Hack your Research! @ IPC 2026
By Marco
Event Date: Sunday, 8 November 2026, time to be announced
Event Location: Denver, Colorado, USA (room to be announced)
Link: ieee-ipc.org/program/sunday-program/
Call for Demos:
Presenter submissions are open through 30 September 2026. Submit a proposal.
Organizers:
Besma Kalla, Eindhoven University of Technology, Netherlands
Advisory Committee:
Rekha Yadav, University College London, United Kingdom
Vincent van Vliet, Eindhoven University of Technology, Netherlands
Amol Delmade, Dublin City University, Ireland
Marco Eppenberger, PsiQuantum, USA
Giammarco Di Sciullo, Università degli Studi dell’Aquila, Italy
Demos:
A Modular Approach to Faster and High Resolution Transmission Measurement
by Pranat Kumar Panda, Indian Institute of Technology, Madras, India
A common method for measuring the transmission of photonic integrated circuit (PIC) components involves using a continuously tunable laser source, data acquisition (DAQ) system, and a photodetector. An effective approach is to use the laser output trigger, which contains wavelength information during the sweep, as a clock source for the DAQ. By sampling the DAQ once per trigger, we limit the wavelength resolution to the step size of the trigger. In this setup, I propose that we sample both the photodetector output and the trigger as analog inputs. This allows us to interpolate between wavelengths, which can enhance the wavelength resolution. This improvement is particularly important for devices under test (DUTs) with significant spectral features, such as high-Q ring resonators.
Information-Driven Design of Photonic Systems Using IDEAL
by Leyla Kabuli, University of California, Berkeley
Computational imaging systems combine optical hardware with reconstruction algorithms to push the limits of image capture and interpretation. In these systems, measurements are designed for algorithms rather than direct human viewing, so a measurement that looks better is not necessarily more useful. We introduce EncodingInformation, an open-source Python/JAX toolkit for quantifying the information captured by a photonic system’s measurements. From just a measurement dataset and a detection noise model, we estimate mutual information, allowing for optical hardware comparison without reconstruction algorithms, ground truth images, or a forward model. We then demonstrate IDEAL, Information-Driven Encoder Analysis Learning, which uses differentiable design methods to automatically optimize optical hardware. These techniques can guide the design of optical elements and illumination patterns, including phase masks and metasurfaces.
SiEPIC openEBL: Remote and Free Silicon Photonic Design-Fabrication-Testing Enabled by Open-Source Process Design Kits and EDA Ecosystem
by Mustafa Hammood and Sujit Ramachandra, SiEPIC Foundation
SiEPIC connects silicon photonic design, simulation, fabrication, testing, and analysis through an open-source EDA ecosystem. This interactive demo follows a photonic circuit from layout to measured results using SiEPIC-Tools and the SiEPIC EBeam PDK. We demonstrate automated layout verification, GDS-to-FDTD simulation through gds_fdtd with the open-source BeamZ solver, and automated measurement analysis. SiEPIC Ecosystem incorporates connections to gdsfactory, Luceda IPKISS, Tidy3D, and Ansys Lumerical, enabling researchers to combine commercial and open-source design and simulation tools. Through SiEPIC openEBL, designers worldwide can access free fabrication and testing remotely, turning designs into fabricated, measured devices without operating a cleanroom or photonics test station. Together, these resources aim to lower the barriers of entry to silicon photonics and further advance the ecosystem for open-source EDA. Attendees will explore practical examples, inspect measured results, and discover courses, workshops, and tutorials to begin designing and testing their own photonic circuits. Learn more about SiEPIC openEBL: https://siepic.ca/openebl/ Learn more about SiEPIC EDA ecosystem: https://github.com/siepic Reference: https://doi.org/10.1117/12.307681
OptiQuantum: Quantum Modeling from Classical Measurements
by Richard Chung, Colorado State University
Quantum circuits and devices are difficult to efficiently model on conventional computers. Most models additionally require postprocessing to generate a useful metric. By approximating a circuit as frequency-independent over a photon spectrum and preselecting the metrics of interest, we are able to efficiently model quantum effects from a classical transfer matrix. In this demo, we present a simulation of Hong-Ou-Mandel interference, quantum circuit fidelity, and two-photon transmittance in Mach-Zehnder interferometer meshes.
Tx or Rx? Finding I/Q Skew from Data Alone
by Shiyi Xia, Nokia Bell Labs
Is the I/Q skew in your coherent link coming from the transmitter or the receiver? Sub-picosecond skew is easy to miss: its image is odd in frequency, so band-averaged imbalance estimates read it as zero. Our MATLAB tool measures it from ordinary data captures, with no test tones or extra hardware. It uses the bench’s natural polarisation and phase drift as a lever: receiver imbalance stays fixed to the ADC axes while transmitter imbalance rotates with the channel, so one frequency-resolved fit separates the two. In practice: capture a batch, run one script, and read each side’s skew and gain, plus RF reflections that show up as residual ripple.
Make Your Figures Look Sharp! Using TikZ for Making Figures and Animations
by Kadir Gümüş, Independent Researcher
In this demo I will show how you can make vector images and animations using TikZ to spruce up your papers and presentations. Catch the eye of the reviewers and the conference attendees with your shiny new figures.
Emulating Dynamic Network Traffic in Optical Transmission
by Lucas Alves Zischler, University of L’Aquila
Reproducing the real-time traffic dynamics of deployed optical networks in the laboratory is nontrivial. This can, however, be emulated by dynamically allocating Gaussian-noise channels. In this demonstration, I show how a noise-loading stage and wavelength blockers can emulate dynamic classical traffic following Erlang statistics. I further demonstrate, as an example, its application to a field-deployed experiment with bidirectional traffic over an uncoupled multicore fiber.