UJAmI Location: A Fuzzy Indoor Location System for the Elderly

Abstract

Due to the large number of elderly people with physical and cognitive issues, there is a strong need to provide indoor location systems that help caregivers monitor as many people as possible and with the best quality possible. In this paper, a fuzzy indoor location methodology is proposed in a smart environment based on mobile devices and Bluetooth Low Energy (BLE) beacons where a set of Received Signal Strength Indicators (RSSI) is received by mobile devices worn by the inhabitants. The use of fuzzy logic and a fuzzy linguistic approach is proposed to deal with the imprecise nature of the RSSI values, which are influenced by external factors such as radio waves, causing significant fluctuations. A case study carried out at the Smart Lab of the University of Jaén (UJAmI Smart Lab) is presented to demonstrate the effectiveness of the proposed methodology, where our proposal is compared with a non-fuzzy logic approach, obtaining an accuracy of 91.63%, approximately 10 points higher than the methodology without using fuzzy logic. Finally, our theoretical proposal is accompanied by a description of the UJAmI Location system, which applies the theory to the functionality of locating elderly people in indoor environments.

Evaluation of the Impact of the Sustainable Development Goals on an Activity Recognition Platform for Healthcare Systems

“…The selected devices have been proven to be successful in recent activity recognition methodologies [ 4 , 11 , 45 ]. They can be divided into two groups according to their purpose: location or daily activity detection.…”

“…Healthy aging is a process of developing and maintaining functional capacity and enabling well-being in old age. Therefore, the goal of the decade is to optimize the functional capacity of the elderly, where technology, software, hardware, engineering and their intelligent applications will play an important role [ 2 , 3 , 4 ]. An essential aspect in smart environments is data management and the integration of sensor installations, Internet of Things and equipment without disrupting people’s daily lives and without creating a very marked digital divide for the elderly, disabled or vulnerable.…”

ACTIVA: Innovation in Quality of Care for Nursing Homes Through Activity Recognition

“…A multitude of proposed models can be found in the literature [27], [32]- [35], ranging from machine learning models that make use of classifiers based on decision trees, k-nearest neighbours, support vector machines, among others, and even deep learning [27], [36]. In addition, other elements such as indoor location systems [15] are included to support activity recognition and thus solve the problem of multi-occupancy.…”

“…Another important aspect in each proposed system is performing some type of quantitative evaluation. In general, this assessment is usually performed using accuracy as the main characteristic, although other parameters such as F1 score, recall, precision and ROC curve are also measured [9]- [11], [15], [19], [20], [23], [30]- [32], [46]. Thus, it is possible to compare different methods based on these metrics.…”

“…To do so, the use of fuzzy logic is used for activity recognition with sensors as it allows for the handling of uncertainty and imprecision inherent in sensor data [10], [13], [15], [17]. In these types of systems, sensors may provide inaccurate or noisy data, making it challenging to identify the activities performed by users accurately.…”