High-Accuracy Clock Synchronization in Low-Power Wireless sEMG Sensors.

Area of Science:

• Biomedical Engineering
• Wearable Technology
• Signal Processing

Background:

• Wireless surface electromyography (sEMG) sensors offer practical mobility but face challenges with radio link latencies impacting multi-sensor time synchronization.
• Minimizing power consumption in sEMG sensors is crucial for extended use, often necessitating multi-clock domain designs for optimized operation.

Purpose of the Study:

• To design and implement an sEMG sensor utilizing Bluetooth Low Energy (BLE) with multiple clock domains for power saving.
• To address and solve the time synchronization challenges introduced by BLE communication and multi-clock domain architecture.

Main Methods:

• Developed an sEMG sensor incorporating Bluetooth Low Energy (BLE) communication and three distinct clock domains (receiver, microcontroller, ADC).
• Conducted a detailed experimental timing analysis of the BLE connection to understand and mitigate latency.
• Implemented a dual-stage filtering mechanism to reduce timestamp measurement noise and improve synchronization accuracy.

Main Results:

• Achieved accurate time synchronization across the receiver, microcontroller, and ADC clock domains.
• Reported a total jitter of only 47 µs RMS for a 1.25 ms sampling period.
• Demonstrated power savings ranging from 10% to 50% through the dedicated ADC clock domain, varying with data rate.

Conclusions:

• The developed sEMG sensor effectively overcomes BLE latency issues for precise multi-sensor time synchronization.
• The multi-clock domain design successfully enhances power efficiency without compromising synchronization accuracy.
• This work provides a robust solution for reliable and energy-efficient wireless sEMG data acquisition.