A High Precision and Low Energy Tunable On-Chip Clock Design for In-Textile Computing

Author: ORCID icon orcid.org/0009-0001-6350-8448
Foysal, Md Fahim, Electrical Engineering - School of Engineering and Applied Science, University of Virginia
Advisor:
Calhoun, Benton, EN-Elec & Comp Engr Dept, University of Virginia
Abstract:

The growing demand for wearable electronics within textiles presents unique challenges that require
advanced data collection and processing systems embedded in textiles. This requires precise
time-stamping among sensor and processing units while ensuring power efficiency. These timing
solutions require precise, low-power, and adaptable clock sources essential to support real-time
processing and data synchronization across multiple integrated devices. This research introduces
a tunable on-chip current-starved ring oscillator, specifically optimized in precision tuning steps
and running at sub-threshold weak inversion to achieve ultra-low power for textile applications,
addressing the need for power efficiency, accuracy, and adaptability in fabric-based electronics.
The oscillator consumes only 170 femtojoules per cycle, a level of efficiency critical for resourceconstrained
environments such as wearable fabrics.
Operating at 1.1V, the proposed oscillator achieves a frequency range of 88.35 KHz to 108.7
KHz with a tuning step size of 80 Hz (0.08%), providing versatile performance to accommodate
a variety of dynamic sensing, processing, and communication tasks inherent to textile-integrated
systems. This capability enables adaptive tuning of the clock frequency and resolution to match
the changing demands of embedded systems within the fabric, ensuring clock stability and efficiency
without compromising power consumption. Demonstrated through integration within an
ARM Cortex M0+ SoC, this clock design meets the stringent timing requirements of advanced
textile-based wearable applications. The successful implementation of this clock as part of a multichiplet
SoC system enables robust synchronization of embedded sensors, paving the way for new
applications in wearable technology that require precise timing control for complex, fabric-based
computing environments.

Degree:
MS (Master of Science)
Language:
English
Rights:
All rights reserved (no additional license for public reuse)
Issued Date:
2025/04/21