Syin Chan

Feasibility study of a real-time wandering detection algorithm for dementia patients

In this paper, we present a new algorithm that automatically classifies wandering patterns (or behaviors) of patients with Alzheimers disease and other different types of dementia. Experimental results on a real-life dataset show that this algorithm can provide a robust and credible assistive technology for monitoring patients with dementia (PWD) who are prone to wandering. Combined with indoor and outdoor location technologies using ubiquitous devices such as smart phones, we also demonstrate the feasibility of a remote mobile healthcare monitoring solution that is capable of reasoning about wandering behaviors of PWD and real-time detection of abnormalities that require timely intervention from caregivers.

MAC Protocol for Two Level QoS Support in Cognitive Radio Network

The exponential increase in the number of mobile devices using large bandwidth and real-time applications has met with challenges of spectrum scarcity and real-time MAC design. To this end we propose an energy efficient cognitive radio MAC protocol with two-level QoS support, called as TQCR-MAC. We have divided the traffic into two types-real time and non-real time data traffic. The TQCR-MAC protocol utilizes the unused spectrum and provides QoS support for different traffic. The TQCR-MAC exploits the combination of multiple channels and TDMA scheme to further improve the QoS provisioning. The simulation result shows that our proposed protocol improves the network performance significantly, especially when the network is either biased towards real time or non-real time data traffic.

Energy Aware Spectrum Decision Framework for Cognitive Radio Networks

Recently cognitive radios have been proposed to mitigate the problem of spectrum scarcity and spectrum under-utilization. One of the key aspect of the cognitive radio is the optimal channel selection. The cognitive radio system selects a licensed band which has acceptable availability and also satisfies the system needs in terms of bandwidth and power constraints. To this end we propose an energy aware spectrum decision framework for cognitive radio network (ESDF-CR), which guarantees a certain level of QoS to the system running on low power, by selecting those channels which require lesser power for transmission. Channel selection and packet admission control techniques have been proposed to realize the spectrum decision framework. A comparison is done with random channel selection scheme to show the effectiveness of the proposed framework.

Reconfigurable Channel Selection Technique for Cognitive Radio Network With Heterogeneous Primary Bands

Recently, the Federal Communications Commission has mandated the use of a database to observe the spectral opportunity at a given place over time along with the use of spectrum sensing. It is arduous to obtain the database for spectral opportunity for all places, and since spectral opportunities rapidly vary with location, a new paradigm to obtain the spectral opportunity needs to be determined. We propose a method to obtain the spectral opportunity at a given location for specific channels, such as television (TV) and frequency-modulated (FM) bands, which follow deterministic time usage pattern. Since these channels have a periodic deterministic time usage pattern for every 24 h, learning the spectral opportunities and utilizing these for consequent time periods have better efficiency than other proposed schemes. Once the spectral opportunity is known and if we have multiple channels from which to select, then a decision on the selection of a channel is made. The spectrum decision is done to satisfy the constraints imposed by a user. Since, with our proposed method, the spectral opportunity is known ahead of time, better spectrum decisions can be made. To this end, we propose three different channel selection techniques with different constraints applied. The first channel selection technique provides the selection of channels with a minimum number of channel switching and, thus, reduces the jitters, switching delays, and packet losses. This channel selection technique is appurtenant to real-time applications such as voice over Internet Protocol. The second channel selection technique provides the selection of channels that maximizes the overall throughput of the system and is particularly useful for bandwidth-intensive applications such as BitTorrent. The third channel selection technique provides a compromised value between minimal channel switching and maximal throughput, thus providing a control over the bandwidth demand and real-time requirement. A comparison of these channel selection techniques is done with three other studies to demonstrate the competency in terms of a number of channel switching and throughput obtained in the system.

A unified grid-based wandering pattern detection algorithm

The aim of this study was to develop and validate a robust algorithm for indoor and outdoor wandering pattern detection and to analyse the relationship of these patterns to other clinical measures. Much of the previous work in this area addressed the measurement of wandering indoors or outdoors and to the best of our knowledge, there has not been a unified algorithm proposed which can deal with both scenarios. We present a novel grid-based layout representation strategy to identify the patterns, which is applicable to both indoor and outdoor scenarios. The algorithm is sufficiently robust to identify interleaving and hybrid patterns and performed with identification accuracy of 90% on a real-world sample.

mHealth Sensors, Techniques, and Applications for Managing Wandering Behavior of People with Dementia: A Review

Managing wandering behavior of people with dementia (PWD) has become increasingly imperative for these reasons: its high prevalence (60%) among PWD, its negative outcomes such as falls or elopement, and its burden on caregivers. In this chapter, we discuss the emergence of sensors, techniques, and applications for managing wandering behavior of PWD. First, we briefly present the 5Ws1H (WHO, WHAT, WHERE, WHEN, WHY, HOW) conceptual map of wandering science including stakeholders (WHO), measurements of wandering (WHAT), environments in which wandering takes place (WHERE), detection of wandering (WHEN), causes of wandering (WHY), interventions of wandering (HOW). Second, we introduce a framework that identifies specific groups of mHealth and eHealth assistive technologies for managing dementia-related wandering. Third, we review existing technological works that address these 4 domains in the 5Ws1H conceptual map: WHAT-WHERE-WHY-HOW. In particular, we explore mHealth sensors to geo-fence and prevent elopement, mHealth devices to track and locate PWD who wander, information services to assist caregivers, eHealth tools to measure dimensions of dementia-related wandering, and mHealth tools that analyze proximal factors as well as study background factors. Based on this review, we further discuss research and design issues, human factors, ethics, security and privacy that need to be considered when implementing mHealth applications for wandering management. We conclude the chapter by highlighting the future research work in this area.

A mobile-health application to detect wandering patterns of elderly people in home environment

Wandering is a common and risky behavior in people with dementia (PWD). In this paper, we present a mobile healthcare application to detect wandering patterns in indoor settings. The application harnesses consumer electronics devices including WiFi access points and mobile phones and has been tested successfully in a home environment. Experimental results show that the mobile-health application is able to detect wandering patterns including lapping, pacing and random in real-time. Once wandering is detected, an alert message is sent using SMS (Short Message Service) to attending caregivers or physicians for further examination and timely interventions.

A predictive location-aware algorithm for dementia care

Some mentally impaired but otherwise physically healthy individuals may have a tendency of wandering or difficulties using public transportation. Intelligent assistive technologies which are able to learn the individuals travel behavior and prompt anomalous events such as when the person deviates from expected destinations would greatly enhance the independence and safety of the individual and also reduce the stress on family members and caregivers. This paper presents the design and preliminary evaluation of a prediction model that is a critical component of a mobile and personal wellness management system for patients with dementia. We define a generic architecture of assurance systems aiming to ensure the safety and well-being of such patients and also develop an enhanced class of state predictor for location-aware applications catering to dementia care. With our new scheme, the prediction accuracy has reached up to 90% compared to the average 76% to 81% accuracy achieved by Markov, Bayesian network, Multilayer Perceptron, Elman net or the original state predictor with confidence estimator. Considerations for the development of future dementia care systems are also addressed.

Classification of pH levels using a mobile phone

We present a novel and automated system for estimating pH levels from non-calibrated camera phone images. Experimental results show that pH levels reported by the software are consistent with the ground truth estimation.

TQCR-media access control: two-level quality of service provisioning media access control protocol for cognitive radio network

Recently cognitive radio has been proposed to solve the problem of spectrum scarcity with the help of an opportunistic utilisation of unused spaces in the spectrum. The implementation of cognitive radio for real-time application faces with the challenge of real-time media access control (MAC) design. To this end the authors propose an energy efficient cognitive radio MAC protocol with two level of quality of service (QoS) support: two level quality of service provisioning in cognitive radio (TQCR)-MAC. The data traffic in the cognitive radio network is classified as real-time traffic and non-real-time traffic, with former having higher priority over latter. The proposed protocol utilises variable interframe spaces for implementing different priorities to data and also exploits the combination of multiple channels and time division multiple access (TDMA) scheme to further improve the QoS provisioning. The TDMA slots in different channels are used as the data carrier segment, in which a real-time data can take over the time slots reserved by non-real-time data. Thus, non-real traffic may suffer from starvation, and hence a guarantee on the non-real-time data is provided with the help of a starvation prevention algorithm. The performance analysis of the proposed MAC protocol is done against two similar MAC protocols, which shows that the performance is improved significantly, especially when the network is either biased towards real-time or non-real-time data traffic.