Abstract
A new methodology for fault detection on wearable medical devices is proposed. The basic strategy relies on correctly classifying the captured physiological signals, in order to identify whether the actual cause is a wearer health abnormality or a system functional flaw. Data fusion techniques, namely fuzzy logic, are employed to process the physiological signals, like the electrocardiogram (ECG) and blood pressure (BP), to increase the trust levels of the captured data after rejecting or correcting distorted vital signals from each sensor, and to provide additional information on the patient's condition by classifying the set of signals into normal or abnormal condition (e.g. arrhythmia, chest angina, and stroke). Once an abnormal situation is detected in one or several sensors the monitoring system runs a set of tests in a fast and energy efficient way to check if the wearer shows a degradation of his health condition or the system is reporting erroneous values.

Abstract
This article addresses the design, development, and evaluation of T-shirt prototypes that embed novel textile sensors for the capture of cardio and respiratory signals. The sensors are connected through textile interconnects to either an embedded custom-designed data acquisition and transmission unit or to snap fastener terminals for connection to external monitoring devices. The performance of the T-shirt prototype is evaluated in terms of signal-to-noise ratio amplitude and signal interference caused by baseline wander and motion artefacts, through laboratory tests with subjects in standing and walking conditions. Performance tests were also conducted in a hospital environment using a T-shirt prototype connected to a commercial three-channel Holter monitoring device. The textile sensors and interconnects were realized with the assistance of an industrial six-needle digital embroidery tool and their resistance to wear addressed with normalized tests of laundering and abrasion. The performance of these wearable systems is discussed, and pathways and methods for their optimization are highlighted.

Abstract
High levels of dependability are required to promote the adherence by public and medical communities to wearable medical devices. The study presented herein addresses fault detection and diagnosis in these systems. The main objective resides on correctly classifying the captured physiological signals, in order to distinguish whether the actual cause of a detected anomaly is a wearer health condition or a system functional flaw. Data fusion techniques, namely fuzzy logic, artificial neural networks, decision trees and naive Bayes classifiers are employed to process the captured data to increase the trust levels with which diagnostics are made. Concerning the wearer condition, additional information is provided after classifying the set of signals into normal or abnormal (e.g., arrhythmia, tachycardia and bradycardia). As for the monitoring system, once an abnormal situation is detected in its operation or in the sensors, a set of tests is run to check if actually the wearer shows a degradation of his health condition or if the system is reporting erroneous values. Selected features from the vital signals and from quantities that characterize the system performance serve as inputs to the data fusion algorithms for Patient and System Status diagnosis purposes. The algorithms performance was evaluated based on their sensitivity, specificity and accuracy. Based on these criteria the naive Bayes classifier presented the best performance.

Abstract
Context-aware ubiquitous computing systems should be able to introspect the surrounding environment and adapt their behavior according to other existing systems and context changes. Although numerous ubiquitous computing systems have been developed that are aware of different types of context such as location, social situation, and available computational resources, few are aware of their computational behavior. Computational behavior introspection is common in reflective systems and can be used to improve the awareness and autonomy of ubicomp systems. In this paper, we propose a decentralized approach based on Simple Network Management Protocol (SNMP) and Universal Plug and Play (UPnP), and on state transition models to model and expose computational behavior. Typically, SNMP and UPnP are targeted to retrieve raw operational variables from managed network devices and consumer electronic devices, e.g., checking network interface bandwidth and automating device discovery and plug and play operations. We extend the use of these protocols by exposing the state of different ubicomp systems and associated state transitions statistics. This computational behavior may be collected locally or remotely from ubicomp systems that share a physical environment, and sent to a coordinator node or simply shared among ubicomp systems. We describe the implementation of this behavior awareness approach in a home health-care environment equipped with a VoIP Phone and a drug dispenser. We provide the means for exposing and using the behavior context in managing a simple home health-care setting. Our approach relies on a system state specification being provided by manufacturers. In the case where the specification is not provided, we show how it can be automatically discovered. We propose two machine learning approaches for automatic behavior discovery and evaluate them by determining the expected state graphs of our two systems (a VoIP Phone and a drug dispenser). These two approaches are also evaluated regarding the effectiveness of generated behavior graphs.

Abstract

Abstract
In this work, a cross-validation procedure is used to identify an appropriate Autoregressive Integrated Moving Average model and an appropriate state space model for a time series. A minimum size for the training set is specified. The procedure is based on one-step forecasts and uses different training sets, each containing one more observation than the previous one. All possible state space models and all ARIMA models where the orders are allowed to range reasonably are fitted considering raw data and log-transformed data with regular differencing (up to second order differences) and, if the time series is seasonal, seasonal differencing (up to first order differences). The value of root mean squared error for each model is calculated averaging the one-step forecasts obtained. The model which has the lowest root mean squared error value and passes the Ljung-Box test using all of the available data with a reasonable significance level is selected among all the ARIMA and state space models considered. The procedure is exemplified in this paper with a case study of retail sales of different categories of women's footwear from a Portuguese retailer, and its accuracy is compared with three reliable forecasting approaches. The results show that our procedure consistently forecasts more accurately than the other approaches and the improvements in the accuracy are significant.