Search Results (32)

This study presents a novel approach to passive human counting in indoor environments using Bluetooth Low Energy (BLE) signals and deep learning. The motivation behind this research is the need for non-intrusive, privacy-preserving occupancy monitoring in sensitive indoor settings, where traditional camera-based solutions may be unsuitable. Our method leverages the deformations that BLE signals undergo when interacting with the human body, enabling occupant detection and counting without requiring wearable devices or visual tracking. We evaluated three deep neural network models—Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), and a hybrid CNN+LSTM architecture—under both classification and regression settings. Experimental results indicate that the hybrid CNN+LSTM model outperforms the others in terms of accuracy and mean absolute error. Notably, in the regression setup, the model can generalize to occupancy values not present in the fine-tuning dataset, requiring only a few minutes of calibration data to adapt to a new environment. We believe that this approach offers a valuable solution for real-time people counting in critical environments such as laboratories, clinics, or hospitals, where preserving privacy may limit the use of camera-based systems. Overall, our method demonstrates high adaptability and robustness, making it suitable for practical deployment in diverse indoor scenarios. Full article

Private, non-intrusive presence detection methods contribute to various applications, from occupancy monitoring to energy optimization and security. This study presents a deep learning approach for predicting occupancy patterns using CO₂ sensor data and temporal features, derived from a year-long dataset (18 September 2023–21 November 2024) collected via the Smart Indoor Air Quality Monitor. We created a dataset of 19,189 samples of CO₂ levels (0–5000 ppm) with timestamps. A sequential neural network with three fully connected layers was implemented in TensorFlow. The developed model demonstrated the feasibility of predicting occupancy based on CO₂ data and temporal features with an accuracy of 0.97 and an F1-score of 0.92. Model visualization was performed using heatmaps. Its advantages include low computational requirements, cost-effective sensors, an IoT-enabled interface, and scalability. However, the study is limited to a university laboratory with a capacity of 1–16 occupants, which may impact its generalizability to other settings. These findings highlight the utility of CO₂ levels and temporal features for occupancy estimation in laboratory conditions and contribute a unique, long-term multimodal dataset to the research community. Full article

In elderly monitoring or indoor intrusion detection, the recognition of human activity is a key task. Owing to several strengths of Wi-Fi-based devices, including their non-contact and privacy protection, these devices have been widely applied in the area of smart homes. By the deep learning technique, numerous Wi-Fi-based activity recognition methods can realize satisfied recognitions, however, these methods may fail to recognize the activities of an unknown person without the learning process. In this study, using channel state information (CSI) data, a novel cross-person activity recognition (CPAR) method is proposed by a deep learning approach with generalization capability. Combining one of the state-of-the-art deep neural networks (DNNs) used in activity recognition, i.e., attention-based bi-directional long short-term memory (ABLSTM), the snapshot ensemble is the first to be adopted to train several base-classifiers for enhancing the generalization and practicability of recognition. Second, to discriminate the extracted features, metric learning is further introduced by using the center loss, obtaining snapshot ensemble-used ABLSTM with center loss (SE-ABLSTM-C). In the experiments of CPAR, the proposed SE-ABLSTM-C method markedly improved the recognition accuracies to an application level, for seven categories of activities. Full article

Seawater intrusion significantly affects the microbial communities within coastal aquifers. Investigating the spatial distribution of groundwater microbial communities in coastal regions is crucial for understanding seawater intrusion. The primary objective of this study is to develop a novel microbial index-based method for detecting seawater intrusion. Groundwater microbial samples were collected and sent to the laboratory in the west coastal area of Longkou City, Shandong Province. By characterizing the microbial community within the whole interval of seawater intrusion into fresh groundwater and discussing the effects of salt-freshwater displacement intensities on groundwater microbial communities, including diversity, structure, and function, using indoor domestication experiments, we reveal the response of microorganisms to the seawater intrusion process under in situ environmental conditions. The results show that the microbial community diversity is highest in environments with a seawater mixing proportion (P_(sm)) of 2.5% and lowest in those with a P_(sm) of 75%. When considering species abundance and evolutionary processes, the microbial community structure is similar at higher P_(sm) levels, while it is similar at lower P_(sm) levels based on the presence or absence of species. Tenericutes, Flavobacteriia, Rhodobacterales, Flavobacteriales, Rhodobacteraceae, Flavobacteriaceae, Cohaesibacteraceae, and Cohaesibacter are significantly positively correlated with the P_(sm). Strong salt-freshwater displacement enhanced the richness and evenness of the microbial community, whereas weak displacement showed the opposite trend. Strong displacement affects the functional profiles of the microbial community. This study effectively addressed the challenge of obtaining samples in coastal areas and also incorporated salt-freshwater displacement intensities, which can more comprehensively describe the microbial community characteristics within the groundwater of coastal aquifers. Full article

This study explored an indoor system for tracking multiple humans and detecting falls, employing three Millimeter-Wave radars from Texas Instruments. Compared to wearables and camera methods, Millimeter-Wave radar is not plagued by mobility inconveniences, lighting conditions, or privacy issues. We conducted an initial evaluation of radar characteristics, covering aspects such as interference between radars and coverage area. Then, we established a real-time framework to integrate signals received from these radars, allowing us to track the position and body status of human targets non-intrusively. Additionally, we introduced innovative strategies, including dynamic Density-Based Spatial Clustering of Applications with Noise (DBSCAN) clustering based on signal SNR levels, a probability matrix for enhanced target tracking, target status prediction for fall detection, and a feedback loop for noise reduction. We conducted an extensive evaluation using over 300 min of data, which equated to approximately 360,000 frames. Our prototype system exhibited a remarkable performance, achieving a precision of 98.9% for tracking a single target and 96.5% and 94.0% for tracking two and three targets in human-tracking scenarios, respectively. Moreover, in the field of human fall detection, the system demonstrates a high accuracy rate of 96.3%, underscoring its effectiveness in distinguishing falls from other statuses. Full article

The construction sector is responsible for almost 30% of the world’s total energy consumption, with a significant portion of this energy being used by heating, ventilation and air-conditioning (HVAC) systems to ensure people’s thermal comfort. In practical applications, the conventional approach to HVAC management in buildings typically involves the manual control of temperature setpoints by facility operators. Nevertheless, the implementation of real-time alterations that are based on the thermal comfort levels of humans inside a building has the potential to dramatically improve the energy efficiency of the structure. Therefore, we propose a model for non-intrusive, dynamic inference of occupant thermal comfort based on building indoor surveillance camera data. It is based on a two-stream transformer-augmented adaptive graph convolutional network to identify people’s heat-related adaptive behaviors. The transformer specifically strengthens the original adaptive graph convolution network module, resulting in further improvement to the accuracy of the detection of thermal adaptation behavior. The experiment is conducted on a dataset including 16 distinct temperature adaption behaviors. The findings indicate that the suggested strategy significantly improves the behavior recognition accuracy of the proposed model to 96.56%. The proposed model provides the possibility to realize energy savings and emission reductions in intelligent buildings and dynamic decision making in energy management systems. Full article

In health monitoring systems for the elderly, a crucial aspect is unobtrusively and continuously monitoring their activities to detect potentially hazardous incidents such as sudden falls as soon as they occur. However, the effectiveness of current non-contact sensor-based activity detection systems is limited by obstacles present in the environment. To overcome this limitation, a straightforward yet highly efficient approach involves utilizing multiple sensors that collaborate seamlessly. This paper proposes a method that leverages 2D Light Detection and Ranging (Lidar) technology for activity detection. Multiple 2D Lidars are positioned in an indoor environment with varying obstacles such as furniture, working cohesively to create a comprehensive representation of ongoing activities. The data from these Lidars is concatenated and transformed into a more interpretable format, resembling images. A convolutional Long Short-Term Memory (LSTM) Neural Network is then used to process these generated images to classify the activities. The proposed approach achieves high accuracy in three tasks: activity detection, fall detection, and unsteady gait detection. Specifically, it attains accuracies of 96.10%, 99.13%, and 93.13% for these tasks, respectively. This demonstrates the efficacy and promise of the method in effectively monitoring and identifying potentially hazardous events for the elderly through 2D Lidars, which are non-intrusive sensing technology. Full article

Device-free indoor identification of people with high accuracy is the key to providing personalized services. Visual methods are the solution but they require a clear view and good lighting conditions. Additionally, the intrusive nature leads to privacy concerns. A robust identification and classification system using the mmWave radar and an improved density-based clustering algorithm along with LSTM are proposed in this paper. The system leverages mmWave radar technology to overcome challenges posed by varying environmental conditions on object detection and recognition. The point cloud data are processed using a refined density-based clustering algorithm to extract ground truth in a 3D space accurately. A bi-directional LSTM network is employed for individual user identification and intruder detection. The system achieved an overall identification accuracy of 93.9% and an intruder detection rate of 82.87% for groups of 10 individuals, demonstrating its effectiveness. Full article

Monitoring the presence and movements of individuals or crowds in a given area can provide valuable insight into actual behavior patterns and hidden trends. Therefore, it is crucial in areas such as public safety, transportation, urban planning, disaster and crisis management, and mass events organization, both for the adoption of appropriate policies and measures and for the development of advanced services and applications. In this paper, we propose a non-intrusive privacy-preserving detection of people’s presence and movement patterns by tracking their carried WiFi-enabled personal devices, using the network management messages transmitted by these devices for their association with the available networks. However, due to privacy regulations, various randomization schemes have been implemented in network management messages to prevent easy discrimination between devices based on their addresses, sequence numbers of messages, data fields, and the amount of data contained in the messages. To this end, we proposed a novel de-randomization method that detects individual devices by grouping similar network management messages and corresponding radio channel characteristics using a novel clustering and matching procedure. The proposed method was first calibrated using a labeled publicly available dataset, which was validated by measurements in a controlled rural and a semi-controlled indoor environment, and finally tested in terms of scalability and accuracy in an uncontrolled crowded urban environment. The results show that the proposed de-randomization method is able to correctly detect more than 96% of the devices from the rural and indoor datasets when validated separately for each device. When the devices are grouped, the accuracy of the method decreases but is still above 70% for rural environments and 80% for indoor environments. The final verification of the non-intrusive, low-cost solution for analyzing the presence and movement patterns of people, which also provides information on clustered data that can be used to analyze the movements of individuals, in an urban environment confirmed the accuracy, scalability and robustness of the method. However, it also revealed some drawbacks in terms of exponential computational complexity and determination and fine-tuning of method parameters, which require further optimization and automation. Full article

Point cloud data have become the primary spatial data source for the 3D reconstruction of building engineering, where 3D reconstructed building information models can improve construction efficiency. In such applications, detecting windows and doors is essential. Previous research mainly used red-green-blue (RGB) information or semantic features for detection, where the combination of these two features was not considered. Therefore, this research proposed a practical approach to detecting windows and doors using point cloud data with the combination of semantic features and material characteristics. The point cloud data are first segmented using Gradient Filtering and Random Sample Consensus (RANSAC) to obtain the 3D indoor data without intrusions and protrusions. As input, the 3D indoor data are projected to horizontal planes as 2D point cloud data. The 2D point cloud data are then transformed to 2D images, representing the indoor area for feature extraction. On the 2D images, the 2D boundary of each potential opening is extracted using an improved Bounding Box algorithm, and the extraction result is transformed back to 3D data. Based on the 3D data, the reflectivity of building material is applied to differentiate windows and doors from potential openings, and the number of data points is used to check the opening condition of windows and doors. The abovementioned approach was tested using the point cloud data representing one campus building, including two big rooms and one corridor. The experimental results showed that accurate detection of windows and doors was successfully reached. The completeness of the detection is 100%, and the correctness of the detection is 90.32%. The total time for the feature extraction is 22.8 s for processing 2 million point cloud data, including time from reading data of 10.319 s and time from showing the results of 4.938 s. Full article

The World Health Organization recognizes physical activity as an influencing domain on quality of life. Monitoring, evaluating, and supervising it by wearable devices can contribute to the early detection and progress assessment of diseases such as Alzheimer’s, rehabilitation, and exercises in telehealth, as well as abrupt events such as a fall. In this work, we use a non-invasive and non-intrusive flexible wearable device for 3D spine pose measurement to monitor and classify physical activity. We develop a comprehensive protocol that consists of 10 indoor, 4 outdoor, and 8 transition states activities in three categories of static, dynamic, and transition in order to evaluate the applicability of the flexible wearable device in human activity recognition. We implement and compare the performance of three neural networks: long short-term memory (LSTM), convolutional neural network (CNN), and a hybrid model (CNN-LSTM). For ground truth, we use an accelerometer and strips data. LSTM reached an overall classification accuracy of 98% for all activities. The CNN model with accelerometer data delivered better performance in lying down (100%), static (standing = 82%, sitting = 75%), and dynamic (walking = 100%, running = 100%) positions. Data fusion improved the outputs in standing (92%) and sitting (94%), while LSTM with the strips data yielded a better performance in bending-related activities (bending forward = 49%, bending backward = 88%, bending right = 92%, and bending left = 100%), the combination of data fusion and principle components analysis further strengthened the output (bending forward = 100%, bending backward = 89%, bending right = 100%, and bending left = 100%). Moreover, the LSTM model detected the first transition state that is similar to fall with the accuracy of 84%. The results show that the wearable device can be used in a daily routine for activity monitoring, recognition, and exercise supervision, but still needs further improvement for fall detection. Full article

Nowadays, unmanned aerial vehicles/drones are involved in a continuously growing number of security incidents. Therefore, the research interest in drone versus human movement detection and characterization is justified by the fact that such devices represent a potential threat for indoor/office intrusion, while normally, a human presence is allowed after passing several security points. Our paper comparatively characterizes the movement of a drone and a human in an indoor environment. The movement map was obtained using advanced signal processing methods such as wavelet transform and the phase diagram concept, and applied to the signal acquired from UWB sensors. Full article

In recent times, we have been witnessing the development of multiple applications and deployment of services through the indoors location of people as it allows the development of services of interest in areas related mainly to security, guiding people, or offering services depending on their localization. On the other hand, at present, the deployment of Wi-Fi networks is so advanced that a network can be found almost anywhere. In addition, security systems are more demanded and are implemented in many buildings. Thus, in order to provide a non intrusive presence detection system, in this manuscript, the development of a methodology is proposed which is able to detect human presence through the channel state information (CSI) of wireless communication networks based on the 802.11n standard. One of the main contributions of this standard is multiple-input multiple-output (MIMO) with orthogonal frequency division multiplexing (OFDM). This makes it possible to obtain channel state information for each subcarrier. In order to implement this methodology, an analysis and feature extraction in time-domain of CSI is carried out, and it is validated using different classification models trained through a series of samples that were captured in two different environments. The experiments show that the methodology presented in this manuscript obtains an average accuracy above 90%. Full article

Indoor occupancy prediction can play a vital role in the energy-efficient operation of building engineering systems and maintaining satisfactory indoor climate conditions at the lowest possible energy use by operating these systems on the basis of occupancy data. Many methods have been proposed to predict occupancy in residential buildings according to different data types, e.g., digital cameras, motion sensors, and indoor climate sensors. Among these proposed methods, those with indoor climate data as input have received significant interest due to their less intrusive and cost-effective approach. This paper proposes a deep learning method called CNN-XGBoost to predict occupancy using indoor climate data and compares the performance of the proposed method with a range of supervised and unsupervised machine learning algorithms plus artificial neural network algorithms. The comparison is performed using mean absolute error, confusion matrix, and F₁ score. Indoor climate data used in this work are CO₂, relative humidity, and temperature measured by sensors for 13 days in December 2021. We used inexpensive sensors in different rooms of a residential building with a balanced mechanical ventilation system located in northwest Copenhagen, Denmark. The proposed algorithm consists of two parts: a convolutional neural network that learns the features of the input data and a scalable end-to-end tree-boosting classifier. The result indicates that CNN-XGBoost outperforms other algorithms in predicting occupancy levels in all rooms of the test building. In this experiment, we achieved the highest accuracy in occupancy detection using inexpensive indoor climate sensors in a mechanically ventilated residential building with minimum privacy invasion. Full article

This work presents the design of a wireless acoustic sensor network (WASN) that monitors indoor spaces. The proposed network would enable the acquisition of valuable information on the behavior of the inhabitants of the space. This WASN has been conceived to work in any type of indoor environment, including houses, hospitals, universities or even libraries, where the tracking of people can give relevant insight, with a focus on ambient assisted living environments. The proposed WASN has several priorities and differences compared to the literature: (i) presenting a low-cost flexible sensor able to monitor wide indoor areas; (ii) balance between acoustic quality and microphone cost; and (iii) good communication between nodes to increase the connectivity coverage. A potential application of the proposed network could be the generation of a sound map of a certain location (house, university, offices, etc.) or, in the future, the acoustic detection of events, giving information about the behavior of the inhabitants of the place under study. Each node of the network comprises an omnidirectional microphone and a computation unit, which processes acoustic information locally following the edge-computing paradigm to avoid sending raw data to a cloud server, mainly for privacy and connectivity purposes. Moreover, this work explores the placement of acoustic sensors in a real scenario, following acoustic coverage criteria. The proposed network aims to encourage the use of real-time non-invasive devices to obtain behavioral and environmental information, in order to take decisions in real-time with the minimum intrusiveness in the location under study. Full article