With the accelerating pace of life

muscle fatigue becomes a common issue affecting people’s quality of life and work efficiency.It not only hinders personal daily activities but also increases the social medical burden

so an urgent demand exists for real-time muscle fatigue detection technology.A surface electromyography(sEMG)signal acquisition and analysis system is developed to meet the requirements of real-time muscle fatigue detection.The system hardware employs an STM32G431CBT6 controller and a domestic chip LHE7909

and the visual analysis upper computer software is compiled and debugged in the Qt environment.To mitigate sEMG signal noise interference

a combined method of bandpass filtering and smoothed wavelet threshold denoising is adopted.Time-domain and frequency-domain features are fused

dimensionality reduction is performed via principal component analysis(PCA)

and the reduced feature are then fed into a support vector machine(SVM)classifier for fatigue state assessment.Static squat fatigue induction tests show that the system achieves an average recognition accuracy of 86.4% for relaxed

transitional

and fatigued states

providing technical support for sports rehabilitation and health monitoring.