Abstract
Remote health monitoring systems play an important role in the healthcare sector. Edge computing is a key enabler for realizing these systems, where it is required to collect big data while providing real-time guarantees. In this study, we focus on remote cuff-less blood pressure (BP) monitoring through electrocardiogram (ECG) as a case study to evaluate the benefits of edge computing and compression. First, we investigate the state-of-the-art algorithms for BP estimation and ECG compression. Second, we develop a system to measure the ECG, estimate the BP, and store the results in the cloud with three different configurations: (i) estimation in the edge, (ii) estimation in the cloud, and (iii) estimation in the cloud with compressed transmission. Third, we evaluate the three approaches in terms of application latency, transmitted data volume, and power usage. In experiments with batches of 64 ECG samples, the edge computing approach has reduced average application latency by 15%, average power usage by 19%, and total transmitted volume by 85%, confirming that edge computing improves system performance significantly. Compressed transmission proved to be an alternative when network bandwidth is limited and edge computing is impractical.
Subject
Control and Optimization,Computer Networks and Communications,Instrumentation
Reference18 articles.
1. Dey, N., Ashour, A.S., Bhatt, C., and James Fong, S. (2019). Healthcare Data Analytics and Management, Academic Press. Advances in Ubiquitous Sensing Applications for Healthcare.
2. Edge computing in smart health care systems: Review, challenges, and research directions;Hartmann;Trans. Emerg. Telecommun. Technol.,2019
3. Edge Computing for Smart Health: Context-Aware Approaches, Opportunities, and Challenges;Abdellatif;IEEE Netw.,2019
4. Electrocardiogram Data Compression Techniques for Cardiac Healthcare Systems: A Methodological Review;Jha;IRBM,2021
5. Homecare-Oriented Intelligent Long-Term Monitoring of Blood Pressure Using Electrocardiogram Signals;Fan;IEEE Trans. Ind. Inform.,2020
Cited by
2 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献