Michella Rustom demonstrates real-time sensing of magnetic coupling during joint flexion using the wearable magnetic induction-based sensors. (Photo Credit: Michella Rustom)
Motion tracking, the technology that captures how objects move, plays a key role across healthcare, sports and entertainment: from helping stroke patients relearn how to walk to analyzing an athlete’s swing or powering immersive virtual reality.
For decades, motion-tracking technologies have stumbled over the same challenges: bulky equipment, reliance on external cameras, sensors that drift over time and the inability to capture movements in real time.
Existing motion tracking solutions today are either camera or inertial sensor-based — accelerometer and gyroscope — both of which have significant limitations. Camera-based systems require bulky setups and a clear line of sight between cameras and the subject, which limits motion tracking to confined environments with no obstructions. This makes it nearly impossible to capture the movements of hidden body parts or at difficult angles. Meanwhile, accelerometer-based systems using inertial measurement units (IMUs), suffer from drift, leading to errors in the predicted position accumulating over time and requiring frequent recalibration. These limitations pose a significant challenge to important motion tracking applications like patient rehabilitation, where precision and wearability are critical.
To overcome these barriers, researchers from the USC Viterbi School of Engineering’s Analog/RF Integrated Circuits, Microsystems, and Electromagnetics (ACME) Lab have developed a solution using magnetic induction (MI).
“This is the first wireless, wearable, and real-time motion-tracking network based on magnetic induction,” explained Michella Rustom, Ph.D. candidate in the Ming Hsieh Department of Electrical and Computer Engineering and first author of the paper, “Frequency-Division Multiplexed Magnetic Induction Based Wireless Wearable Sensor Network for Real-Time Motion Tracking.”
“We presented a new localization scheme, which can accurately track motion by measuring the changes in magnetic coupling between pairs of wearable sensors at a high frame rate,” said Rustom, an international student from Lebanon who joined USC in 2019 as an Annenberg fellow. Her research focuses on analog integrated circuits for biomedical applications, from ingestible to wearable technology.
This work was funded by an Office of Naval Research (ONR) Young Investigator Program (YIP) award and led by Constantine Sideris, USC Viterbi associate professor of electrical and computer engineering and director of the ACME lab.
“Accurate, real-time, human motion tracking that can be used outside of a lab or studio environment can enable so many exciting applications,” explained Sideris. “However, wearable solutions based on inertial measurement fall short due to being relative positioning systems, inferring position based on changes in acceleration and rotation, which causes errors to accumulate. Our approach based on magnetic induction is an absolute positioning system, analogous to GPS, and therefore enables a drift-free motion-tracking solution that does not require calibration.”
This tracking system uses a network of lightweight sensors, each featuring the custom-designed MI transceiver chip. By measuring the changes in magnetic coupling between each pair of sensors, the position and orientation of the wearer’s limbs can be dynamically tracked. Unlike camera-based solutions, the sensors can function in any environment without line-of-sight restrictions and do not suffer from long-term drift. Compact, low power and low-cost, the system is practical and reliable for tracking all kinds of motions across a wide range of environments or conditions in the real world.
Rustom designed and tested her custom microchip and experimentally demonstrated joint flexion and extension on both a 3D-printed double-joint leg model and a human volunteer.
In June, she presented her motion tracking sensor network and gave a live demo at the IEEE Symposium on VLSI Technology and Circuits in Kyoto, Japan, one of the leading international conferences on semiconductor technology and circuits. In addition to Rustom and Sideris, the paper is co-authored by collaborators Thanh Dat Nguyen, Alireza Farhadian Bouroujerdi and Mahta Moghaddam from the MiXiL lab.
Michella Rustom and Constantine Sideris presenting their motion tracking sensor network at the 2025 VLSI demo session in Kyoto, Japan. (Photo Credit: Michella Rustom)
“From concept to full system-level design and validation, bringing this work to life has been one of the most rewarding milestones of my PhD,” Rustom said. “It reflects my passion for chip design and my commitment to advancing wearable technologies that can transform human lives.”
This new wearable technology can open countless new possibilities across industries. In health care, physical therapists could monitor patient recovery remotely with lab-level accuracy, reducing the need for frequent clinic visits. In sports, athletes could analyze performance in real time without intrusive hardware or frequent recalibration. In the virtual reality world, the wearable sensors could enable smoother, more immersive experiences by accurately tracking full-body movements. From clinics and athletic fields to living rooms, USC ACME Lab’s innovative sensor is poised to revolutionize motion tracking.
Future next-generation version motion tracker module (Photo Credit: Michella Rustom)
The motion tracker chip-on-board (Photo Credit: Michella Rustom)
Published on October 27th, 2025
Last updated on October 27th, 2025
