No abstract

No abstract

No abstract

Alert intelligent device is an ambient assisted living (AAL) system that allows the evaluation of potentially dangerous situations for elderly people living alone at home. This evaluation is obtained by an ad hoc network of sensor nodes, working in conjunction with an ambient intelligence layer embedded in a personal computer that learns from user behaviour patterns and warns when a detected pattern differs significantly from previously acquired normal patterns. Each new datum read from sensors is processed in the ambient intelligence layer through three processing levels: shallow, intermediate and deep. The shallow processing level focuses on physical data and sensory features. The intermediate level covers information interpretation and its translation into the form required by the third level: the reasoning processing or deep level. In alert intelligent device, energy is a critical issue, so that sensor devices need to be properly designed and managed to achieve significant energy saving. The use of bed/chair occupancy sensors is mandatory for this kind of ubiquitous computing system. A first way to approach this problem relied on the use of pressure mats, but several environmental drawbacks showed them inappropriate as an efficient and reliable solution for large volume deployments. Moreover, solutions based on force‐to‐resistor transducers entail power consumption budgets that keep them from being integrated on wireless sensor nodes. In this paper, a force‐capacitive transducer based sensor has been proposed, implemented and tested. This sensor is based on electromechanical film (EMFi) transducers, which are able to detect force variations in a quasi‐passive way. This kind of transducer behaves as capacitors with variable capacitance depending on the force exerted on its surface. We have developed a new technique to carry out the characterization of these transducers, where the detection of a force change is used to trigger an active mechanism that allows us to measure the weight by means of a novel modelling approach. A low‐power wireless sensor node prototype that includes this new transducer has been assembled and tested on a wide range of weights. Occupancy detection achieved by this technology has proven to be successful, increasing the total power consumption of the node by less than 15%, which makes it suitable for implementation.

Accidental falls of our elderly, and physical injuries resulting, represent a major health and economic. Falls are the most common causes of serious injuries and a major health threats in the stratum of older population. Early detection of a fall is a key factor when trying to provide adequate care to the elderly person who has suffered an accident at home. In this paper, we present a support system for detecting falls of an elder person by a static wireless nonintrusive sensorial infrastructure based on heterogenous sensor nodes. This previous infrastructure, named AID (Alarm Intelligent Device), is an AAL (Ambient Assisted Living) system that allows to infer a potential fall. We have developed, different to other contributions, a specific low-power multi-hop network consists of nodes (Motes) that wirelessly communicate to each other and are capable of hopping radio messages to a base station where they are passed to a PC (or other possible client). The goal of this project is 1) to provide alerts to caregivers in the event of an accident, acute illness or strange (possibly dangerous) activities, and 2) to enable that authorized and authenticated caregivers by means of a itinerant wearable mote can be inserted into mesh and interact with it. In this paper, we describe an ubiquitous assistential monitoring system at home.