Design and implementation of a real-time sleep monitoring system based on edge computing
ZHAO Haonan
XUE Jihang
WANG Peng
DU Lidong
LI Zhenfeng
CHEN Xianxiang
FANG Zhen
Abstract：For the problem that the existing sleep monitoring systems is dependent on network and high-performance computers,we designed a real-time sleep monitoring system based on edge computing.The system consisted of wearable front-end acquisition de-vices and portable receiving terminals,and it could achieve real-time,stable monitoring of sleep states without network and high-per-formance computer support.The portable receiving terminals were able to perform sleep monitoring through sleep staging algorithms based on U2-Net and ECA-Net.Experimental results showed that the accuracy,average F1 score and Kappa value of the proposed sleep staging algorithm on the sleep-EDFx dataset were 85.2%,79.9%and 0.795,respectively.By integrating model compression and hardware acceleration techniques,the system can efficiently perform tasks network and computers free environment,and it has the abil-ity to extend to other sleep assessment tasks,which is expected to provide reliable sleep health assurance and support for plan adjust-ments in special scenarios.
Keywords：Sleep monitoringEdge computingReal-time monitoringSleep stagingWearable technologyPortable devicesModel compressionSpecial scenarios
Publication Date：2025-02-28
Online Publishing Date：2026-09-11(First online date of this platform, not the publication date of the document)
Pages：8( 16-23 )