Ambient-assisted living (AAL) is promising to become a supplement of the current care models, providing enhanced living experience to people within context-aware homes and smart environments. Activity recognition based on sensory data in AAL systems is an important task because 1) it can be used for estimation of levels of physical activity, 2) it can lead to detecting changes of daily patterns that may indicate an emerging medical condition, or 3) it can be used for detection of accidents and emergencies. To be accepted, AAL systems must be affordable while providing reliable performance. These two factors hugely depend on optimizing the number of utilized sensors and extracting robust features from them. This paper proposes a generic feature engineering method for selecting robust features from a variety of sensors, which can be used for generating reliable classi cation models. From the originally recorded time series and some newly generated time series [i.e., magnitudes, rst derivatives, delta series, and fast Fourier transformation (FFT)-based series], a variety of time and frequency domain features are extracted. Then, using two-phase feature selection, the number of generated features is greatly reduced. Finally, different classi cation models are trained and evaluated on an independent test set. The proposed method was evaluated on ve publicly available data sets, and on all of them, it yielded better accuracy than when using hand-tailored features. The bene ts of the proposed systematic feature engineering method are quickly discovering good feature sets for any given task than manually nding ones suitable for a particular task, selecting a small feature set that outperforms manually determined features in both execution time and accuracy, and identi cation of relevant sensor types and body locations automatically. Ultimately, the proposed method could reduce the cost of AAL systems by facilitating execution of algorithms on devices with limited resources and by using as few sensors as possible.

Body Sensor Networks (BSN) are a technology used to supply individual’s vital signs to ubiquitous and pervasive systems. This systems applied to the healthcare area uses BSN to monitor patient’s vital signs remotely, that is without them being in a hospital environment. This technology applied as a base to this application assists healthcare professional to remotely monitor patients, which contribute to solve healthcare services problems. One of those problems is the crowded treatments centers. The usage of BSN, by conditioning remote monitoring, can colaborate to reduce the patients’ needs to consistently go to hospitals, thus contributing to reduce the number of patients in those environments. Given this tecnology potential comes the need to develop good quality applications. However, considering an analysis of BSN projects applied to the healthcare field, the usage of inadequated hardware platforms are noticed. Therefore aiming to make the development of this type of applications easier, it is proposed a simulator specific for BSN applied to the healthcare field. Hence, this work’s goal is to show this application simulator through the usage of a simulated hardware platform which takes intoconsiderationhealthcare demands to provide a test environment.

Body sensor networks (BSNs) enable continuous monitoring of patients anywhere, with minimum constraints to daily life activities. Although the IEEE 802.15.4 and ZigBee® (ZigBee Alliance, San Ramon, CA) standards were mainly developed for use in wireless sensors network (WSN) applications, they are also widely used in BSN applications because of device characteristics such as low power, low cost, and small form factor. However, compared with WSNs, BSNs present some very distinctive characteristics in terms of traffic and mobility patterns, heterogeneity of the nodes, and quality of service requirements. This article evaluates the suitability of the carrier sense multiple access–collision avoidance protocol, used by the IEEE 802.15.4 and ZigBee standards, for data-intensive BSN applications, through the execution of experimental tests in different evaluation scenarios, in order to take into account the effects of contention, clock drift, and hidden nodes on the communication reliability. Results show that the delivery ratio may decrease substantially during transitory periods, which can last for several minutes, to a minimum of 90% with retransmissions and 13% without retransmissions. This article also proposes and evaluates the performance of the BSN contention avoidance mechanism, which was designed to solve the identified reliability problems. This mechanism was able to restore the delivery ratio to 100% even in the scenario without retransmissions.

This paper describes the development of wireless sensor network prototype that gathers biometrical data and posture information from several wearable sensor networks and sends it in real-time to a personal computer where the information is monitored and stored. The wireless sensor network is based on a low power real-time MAC protocol that was designed and implemented in the MICAz platform. This paper also presents some analytical results and several experimental results regarding the behavior of the developed system.

We present an experimental performance evaluation of ZigBee networks in the context of data-intensive body sensor networks (BSNs). IEEE 802.15.4/ZigBee devices were mainly developed for use in wireless sensors network (WSN) applications; however, due to characteristics such as low power and small form factor, they are also being widely used in BSN applications, making it necessary to evaluate their suitability in this context. The delivery ratio and end-to-end delay were evaluated, under contention, for both star and tree topologies. The reliability of the ZigBee network in a star topology without hidden nodes was very good (delivery ratio close to 100%), provided the acknowledgement mechanism was enabled. On the other hand, the performance in a tree topology was degraded due to router overload and the activation of the route maintenance protocol triggered by periods of high traffic load. The effect of the devices’ clock drift and hidden nodes on the reliability of the star network was modeled and validated through experimental tests. In these tests, the worst-case delivery ratio when the acknowledgment is used decreased to 90% with two sensor nodes, while for the non-acknowledged mode the result was of 13%. These results show that a mechanism for distributing the nodes’ traffic over the time is required to avoid BSN performance degradation caused by router overload, clock drift and hidden node issues.

Health promotion in hospital environments can be improved using the most recent information and communication technologies. The Internet connectivity to small sensor nodes carried by patients allows remote access to their bio-signals. To promote these features the healthcare wireless sensor networks (HWSN) are used. In these networks mobility support is a key issue in order to keep patients under realtime monitoring even when they move around. To keep sensors connected to the network, they should change their access points of attachment when patients move to a new coverage area along an infirmary. This process, called handover, is responsible for continuous network connectivity to the sensors. This paper presents a detailed performance evaluation study considering three handover mechanisms for healthcare scenarios (Hand4MAC, RSSI-based, and Backbone-based). The study was performed by simulation using several scenarios with different number of sensors and different moving velocities of sensor nodes. The results show that Hand4MAC is the best solution to guarantee almost continuous connectivity to sensor nodes with less energy consumption.

A Body Sensor Network can sense health parameters directly on the patient’s body, allowing 24/7 monitoring in an unobtrusive way. Several tiny sensors collect and route data to a special sink node. A new intra-vaginal biosensor was developed to study the relation between temperature variations and women health conditions, such as ovulation period, among others. We present a biosensor prototype and some initial results on real scenarios with a woman. One of the main issues in a body sensor network is the transformation of the sensor raw data into meaningful medical data for medical staff. Several approaches exist, from mobile device-based approaches to more powerful hardware such as a personal computer. This paper presents our current work in body sensor networks, namely a prototype for intra-vaginal temperature monitoring with initial results, and a mobile tool for data presentation of a three-tier body sensor network. The gathered results demonstrate the feasibility of the approach, contributing to the widespread application of body sensor networks.

Born on military applications, wireless sensor networks (WSNs) application grew on the promise of environment sensing and data processing capability at low cost. These networks can hold hundreds or even thousands of smart sensing nodes with processing and sensing capabilities and even integrated power through a dedicated battery. This paper surveys on the application of wireless sensor networks to healthcare promotion, namely with the use of biosensor technology applied to body sensor networks. On a wireless body sensor network, a person wears biosensors to gather data, while doing their daily activities. Currently, engineers and medical staff are cooperating on finding new ways to properly gather meaningful data on-site and achieve a convenient way to process these data for research and on-site medical decision. New challenges that such approach brings are also considered. Moreover, it is shown that wireless sensor networks provide the technology to built wireless sensing and create a convenient infrastructure for multiple data gathering in healthcare applications. Together with real successful examples, we demonstrate the great usefulness of wireless sensor networks in healthcare promotion. The paper concludes with some guidelines for future work.

Sensor nodes are small devices able to collect and retrieve sensorial data. The use of these sensors for medical purposes offers valuable contributions to improve patients’ healthcare, both for diagnosis and therapeutics monitoring. An important and common parameter used on healthcare diagnosis is the body temperature. It is monitored on several matters related with gynecological and obstetrics issues but, usually it is measure at the skin surface. Then, this paper proposes the design concepts of a new intra-body sensor for long-term intra-vaginal temperature collection. The embedded IEEE 802.15.4 communication module allows the integration of this sensor in wireless sensor networks for remote data access and monitoring. It is presented the sensor architecture, the construction of the corresponding testbed, and its performance evaluation. This sensor may be used on several applications, including fertile and ovulation period detection, and preterm labor prevention.

Biofeedback data acquisition is an extremely important task in body sensor networks (BSNs). Data collected by sensors need to be processed in order to be shown in an easy and meaningful way for the user. The use of mobile devices may support and offer new user experiences. When connected to a BSN they can aggregate and process data collected by each sensor, providing a mobile solution for a healthcare system. This mobility offers a better patients' quality of life allowing a regular daily routine and always under monitoring. This paper proposes a Symbian-based mobile solution for intra-body temperature monitoring. Mobile device connects wirelessly to an intra-vaginal temperature sensor and interacts with sensor for temperature data collection and monitoring. This system helps women to detect their fertile and ovulation periods by the increasing of their intra-vaginal temperature. The mobile system was tested and validated with success and it is available for regular use.