rate during data transmission. An Multi Input Multi Output (MIMO) system uses numerous antennas to enhance the functionality of the system. Moreover, system intricacy and power utilization are difficult and highly complicated tasks to achieve in an Analog to Digital Converter (ADC) at the receiver side. An infinite number of MIMO channels are used in wireless networks to improve efficiency with Cross Entropy Optimization (CEO). ADC is a serious issue because the data of the accepted signal are completely lost. ADC is used in the MIMO channels to overcome the above issues, but it is very hard to implement and design. So, an efficient way to enhance the estimation of channels in the MIMO system is proposed in this paper with the utilization of the heuristic-based optimization technique. The main task of the implemented channel prediction framework is to predict the channel coefficient of the MIMO system at the transmitter side based on the receiver side error ratio, which is obtained from feedback information using a Hybrid Serial Cascaded Network (HSCN). Then, this multi-scaled cascaded autoencoder is combined with Long Short Term Memory (LSTM) with an attention mechanism. The parameters in the developed Hybrid Serial Cascaded Multi-scale Autoencoder and Attention LSTM are optimized using the developed Hybrid Revised Position-based Wild Horse and Energy Valley Optimizer (RP-WHEVO) algorithm for minimizing the "Root Mean Square Error (RMSE), Bit Error Rate (BER) and Mean Square Error (MSE)" of the estimated channel. Various experiments were carried out to analyze the accomplishment of the developed MIMO model. It was visible from the tests that the developed model enhanced the convergence rate and prediction performance along with a reduction in the computational costs. [ABSTRACT FROM AUTHOR]

In general, judging the use/idle state of the wireless spectrum is the foundation for cognitive radio users (secondary users, SUs) to access limited spectrum resources efficiently. Rich information can be mined by the inherent correlation of electromagnetic spectrum data from SUs in time, frequency, space, and other dimensions. Therefore, how to efficiently use the spectrum status of each SU implementation of reception multidimensional combination forecasting is the core of this paper. In this paper, we propose a deep-learning hybrid model called TensorGCN-LSTM based on the tensor data structure. The model treats SUs deployed at different spatial locations under the same frequency, and the spectrum status of SUs themselves under different frequencies in the task area as nodes and constructs two types of graph structures. Graph convolutional operations are used to sequentially extract corresponding spatial-domain and frequency-domain features from the two types of graph structures. Then, the long short-term memory (LSTM) model is used to fuse the spatial, frequency, and temporal features of the cognitive radio environment data. Finally, the prediction task of the spectrum distribution situation is accomplished through fully connected layers. Specifically, the model constructs a tensor graph based on the spatial similarity of SUs' locations and the frequency correlation between different frequency signals received by SUs, which describes the electromagnetic wave's dependency relationship in spatial and frequency domains. LSTM is used to capture the electromagnetic wave's dependency relationship in the temporal domain. To evaluate the effectiveness of the model, we conducted ablation experiments on LSTM, GCN, GC-LSTM, and TensorGCN-LSTM models using simulated data. The experimental results showed that our model achieves better prediction performance in RMSE, and the correlation coefficient R2 of 0.8753 also confirms the feasibility of the model. [ABSTRACT FROM AUTHOR]

Vehicle malfunctions have a direct impact on both human and road safety, making vehicle network security an important and critical challenge. Vehicular ad hoc networks (VANETs) have grown to be indispensable in recent years for enabling intelligent transport systems, guaranteeing traffic safety, and averting collisions. However, because of numerous types of assaults, such as Distributed Denial of Service (DDoS) and Denial of Service (DoS), VANETs have significant difficulties. A powerful Network Intrusion Detection System (NIDS) powered by Artificial Intelligence (AI) is required to overcome these security issues. This research presents an innovative method for creating an AI-based NIDS that uses Deep Learning methods. The suggested model specifically incorporates the Self Attention-Based Bidirectional Long Short-Term Memory (SA-BiLSTM) for classification and the Cascaded Convolution Neural Network (CCNN) for learning high-level features. The Multi-variant Gradient-Based Optimization algorithm (MV-GBO) is applied to improve CCNN and SA-BiLSTM further to enhance the model's performance. Additionally, information gained using MV-GBO-based feature extraction is employed to enhance feature learning. The effectiveness of the proposed model is evaluated on reliable datasets such as KDD-CUP99, ToN-IoT, and VeReMi, which are utilized on the MATLAB platform. The proposed model achieved 99% accuracy on all the datasets. [ABSTRACT FROM AUTHOR]

The extreme operating environment of the combined heat and power (CHP) engine is likely to cause anomalies and defects, which can lead to engine failure; thus, detecting engine anomalies is essential. In this study, we propose a parallel convolutional neural network–long short-term memory (CNN-LSTM) residual blocks attention (PCLRA) anomaly detection model with engine sensor data. To our knowledge, this is the first time that parallel CNN-LSTM-based networks have been used in the field of CHP engine anomaly detection. In PCLRA, spatiotemporal features are extracted via CNN-LSTM in parallel and the information loss is compensated using the residual blocks and attention mechanism. The performance of PCLRA is compared with various hybrid models for 15 cases. First, the performances of serial and parallel models are compared. In addition, we evaluated the contributions of the residual blocks and attention mechanism to the performance of the CNN–LSTM hybrid model. The results indicate that PCLRA achieves the best performance, with a macro f1 score (mean ± standard deviation) of 0.951 ± 0.033, an anomaly f1 score of 0.903 ± 0.064, and an accuracy of 0.999 ± 0.002. We expect that the energy efficiency and safety of CHP engines can be improved by applying the PCLRA anomaly detection model. [ABSTRACT FROM AUTHOR]

As the use of construction robots continues to increase, ensuring safety and productivity while working alongside human workers becomes crucial. To prevent collisions, robots must recognize human behavior in close proximity. However, single, or RGB-depth cameras have limitations, such as detection failure, sensor malfunction, occlusions, unconstrained lighting, and motion blur. Therefore, this study proposes a multiple-camera approach for human activity recognition during human–robot collaborative activities in construction. The proposed approach employs a particle filter, to estimate the 3D human pose by fusing 2D joint locations extracted from multiple cameras and applies long short-term memory network (LSTM) to recognize ten activities associated with human and robot collaboration tasks in construction. The study compared the performance of human activity recognition models using one, two, three, and four cameras. Results showed that using multiple cameras enhances recognition performance, providing a more accurate and reliable means of identifying and differentiating between various activities. The results of this study are expected to contribute to the advancement of human activity recognition and utilization in human–robot collaboration in construction. [ABSTRACT FROM AUTHOR]

Light curves are plots of brightness measured over time. In the field of Space Situational Awareness (SSA), light curves of Resident Space Objects (RSOs) can be utilized to infer information about an RSO such as the type of object, its attitude, and its shape. Light curves of RSOs in geostationary orbit (GEO) have been a main research focus for many years due to the availability of long time series data spanning hours. Given that a large portion of RSOs are in low Earth orbit (LEO), it is of great importance to study trends in LEO light curves as well. The challenge with LEO light curves is that they tend to be short, typically no longer than a few minutes, which makes them difficult to analyze with typical time series techniques. This study presents a novel approach to observational LEO light curve classification. We extract features from light curves using a wavelet scattering transformation which is used as an input for a machine learning classifier. We performed light curve classification using both a conventional machine learning approach, namely a support vector machine (SVM), and a deep learning technique, long short-term memory (LSTM), to compare the results. LSTM outperforms SVM for LEO light curve classification with a 92% accuracy. This proves the viability of RSO classification by object type and spin rate from real LEO light curves. [ABSTRACT FROM AUTHOR]

The quality of rolling bearings is vital for the working state and rotation accuracy of the shaft. Timely and accurately acquiring bearing status and early fault diagnosis can effectively prevent losses, making it highly practical. To improve the accuracy of bearing fault diagnosis, this paper proposes a CNN-LSTM bearing fault diagnosis model optimized by hybrid particle swarm optimization (HPSO). The HPSO algorithm has a strong global optimization ability and can effectively solve nonlinear and multivariate optimization problems. It is used to optimize and match the parameters of the CNN-LSTM model and dynamically find the optimal value of the parameters. This model overcomes the problem that the parameters of the CNN-LSTM model depend on empirical settings and cannot be adjusted dynamically. This model is used for bearing fault diagnosis, and the accuracy rate of fault diagnosis classification reaches 99.2%. Compared with the traditional CNN, LSTM, and CNN-LSTM models, the accuracy rates are increased by 6.6%, 9.2%, and 5%, respectively. At the same time, comparing the models with different optimization parameters shows that the model proposed in this paper has the highest accuracy. The experimental results verified the superiority of the HPSO algorithm to optimize model parameters and the feasibility and accuracy of the HPSO-CNN-LSTM model for bearing fault diagnosis. [ABSTRACT FROM AUTHOR]

The advent of Industry 4.0 introduced new ways for businesses to evolve by implementing maintenance policies leading to advancements in terms of productivity, efficiency, and financial performance. In line with the growing emphasis on sustainability, industries implement predictive techniques based on Artificial Intelligence for the purpose of mitigating machine and equipment failures by predicting anomalies during their production process. In this work, a new dataset that was made publicly available, collected from an industrial blower, is presented, analyzed and modeled using a Sequence-to-Sequence Stacked Sparse Long Short-Term Memory Autoencoder. Specifically the right and left mounted ball bearing units were measured during several months of normal operational condition as well as during an encumbered operational state. An anomaly detection model was developed for the purpose of analyzing the operational behavior of the two bearing units. A stacked sparse Long Short-Term Memory Autoencoder was successfully trained on the data obtained from the left unit under normal operating conditions, learning the underlying patterns and statistical connections of the data. The model was evaluated by means of the Mean Squared Error using data from the unit's encumbered state, as well as using data collected from the right unit. The model performed satisfactorily throughout its evaluation on all collected datasets. Also, the model proved its capability for generalization along with adaptability on assessing the behavior of equipment similar to the one it was trained on. [ABSTRACT FROM AUTHOR]

This article is devoted to solving the problem of converting sign language into a consistent text with intonation markup for subsequent voice synthesis of sign phrases by speech with intonation. The paper proposes an improved method of continuous recognition of sign language, the results of which are transmitted to a natural language processor based on analyzers of morphology, syntax, and semantics of the Kazakh language, including morphological inflection and the construction of an intonation model of simple sentences. This approach has significant practical and social significance, as it can lead to the development of technologies that will help people with disabilities to communicate and improve their quality of life. As a result of the cross-validation of the model, we obtained an average test accuracy of 0.97 and an average val_accuracy of 0.90 for model evaluation. We also identified 20 sentence structures of the Kazakh language with their intonational model. [ABSTRACT FROM AUTHOR]

This paper proposes a method for accurate 3D posture sensing of the soft actuators, which could be applied to the closed-loop control of soft robots. To achieve this, the method employs an array of miniaturized sponge resistive materials along the soft actuator, which uses long short-term memory (LSTM) neural networks to solve the end-to-end 3D posture for the soft actuators. The method takes into account the hysteresis of the soft robot and non-linear sensing signals from the flexible bending sensors. The proposed approach uses a flexible bending sensor made from a thin layer of conductive sponge material designed for posture sensing. The LSTM network is used to model the posture of the soft actuator. The effectiveness of the method has been demonstrated on a finger-size 3 degree of freedom (DOF) pneumatic bellow-shaped actuator, with nine flexible sponge resistive sensors placed on the soft actuator's outer surface. The sensor-characterizing results show that the maximum bending torque of the sensor installed on the actuator is 4.7 Nm, which has an insignificant impact on the actuator motion based on the working space test of the actuator. Moreover, the sensors exhibit a relatively low error rate in predicting the actuator tip position, with error percentages of 0.37%, 2.38%, and 1.58% along the x-, y-, and z-axes, respectively. This work is expected to contribute to the advancement of soft robot dynamic posture perception by using thin sponge sensors and LSTM or other machine learning methods for control. [ABSTRACT FROM AUTHOR]

Predictive maintenance in the car industry is an active field of research for machine learning and anomaly detection. The capability of cars to produce time series data from sensors is growing as the car industry is heading towards more connected and electric vehicles. Unsupervised anomaly detectors are therefore very adapted to process those complex multidimensional time series and highlight abnormal behaviors. We propose to use recurrent and convolutional neural networks based on unsupervised anomaly detectors with simple architectures on real, multidimensional time series generated by the car sensors and extracted from the Controller Area Network bus (CAN). Our method is then evaluated through known specific anomalies. As the computational costs of Machine Learning algorithms are a rising issue regarding embedded scenarios such as car anomaly detection, we also focus on creating anomaly detectors that are as small as possible. Using a state-of-the-art methodology incorporating a time series predictor and a prediction-error-based anomaly detector, we show that we can obtain roughly the same anomaly detection performance with smaller predictors, reducing parameters and calculations by up to 23 % and 60 % , respectively. Finally, we introduce a method to correlate variables with specific anomalies by using anomaly detector results and labels. [ABSTRACT FROM AUTHOR]

As the number of Internet of things (IoT) devices increases exponentially, scheduling and managing the radio resources for IoT devices has become more important. To efficiently allocate radio resources, the base station (BS) needs the channel state information (CSI) of devices every time. Hence, each device needs to periodically (or aperiodically) report its channel quality indicator (CQI) to the BS. The BS determines the modulation and coding scheme (MCS) based on the CQI reported by the IoT device. However, the more a device reports its CQI, the more the feedback overhead increases. In this paper, we propose a long short-term memory (LSTM)-based CQI feedback scheme, where the IoT device aperiodically reports its CQI relying on an LSTM-based channel prediction. Additionally, because the memory capacity of IoT devices is generally small, the complexity of the machine learning model must be reduced. Hence, we propose a lightweight LSTM model to reduce the complexity. The simulation results show that the proposed lightweight LSTM-based CSI scheme dramatically reduces the feedback overhead compared with that of the existing periodic feedback scheme. Moreover, the proposed lightweight LSTM model significantly reduces the complexity without sacrificing performance. [ABSTRACT FROM AUTHOR]

Aiming at the problem of fast divergence of pure inertial navigation system without correction under the condition of GNSS restricted environment, this paper proposes a multi-mode navigation method with an intelligent virtual sensor based on long short-term memory (LSTM). The training mode, predicting mode, and validation mode for the intelligent virtual sensor are designed. The modes are switching flexibly according to GNSS rejecting situation and the status of the LSTM network of the intelligent virtual sensor. Then the inertial navigation system (INS) is corrected, and the availability of the LSTM network is also maintained. Meanwhile, the fireworks algorithm is adopted to optimize the learning rate and the number of hidden layers of LSTM hyperparameters to improve the estimation performance. The simulation results show that the proposed method can maintain the prediction accuracy of the intelligent virtual sensor online and shorten the training time according to the performance requirements adaptively. Under small sample conditions, the training efficiency and availability ratio of the proposed intelligent virtual sensor are improved significantly more than the neural network (BP) as well as the conventional LSTM network, improving the navigation performance in GNSS restricted environment effectively and efficiently. [ABSTRACT FROM AUTHOR]

In this work, we developed a prototype that adopted sound-based systems for localization of visually impaired individuals. The system was implemented based on a wireless ultrasound network, which helped the blind and visually impaired to navigate and maneuver autonomously. Ultrasonic-based systems use high-frequency sound waves to detect obstacles in the environment and provide location information to the user. Voice recognition and long short-term memory (LSTM) techniques were used to design the algorithms. The Dijkstra algorithm was also used to determine the shortest distance between two places. Assistive hardware tools, which included an ultrasonic sensor network, a global positioning system (GPS), and a digital compass, were utilized to implement this method. For indoor evaluation, three nodes were localized on the doors of different rooms inside the house, including the kitchen, bathroom, and bedroom. The coordinates (interactive latitude and longitude points) of four outdoor areas (mosque, laundry, supermarket, and home) were identified and stored in a microcomputer's memory to evaluate the outdoor settings. The results showed that the root mean square error for indoor settings after 45 trials is about 0.192. In addition, the Dijkstra algorithm determined that the shortest distance between two places was within an accuracy of 97%. [ABSTRACT FROM AUTHOR]

When an electric wheelchair is operated using gaze motion, eye movements such as checking the environment and observing objects are also incorrectly recognized as input operations. This phenomenon is called the "Midas touch problem", and classifying visual intentions is extremely important. In this paper, we develop a deep learning model that estimates the user's visual intention in real time and an electric wheelchair control system that combines intention estimation and the gaze dwell time method. The proposed model consists of a 1DCNN-LSTM that estimates visual intention from feature vectors of 10 variables, such as eye movement, head movement, and distance to the fixation point. The evaluation experiments classifying four types of visual intentions show that the proposed model has the highest accuracy compared to other models. In addition, the results of the driving experiments of the electric wheelchair implementing the proposed model show that the user's efforts to operate the wheelchair are reduced and that the operability of the wheelchair is improved compared to the traditional method. From these results, we concluded that visual intentions could be more accurately estimated by learning time series patterns from eye and head movement data. [ABSTRACT FROM AUTHOR]

The integration of the cloud and Internet of Things (IoT) technology has resulted in a significant rise in futuristic technology that ensures the long-term development of IoT applications, such as intelligent transportation, smart cities, smart healthcare, and other applications. The explosive growth of these technologies has contributed to a significant rise in threats with catastrophic and severe consequences. These consequences affect IoT adoption for both users and industry owners. Trust-based attacks are the primary selected weapon for malicious purposes in the IoT context, either through leveraging established vulnerabilities to act as trusted devices or by utilizing specific features of emerging technologies (i.e., heterogeneity, dynamic nature, and a large number of linked objects). Consequently, developing more efficient trust management techniques for IoT services has become urgent in this community. Trust management is regarded as a viable solution for IoT trust problems. Such a solution has been used in the last few years to improve security, aid decision-making processes, detect suspicious behavior, isolate suspicious objects, and redirect functionality to trusted zones. However, these solutions remain ineffective when dealing with large amounts of data and constantly changing behaviors. As a result, this paper proposes a dynamic trust-related attack detection model for IoT devices and services based on the deep long short-term memory (LSTM) technique. The proposed model aims to identify the untrusted entities in IoT services and isolate untrusted devices. The effectiveness of the proposed model is evaluated using different data samples with different sizes. The experimental results showed that the proposed model obtained a 99.87% and 99.76% accuracy and F-measure, respectively, in the normal situation, without considering trust-related attacks. Furthermore, the model effectively detected trust-related attacks, achieving a 99.28% and 99.28% accuracy and F-measure, respectively. [ABSTRACT FROM AUTHOR]

This study presents a novel approach to cope with the human behaviour uncertainty during Human-Robot Collaboration (HRC) in dynamic and unstructured environments, such as agriculture, forestry, and construction. These challenging tasks, which often require excessive time, labour and are hazardous for humans, provide ample room for improvement through collaboration with robots. However, the integration of humans in-the-loop raises open challenges due to the uncertainty that comes with the ambiguous nature of human behaviour. Such uncertainty makes it difficult to represent high-level human behaviour based on low-level sensory input data. The proposed Fuzzy State-Long Short-Term Memory (FS-LSTM) approach addresses this challenge by fuzzifying ambiguous sensory data and developing a combined activity recognition and sequence modelling system using state machines and the LSTM deep learning method. The evaluation process compares the traditional LSTM approach with raw sensory data inputs, a Fuzzy-LSTM approach with fuzzified inputs, and the proposed FS-LSTM approach. The results show that the use of fuzzified inputs significantly improves accuracy compared to traditional LSTM, and, while the fuzzy state machine approach provides similar results than the fuzzy one, it offers the added benefits of ensuring feasible transitions between activities with improved computational efficiency. [ABSTRACT FROM AUTHOR]

To prevent and diagnose hypertension early, there has been a growing demand to identify its states that align with patients. This pilot study aims to research how a non-invasive method using photoplethysmographic (PPG) signals works together with deep learning algorithms. A portable PPG acquisition device (Max30101 photonic sensor) was utilized to (1) capture PPG signals and (2) wirelessly transmit data sets. In contrast to traditional feature engineering machine learning classification schemes, this study preprocessed raw data and applied a deep learning algorithm (LSTM-Attention) directly to extract deeper correlations between these raw datasets. The Long Short-Term Memory (LSTM) model underlying a gate mechanism and memory unit enables it to handle long sequence data more effectively, avoiding gradient disappearance and possessing the ability to solve long-term dependencies. To enhance the correlation between distant sampling points, an attention mechanism was introduced to capture more data change features than a separate LSTM model. A protocol with 15 healthy volunteers and 15 hypertension patients was implemented to obtain these datasets. The processed result demonstrates that the proposed model could present satisfactory performance (accuracy: 0.991; precision: 0.989; recall: 0.993; F1-score: 0.991). The model we proposed also demonstrated superior performance compared to related studies. The outcome indicates the proposed method could effectively diagnose and identify hypertension; thus, a paradigm to cost-effectively screen hypertension could rapidly be established using wearable smart devices. [ABSTRACT FROM AUTHOR]

Insulators installed outdoors are vulnerable to the accumulation of contaminants on their surface, which raise their conductivity and increase leakage current until a flashover occurs. To improve the reliability of the electrical power system, it is possible to evaluate the development of the fault in relation to the increase in leakage current and thus predict whether a shutdown may occur. This paper proposes the use of empirical wavelet transform (EWT) to reduce the influence of non-representative variations and combines the attention mechanism with a long short-term memory (LSTM) recurrent network for prediction. The Optuna framework has been applied for hyperparameter optimization, resulting in a method called optimized EWT-Seq2Seq-LSTM with attention. The proposed model had a 10.17% lower mean square error (MSE) than the standard LSTM and a 5.36% lower MSE than the model without optimization, showing that the attention mechanism and hyperparameter optimization is a promising strategy. [ABSTRACT FROM AUTHOR]

Recent technological advancements facilitate the autonomous navigation of maritime surface ships. The accurate data given by a range of various sensors serve as the primary assurance of a voyage's safety. Nevertheless, as sensors have various sample rates, they cannot obtain information at the same time. Fusion decreases the accuracy and reliability of perceptual data if different sensor sample rates are not taken into account. Hence, it is helpful to increase the quality of the fusion information to precisely anticipate the motion status of ships at the sampling time of each sensor. This paper proposes a non-equal time interval incremental prediction method. In this method, the high dimensionality of the estimated state and nonlinearity of the kinematic equation are taken into consideration. First, the cubature Kalman filter is employed to estimate a ship's motion at equal intervals based on the ship's kinematic equation. Next, a ship motion state predictor based on a long short-term memory network structure is created, using the increment and time interval of the historical estimation sequence as the network input and the increment of the motion state at the projected time as the network output. The suggested technique can lessen the effect of the speed difference between the test set and the training set on the prediction accuracy compared with the traditional long short-term memory prediction method. Finally, comparison experiments are carried out to validate the precision and effectiveness of the proposed approach. The experimental results show that the root-mean-square error coefficient of the prediction error is decreased on average by roughly 78% for various modes and speeds when compared with the conventional non-incremental long short-term memory prediction approach. Additionally, the proposed prediction technology and the traditional approach have virtually the same algorithm times, which may fulfill the real engineering requirements. [ABSTRACT FROM AUTHOR]

Adaptation of handover parameters in ultra-dense networks has always been one of the key issues in optimizing network performance. Aiming at the optimization goal of effective handover ratio, this paper proposes a deep Q-learning (DQN) method that dynamically selects handover parameters according to wireless signal fading conditions. This approach seeks good backward compatibility. In order to enhance the efficiency and performance of the DQN method, Long Short Term Memory (LSTM) is used to build a digital twin and assist the DQN algorithm to achieve a more efficient search. Simulation experiments prove that the enhanced method has a faster convergence speed than the ordinary DQN method, and at the same time, achieves an average effective handover ratio increase of 2.7%. Moreover, in different wireless signal fading intervals, the method proposed in this paper has achieved better performance. [ABSTRACT FROM AUTHOR]