Hack your Research! @ IPC 2025
By Vincent
Event Date: Sunday, 9 November 2025, 3:30pm – 5:00pm SGT
Event Location: Pisces 2 & 3, Resorts World Convention Centre, Singapore, Singapore
Link: ieee-ipc.org/sunday-program/
Organizers:
Shamsul Arafin, The State Ohio University, Columbus, OH, USA
Rekha Yadav, University College London, United Kingdom
Besma Kalla, Eindhoven University of Technology, Netherlands
Vincent van Vliet, Eindhoven University of Technology, Netherlands
Amol Delmade, Pilot Photonics, Ireland
Giammarco Di Sciullo, Università degli Studi dell’Aquila, Italy
Xuhui Zhou, National Technological University, Singapore
Zhang Yueqian, National Technological University, Singapore
Advisory Committee:
Marco Eppenberger, PsiQuantum, USA
Menno van den Hout, Eindhoven University of Technology, Netherlands
Nicolas Fontaine, Nokia Bell Labs, USA
Roland Ryf, Nokia Bell Labs, USA
Jochen Schroeder, Chalmers University of Technology, Sweden
Demos:
AutoPaper: A Free and Open-Source Toolkit for Streamlining Academic Writing and Referencing
by Mizanur Rahman Howlader, Hajee Mohammad Danesh Science and Technology University, Bangladesh
AutoPaper is a lightweight, Python-based toolkit designed to support early-career researchers and students in organizing their scientific writing workflow. This tool helps automate the structure of research articles (IEEE/Elsevier format), manage citations using BibTeX or Zotero, generate LaTeX-compatible plots, and insert journal-specific templates quickly. The demo will include how AutoPaper can save hours of manual formatting time, reduce referencing errors, and improve the reproducibility of documentation pipelines—all using open-source libraries. The toolkit is especially useful for researchers from under-resourced regions seeking accessible and efficient tools to enhance productivity without relying on expensive software.
FiberGuard: Real-Time Intrusion Detection for Optical Networks
by Petr Münster, Brno University of Technology, Czech Republic
The live demonstration showcases the core functionality of the Fiber Guard system using a simplified laboratory setup. A laser source injects continuous-wave light into a short section of single-mode fiber that is directly connected to a polarization analyzer, consisting of a polarization beam splitter and a balanced photodetector. External mechanical vibrations or touches applied to the fiber immediately alter the state of polarization of the transmitted light. These variations are converted into electrical signals and processed in real time by the Fiber Guard software.
Interactive Polarization Control: Polarization Controller and Its Digital Twin in the Brillouin Optical Time-Domain Analyzer
by Raphael Zhizhi Yang, Gent University, Belgium
In this demo, we present an interactive setup for polarization control in Brillouin Optical Time-Domain Analysis (BOTDA), combining hardware and software to demonstrate real-time manipulation of polarization states. The core of the demo is a programmable polarization controller integrated with a digital twin environment that simulates its behaviour and interaction with the BOTDA system, illustrating how polarization affects Brillouin gain measurements.
The book that changed my research writing experience: Trees, Maps and Theorems
by Bernat Molero Agudo, Eindhoven University of Technology, the Netherlands
The book Trees, Maps, and Theorems by Jean-Luc Doumont teaches a way to communicate research through basic language and structured reasoning. This book is worth sharing, and that’s why I wanted to share it with you. The author provided me with a discount code for you to get a physical version more easily.
Budget-Friendly Lab Automation: Python-Powered Solutions for Expensive Optical Equipment
by Joydip Dutta & Jonaq Niveer Sarma, Indian Institute of Technology Madras, India
Commercial automation solutions for fiber optic laboratories are often priced prohibitively high, restricting experimental flexibility and access. This set of three demonstrations highlights how Python and Raspberry Pi hardware can be leveraged to automate essential tasks, offering versatile, low-cost alternatives to expensive proprietary systems. 1. Automated Polarization Controller Using Raspberry Pi: Commercial polarization controllers are rarely offered in the desired form factor, and automation modules are costly and hard to modify. We present a system where a Raspberry Pi controls servo motors attached to custom 3D-printed paddles, automating polarization alignment through a Python script for highly flexible mode control. 2. Multi-Channel OTDR Monitoring with NX1 Optical Switch: Most commercial OTDRs are either non-programmable or offer expensive automation options. Our low-cost monitoring setup uses a Raspberry Pi to control a standard OTDR via VNC. Python (with pyautogui) automates the data collection, and traces are transferred via FTP to Google Sheets for seamless multi-channel reporting. By employing an NX1 optical switch, multiple aerial and underground fiber links can be monitored on a schedule—completely bypassing proprietary vendor solutions. 3. MCF Splicing: Python-Powered Solution for Easy Core Alignment: High-end fusion splicers can mechanically rotate fibers for PM or MCF splicing but usually lack the capability for aligning multi-core markers, which is vital to avoid core swapping. We integrate a Python-based image processing pipeline on Raspberry Pi to detect markers and guide splicer alignment for MCF tasks, enabling orientation-critical splicing, as well as warning against incompatible fibers. The system demonstrates that open-source software and readily available hardware can augment advanced fiber optic lab automation. Each system demonstrates that open-source software and readily available hardware can democratize advanced fiber optic lab automation, matching or exceeding the adaptability of expensive industry offerings.
Fourier Minds for Light: Automating Photonic Splitter Design with Neural Operators
by Aggraj Gupta, Indian Institute of Technology Madras, India
Designing photonic splitters traditionally requires time-consuming parameter sweeps and expert intuition. In this demo, we present an AI-driven framework that leverages the Fourier Neural Operator (FNO) to automate and accelerate the inverse design of photonic splitters. By learning the operator mapping between geometric design spaces and optical field responses, our model predicts optimal device topologies in milliseconds, bypassing iterative electromagnetic simulations. We demonstrate how the trained FNO can generalize across different wavelength bands, refractive index contrasts, and splitter ratios, enabling on-the-fly photonic layout generation. The demo concludes with a real-time design session, where a desired power-split ratio is input and the resulting structure and its field propagation pattern are instantly visualized. This approach paves the way toward self-driving photonics, where deep operator learning replaces manual design cycles with automated, data-efficient intelligence.
SimPhotonicsFMM: A Fast and Efficient Fourier Modal Method MATLAB Toolbox for Photonic Structure Simulation
by Ram Prakash Shanmugam, Laboratoire Charles Fabry, Institut d’Optique Graduate School, CNRS, France Software tools dedicated to the simulation of photonic micro- and nanostructures in periodic arrays—whether open-source or proprietary—address a wide range of needs in both academia and industry. The choice among these tools generally depends on several key factors, including computational speed and ease of use. These two criteria lie at the heart of the development of SimPhotonicsFMM, a MATLAB toolbox based on the Fourier Modal Method (also known as Rigorous Coupled Wave Analysis). SimPhotonicsFMM enables the simulation of micro- and nanophotonic structures, including multilayer systems, one-dimensional (1D) and two-dimensional (2D) metamaterials, and periodic metasurfaces. To significantly reduce computation time without compromising accuracy, the toolbox employs a numerically efficient implementation of the differential method using a centered finite-difference scheme. This approach achieves a gain in CPU time exceeding one order of magnitude for large numbers of Fourier terms compared to conventional FMM implementations. For geometric modeling, SimPhotonicsFMM integrates the Gielis superformula, allowing the generation of highly diverse and complex nanoparticle shapes using only six parameters. This compact and standardized representation simplifies optimization processes and facilitates integration with deep-learning-based inverse design frameworks. In fact, it is possible to generate varied and complex shapes, without having to discretize the network cell, and in which a potentially different refractive index is associated with each pixel. SimPhotonicsFMM is developed at the Laboratoire Charles Fabry, Institut d’Optique Graduate School, by Dr.Mondher Besbes.
GDSFactory: Automated chip design and layout in python
by Troy Tamas, GDSFactory, USA
We will demonstrate how the open source GDSFactory can be used to accelerate your workflows, from design and simulation to layout and verification, and how keeping your workflow defined in code enables you to better track, manage, and scale it over time.