1,720,989 research outputs found
SHPIA 2.0: An Easily Scalable, Low-Cost, Multi-purpose Smart Home Platform for Intelligent Applications
Full text linkSensors, electronic devices, and smart systems have invaded the market and our daily lives. As a result, their utility in smart home contexts to improve the quality of life, especially for the elderly and people with special needs, is getting stronger and stronger. Therefore, many systems based on smart applications and intelligent devices have been developed, for example, to monitor people’s environmental contexts, help in daily-life activities, and analyze their health status. However, most existing solutions have drawbacks related to accessibility and usability. They tend to be expensive and lack generality and interoperability. These solutions are not easily scalable and are typically designed for specific constrained scenarios. This paper tackles such drawbacks by presenting SHPIA 2.0, an easily scalable, low-cost, multi-purpose smart home platform for intelligent applications. It leverages low-cost Bluetooth Low Energy (BLE) devices featuring both BLE connected and BLE broadcast modes, to transform common objects of daily life into smart objects. Moreover, SHPIA 2.0 allows the col- lection and automatic labeling of different data types to provide indoor monitoring and assistance. Specifically, SHPIA 2.0 is designed to be adaptable to different home-based application scenarios, including human activity recognition, coaching systems, and occupancy detection and counting. The SHPIA platform is open source and freely available to the scientific community, fostering collaboration and innovation
ONTO-PLC: An ontology-driven methodology for converting PLC industrial plants to IoT
No full textA methodology is presented that guides a user in the transition from a plant governed by PLC towards one governed by SoC
Automatic generation of self-adaptive transactors from PSL assertions
No full textThis paper presents an approach to automatically generate transactors that implement TLM protocols for RTL IPs, such that the RTL IPs can be abstracted towards corresponding TLM models and easily integrated inside a TLM virtual pro- totype. The obtained transactor is self-adaptive, since it allows plugging the target IP in the virtual prototype independently from the protocol implemented by the corresponding TLM initiator. The transactor is automatically created from the set of PSL assertions that describe the temporal behaviour of the communication protocol of the original RTL IP
Enhancing Freezing of Gait Detection in Parkinson’s Through Fine-Tuned Deep Learning Models
No full textFreezing of Gait (FoG) is a common and disabling symptom in Parkinson's Disease (PD), characterized by a sudden and temporary inability to initiate or continue walking. FoG arises from various factors such as environmental triggers, or physiological status of people with Parkinson's. Traditional methods for preventing or alleviating FoG have limitations, prompting exploration into new technologies, such as the combination of sensing technologies and Deep Learning (DL) and Machine Learning (ML) algorithms. However, recognizing FoG with sensors and ML/DL poses challenges, such as the generalizability of the FoG recognition models over different individuals. Moreover, current approaches often require extensive time and effort to personalize the FoG recognition models. To mitigate these challenges, we propose a system that reduces the workload for creating personalized models through a fine-tuning approach. Our methodology has undergone rigorous testing in a subject-independent setup on a self-collected dataset of 22 subjects. Through the fine-tuning phase, we observed a remarkable average increase of up to 20.9 % in F1-score performance compared to the training and testing approach without fine-tuning
A Lightweight CNN for Real-Time Pre-Impact Fall Detection
No full textFalls can have significant and far-reaching effects on various groups, particularly the elderly, workers, and the general population. These effects can impact both physical and psychological well-being, leading to long-term health problems, reduced productivity, and a decreased quality of life. Numerous fall detection systems have been developed to prompt first aid in the event of a fall and reduce its impact on people's lives. However, detecting a fall after it has occurred is insufficient to mitigate its consequences, such as trauma. These effects can be further minimized by activating safety systems (e.g., wearable airbags) during the fall itself—specifically in the pre-impact phase—to reduce the severity of the impact when hitting the ground. Achieving this, however, requires recognizing the fall early enough to provide the necessary time for the safety system to become fully operational before impact. To address this challenge, this paper introduces a novel lightweight convolutional neural network (CNN) designed to detect pre-impact falls. The proposed model overcomes the limitations of current solutions regarding deployability on resource-constrained embedded devices, specifically for controlling the inflation of an airbag jacket. We extensively tested and compared our model, deployed on an STM32F722 microcontroller, against state-of-the-art approaches using two different datasets
Towards a wearable system for predicting the freezing of gait in people affected by Parkinson's disease
Full text linkSome wearable solutions exploiting on-body acceleration sensors have been proposed to recognize Freezing of Gait (FoG) in people affected by Parkinson Disease (PD). Once a FoG event is detected, these systems generate a sequence of rhythmic stimuli to allow the patient restarting the march. While these solutions are effective in detecting FoG events, they are unable to predict FoG to prevent its occurrence. This paper fills in the gap by presenting a machine learning-based approach that classifies accelerometer data from PD patients, recognizing a pre-FOG phase to further anticipate FoG occurrence in advance. Gait was monitored by three tri-axial accelerometer sensors worn on the back, hip and ankle. Gait features were then extracted from the accelerometer's raw data through data windowing and non-linear dimensionality reduction. A k-nearest neighbor algorithm (k-NN) was used to classify gait in three classes of events: pre-FoG, no-FoG and FoG. The accuracy of the proposed solution was compared to state of-the-art approaches. Our study showed that: (i) we achieved performances overcoming the state-of-the-art approaches in terms of FoG detection, (ii) we were able, for the very first time in the literature, to predict FoG by identifying the pre-FoG events with an average sensitivity and specificity of, respectively, 94.1% and 97.1%, and (iii) our algorithm can be executed on resource-constrained devices. Future applications include the implementation on a mobile device, and the administration of rhythmic stimuli by a wearable device to help the patient overcome the FoG
B-HAR: an open-source baseline framework for in depth study of human activity recognition datasets and workflows
Full text linkHuman Activity Recognition (HAR), based on machine and deep learning algorithms is considered one of the most promising technologies to monitor professional and daily life activities for different categories of people (e.g., athletes, elderly, kids, employers) in order to provide a variety of services related, for example to well-being, empowering of technical performances, prevention of risky situation, and educational purposes. However, the analysis of the effectiveness and the efficiency of HAR methodologies suffers from the lack of a standard workflow, which might represent the baseline for the estimation of the quality of the developed pattern recognition models. This makes the comparison among different approaches a challenging task. In addition, researchers can make mistakes that, when not detected, definitely affect the achieved results. To mitigate such issues, this paper proposes an open-source automatic and highly configurable framework, named B-HAR, for the definition, standardization, and development of a baseline framework in order to evaluate and compare HAR methodologies. It implements the most popular data processing methods for data preparation and the most commonly used machine and deep learning pattern recognition models.9 Pages, 3 Figures, 3 Tables, Link to B-HAR Library: https://github.com/B-HAR-HumanActivityRecognition/B-HA
Human Activity Recognition using Inertial, Physiological and Environmental Sensors: A Comprehensive Survey
Full text linkIn the last decade, Human Activity Recognition (HAR) has become a vibrant research area, especially due to the spread of electronic devices such as smartphones, smartwatches and video cameras present in our daily lives. In addition, the advance of deep learning and other machine learning algorithms has allowed researchers to use HAR in various domains including sports, health and well-being applications. For example, HAR is considered as one of the most promising assistive technology tools to support elderly's daily life by monitoring their cognitive and physical function through daily activities. This survey focuses on critical role of machine learning in developing HAR applications based on inertial sensors in conjunction with physiological and environmental sensors
Exploiting sub-graph isomorphism and probabilistic neural networks for the detection of hardware Trojans at RTL
No full textHardware Trojans (HTs) have been generally inserted at the lower levels of the digital system design and fabrication process, where, due to the high complexity of the hardware model, their detection is more difficult. However, RTL models are becoming more and more complex, making difficult the identification of malicious behaviours also at this level. Unfortunately, only a few verification techniques have been proposed for the identification of HTs in RTL descriptions. To fill in the gap, this paper proposes a technique that exploits graph-based features and a probabilistic neural network to identify and classify HTs at RTL. The approach identifies suspicious locations inside the RTL description according to a set of known HTs. In addition, it returns a couple of similarity indexes to measure the probability that a suspicious location actually contains a malicious behaviour
- …
