We’re excited to share a new milestone from our research team: our latest work has just been published in the IEEE Sensors Journal, one of the leading venues for sensing technologies.
The paper is titled:
“Speckle Analysis in Multimode Optical Fibers for Chemical and Physical Sensing: A Comparative Study of Demodulation Algorithms.”
This study has been part of a long-term effort within our group to understand how speckle patterns—those seemingly random intensity patterns produced by multimode fibers—can be transformed into powerful sensing signals.
Why We Took on This Study
Over the past few years, fiber-optic sensing based on speckle analysis has gained momentum. Yet, a major challenge remained: different research groups used different demodulation algorithms, making it hard to compare performance or choose the right approach for specific applications.
We wanted to change that.
Our goal was simple:
to evaluate the most widely used speckle demodulation algorithms under identical conditions and offer the field a clear, unified comparison.
What the Paper Reveals
Using a polymer multimode fiber as our sensing platform, we examined three main algorithms:
-
Average Intensity (AIA)
-
Normalized Inner Product (NIPC)
-
Sum of Squared Differences (SSD)
Each algorithm was tested using the same optical setup, the same stimuli, and the same environmental conditions — allowing, for the first time, a direct and fair comparison.
Here are some of the key insights:
-
AIA stood out with the widest sensing range and the fastest computation times, making it an attractive choice for real-time applications.
-
Through Region-of-Interest (ROI) analysis, we were able to tune the sensing sensitivity by more than 200%, simply by selecting where we look inside the speckle pattern.
-
The ROI study also uncovered directional speckle motion, offering new clues about how different types of stimuli affect multimode fiber speckle fields.
-
Overall, the results form a decision-making framework for choosing the right algorithm depending on range, sensitivity and computational needs.
Why This Work Matters
Beyond the technical details, this study is meaningful for us because it reflects the type of research we value:
clear, comparative, and directly useful to the wider sensing community.
We hope this publication helps researchers and engineers make more informed choices when designing speckle-based sensing systems — whether for chemical monitoring, mechanical sensing, biomedical applications, or emerging smart-material platforms.
Read the Full Paper
If you’re interested in diving deeper, you can access the work here:
-
Open-access version: https://zenodo.org/records/16162480