Explore the vast range of titles available.

Indoor localization systems have already wide applications mainly for providing localized information and directions. The majority of them focus on commercial applications providing information such us advertisements, guidance and asset tracking. Medical oriented localization systems are uncommon. Given the fact that an individual’s indoor movements can be indicative of his/her clinical status, in this paper we present a low-cost indoor localization system with room-level accuracy used to assess the frailty of older people. We focused on designing a system with easy installation and low cost to be used by non technical staff. The system was installed in older people houses in order to collect data about their indoor localization habits. The collected data were examined in combination with their frailty status, showing a correlation between them. The indoor localization system is based on the processing of Received Signal Strength Indicator (RSSI) measurements by a tracking device, from Bluetooth Beacons, using a fingerprint-based procedure. The system has been tested in realistic settings achieving accuracy above 93% in room estimation. The proposed system was used in 271 houses collecting data for 1–7-day sessions. The evaluation of the collected data using ten-fold cross-validation showed an accuracy of 83% in the classification of a monitored person regarding his/her frailty status (Frail, Pre-frail, Non-frail).

Activity recognition has gained significant attention recently, due to the availability of smartphones and smartwatches with movement sensors which facilitate the collection and processing of relevant measurements, by almost everyone. Using the device-embedded sensors, there is no need of carrying dedicated equipment (inertia measurement units or accelerometers) and use complex software to process the data. This approach though, has the disadvantage of needing to carry a device during the monitoring time. WiFi channel state information (CSI) offers a passive, device-free opportunity, for monitoring activities of daily living in non-line-of-sight conditions. In this paper, Mel frequency cepstral coefficient (MFCC) feature extraction, used successfully for audio signals, is proposed for CSI time-series classification. The applicability of the proposed features in activity recognition has been evaluated using three classification methods, convolutional neural networks (CNN), long short term memory recurrent neural networks and Hidden Markov models, in comparison to currently used feature extraction methods (discrete wavelet transform, short time fourier transform). MFCC feature extraction achieves higher accuracy in activity classification than the compared methods, as verified by evaluation with two activity datasets, in particular 95% accuracy precision achieved in activity recognition using MFCC features in combination with CNN classifier.