The physiological monitoring of older people using wearable sensors has shown great potential in improving their quality of life and preventing undesired events related to their health status. Nevertheless, creating robust predictive models from data collected unobtrusively in home environments can be challenging, especially for vulnerable ageing population. Under that premise, we propose an activity recognition scheme for older people exploiting feature extraction and machine learning, along with heuristic computational solutions to address the challenges due to inconsistent measurements in non-standardized environments. In addition, we compare the customized pipeline with deep learning architectures, such as convolutional neural networks, applied to raw sensor data without any pre- or post-processing adjustments. The results demonstrate that the generalizable deep architectures can compensate for inconsistencies during data acquisition providing a valuable alternative.

In recent years, a number of research efforts have focused on effective and stable P2P architectures aiming at large scale and low bandwidth cost, real time video streaming systems. In our work, we consider a BitTorrent-like VoD system and try to answer the following questions: (1) how we can dynamically manage a P2P overlay using distributed-scalable algorithms in order to exploit the available upload bandwidth of highly heterogeneous participating peers and (2) how media servers (cloud) can minimize the amount of the bandwidth they offer while ensuring uninterrupted video playback. To illustrate the success of our approach, we have developed a network level P2P VoD packet simulator for demonstrating the efficiency, scalability and stability of our work under variant and dynamic conditions.

No abstract

No abstract

No abstract