This paper investigates two major issues in using a tri-axial accelerometer-embedded mobile phone for continuous activity monitoring, i.e. the difference orientations and locations of device. Two experiments with total ten test subjects performed six daily activities were conducted this study: one device fixed on waist sixteen different another three (i.e., shirt-pocket, trouser-pocket waist) orientations. For handling varying orientations, projection-based method coordinate system...

This paper presents a real-time method for detecting fall at different phases using wireless tri-axial accelerometer and reports the classification performance when sensor is placed on body parts. The proposed hybrid framework combines rule-based knowledge representation scheme with time control mechanism machine-learning-based activity classification. Real-time temporal reasoning performed standard inference engine. validated detection performance, false alarm evaluation, comparison highly...

This paper describes an orientation-independent method for detecting activities of daily living based on reference coordinate transformation. With the proposed method, a classification model can be trained using data acquired during specific sensor orientation and applied to other input signals regardless device. The technique is validated activity recognition experiments with four different orientations single tri-axial accelerometer placed waist 13 subjects performing sub-class living. A...

This paper presents a hand-gesture based interface for facilitating communication among speech- and hearing-impaired disabilities. In the system, wireless sensor glove equipped with five flex sensors 3D accelerometer is used as input device. By integrating speech synthesizer onto an automatic gesture recognition user's hand gestures can be translated into sounds. this study, we proposed hierarchical framework on combined use of multivariate Gaussian distribution, bigram set rules model...

This paper focuses on optimal sensor positioning for monitoring activities of daily living and investigates different combinations features models positions, i.e., the side waist, front chest, thigh, head, upper arm, wrist, ankle. Nineteen are extracted, feature importance is measured by using Relief-F selection algorithm. Eight classification algorithms evaluated a dataset collected from young subjects elderly subjects, with two experimental settings. To deal sampling rates, signals high...

This paper presents a new design of wireless sensor glove developed for American Sign Language finger spelling gesture recognition. Five contact sensors are installed on the glove, in addition to five flex fingers and 3D accelerometer back hand. Each pair combined into same input channel BSN node order save number channels installation area. After which, signal is analyzed separated features by software. With electrical contacts wirings made conductive fabric threads, has become thinner more...

This paper presents a study of two simple methods for reducing the complexity instance-based classification technique and demonstrates their use in device-context independent activity recognition on mobile phone. A projection-based method signal rectification has been implemented an iPhone order to handle with variation device orientations. The transformation matrix is estimated ten-second dynamic data buffer. To search suitable set training prototypes implementation, experiment conducted...

In high-density road networks, with each vehicle broadcasting multiple messages per second, the arrival rate of safety can easily exceed at which digital signatures be verified. Since not all verified, algorithms for selecting to verify are required ensure that receives appropriate awareness about neighbouring vehicles. This paper presents a novel scheme select important verification in vehicular ad hoc networks (VANETs). The proposed uses location and direction sender, as well proximity...

False alarms in cardiac monitoring affect the quality of medical care, impacting on both patients and healthcare providers. In continuous using wireless Body Sensor Networks (BSNs), ECG signals can be deteriorated owing to several factors, e.g., noises, low battery power, network transmission problems, often resulting high false alarm rates. addition, body movements occurring from activities daily living (ADLs) also create alarms. This paper presents a two-phase framework for arrhythmia...

Five well-known arrhythmia classification algorithms were compared in this paper based on the recommendations AAMI standard. They are C4.5, k-Nearest Neighbor, Multilayer Perceptron, PART, and Support Vector Machine, respectively, with inputs rela

This paper presents a process of feature selection, and classification algorithm evaluation for continuous sleep monitoring system, using tri-axial accelerometer attached to the subject's chest. Two selection algorithms, i.e., Relief-F support vector machine recursive elimination (SVM-RFE), seven Bayesian network, naive machine, pruned decision tree, instance-based learning with one neighbor, three neighbors, multi-layer perceptron, were investigated. By four features according rank obtained...

Pervasive sensing with Body Sensor Networks (BSNs) is a promising technology for continuous health monitoring. Since the sensor nodes are resource-limited, on-node processing and advertisement of digested information via BLE beacon technique that can enable node gateway to communicate more extend node’s lifetime before requiring recharging. This study proposes Dynamic Light-weight Symmetric (DLS) encryption algorithm designed developed address challenges in data protection real-time secure...

At an age when information access is no longer limited by physical barriers or distances, the desire to have continuous sensing and monitoring, rather than simple episodic snapshot measurements represents current trend in almost all applications ranging from environment, transport, infrastructure well being, sports, health care. The term body sensor network (BSN) was coined harness several allied technologies that underpin development of pervasive for care, other require "ubiquitous"...

Context-aware sensing is an integral part of the body sensor network (BSN) design and it allows understanding intrinsic characteristics sensed signal determination how BSNs should react to different events adapt its monitoring behaviour. The purpose this paper propose a novel spatio-temporal self-organising map that minimises number neurons involved whilst maintaining high accuracy in class separation for both static dynamic activities

In this paper, a study of assistive devices with multi-modal feedback is conducted to evaluate the efficiency haptic and auditory information towards users' mouse operations. Haptic feedback, generated by combination wheels driven motors, provided through use mouse. Meanwhile, audio either in form synthesized directional speech or signal. Based on these interfaces, set experiments are compare their efficiencies. The measurement criteria used experiment distance regarding target circle...

This paper focuses on optimal sensor positioning for monitoring activities of daily living and investigates different combinations features models positions, i.e., the side waist, front chest, thigh, head, upper arm, wrist, ankle. Sixteen are extracted feature importance is measured by using Relief-F selection algorithm. Eight classification algorithms evaluated a dataset collected from young subjects that elderly subjects, with two experimental settings. To deal sampling rates, signals high...

Based on rules and ontologies, this paper proposes a framework for predicting types of arrhythmia from electro-cardiogram (ECG) signals acquired using BSN node. Using terms in an ECG signal ontology, are annotated by locating the positions elementary waves, including their onset, offset, peak positions. Rules used extracting features, e.g., heart rate, PR intervals, RR QRS signals. An indicator ontology is constructed order to define concepts representing different characteristics waveforms,...