10.3390/polym8030084

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Population aging has increased the need for continuous, non-invasive, and context-aware monitoring systems capable of supporting autonomy, safety, and early intervention in daily living environments. Multimodal biometric sensing offers an important technical basis for this purpose, as it combines physiological, motion-related, and environmental signals to provide a more complete view of older adults' functional and health-related conditions. However, many existing solutions remain fragmented, device-dependent, and insufficiently connected to core instrumentation requirements, including signal quality, sensor calibration, temporal synchronization, latency, energy consumption, interoperability, reliability, and data privacy. This article proposes an Edge-AI instrumentation framework for multimodal biometric sensing in active aging environments, supported by a structured analysis of recent literature on wearable, ambient, and context-aware sensing systems. The framework integrates wearable, ambient, and context-aware sensors with local processing capabilities to support signal acquisition, preprocessing, quality control, feature extraction, anomaly detection, and decision support close to the data source. By placing Edge AI within the instrumentation pipeline, the proposed framework identifies design requirements that may help reduce response time, limit unnecessary transmission of sensitive biometric data, and improve feasibility in home-based and assisted-living contexts. These expected benefits, however, require empirical testing through future prototype implementation and real-world evaluation. The article also defines a validation-oriented perspective for sensor-based active aging systems, covering technical, operational, and human-centered dimensions such as measurement accuracy, signal robustness, usability, privacy preservation, interoperability, reproducibility, energy efficiency, and system scalability. The proposed framework is intended to support the design, comparison, and validation of more reliable, interpretable, and reproducible sensor-based monitoring systems, while offering a structured basis for prototype development and future real-world evaluation in active aging environments.

Abstract: PubMed · Datensatz