Skip to main content

TRANSNET at OECC 2025

From June 29th to July 3rd, TRANSNET colleagues and friends will be travelling to Sapporo, Japan, to present their work at the 30th annual OptoElectronics and Communications Conference (OECC 2025). OECC is a leading conference for the fields of optoelectronics, optical fibre transmission, and photonic network systems, held jointly with the International Conference on Photonics in Switching and Computing (PSC).


Dr Aleksandr Donodin, a TRANSNET researcher based in Aston, is an invited speaker at this year’s OECC! He will be presenting “Doped Fiber Amplifiers for Multi-band Transmission”. Friend of TRANSNET and UCL team alum, Domanic Lavery (Nokia), is also an invited speaker, presenting “State of the Art Real-time DSP for Long-haul Transmission Systems”. 

Dr Eric Sillekens, TRANSNET research fellow, will be speaking in the workshop “Can SDM Survive the Hollow Core Challenge?”. Eric is a RAEng research fellow, investigating energy reduction for optical transmissions systems by introducing low-complexity digital signal processing (DSP), enabling the combined use of SOAs and hollow core optical fibre (HCF) - you can read more about his research in this article. Dr Yuta Wakayama (KDDI), friend and frequent collaborator of TRANSNET, will also be speaking in this workshop.

 

Oral and Poster Presentations

Extending the Closed Form Approximation of the ISRS GN model in the Zero-Dispersion Regime for Arbitrary Modulation Formats, Span Length, and Fiber Loss

Filippos Balasis, Mindaugas Jarmolovičius, Daniel J. Elson, Eric Sillekens, Han Wang, Robert I. Killey, Polina Bayvel, Noboru Yoshikane, Takehiro Tsuritani, Yuta Wakayama

In a collaboration between KDDI and TRANSNET, the benefits of using O-band transmission over conventional systems are acknowledged alongside the importance of estimating non-linear interference, which can be much stronger in O-band. The authors provide a closed form approximation of the four-wave mixing interference in O-band accounting for arbitrary modulation format. The approximation also accounts for span length (accuracy for short span links is an important feature with applications to existing infrastructure such as inter-data centre links) and fibre loss (ultra-low loss is an expected feature of future hollow core fibers). 

 

Algorithm Finding GMI-Optimal Binary Labelings for BICM

Domaniç Lavery, Alex Alvarado 

In optical fiber communication systems, the optimization of coding and modulation can maximise the achievable information rate. A commonly used achievable rate is the generalized mutual information (GMI). For any given modulation format, the binary labelling of a M-ary constellation has a substantial impact on GMI. For complex constellations, figuring out the best labelling is challenging; the number of possible labellings grows rapidly with M, making it impractical to test them all. The paper presents an algorithm for optimal binary labelling of constellations with respect to GMI, and assesses its computational complexity for a selection of geometrically shaped constellations. This algorithm is the first means of computation of optimal labellings for arbitrary constellations of order M ≤ 16.

 

Line-by-Line Comb Noise Measurements Using an Electro-Optic Comb Local Oscillator

Alex Bennett, Zun Htay, Florian Emaury, Andrea Pertoldi, Benjamin Rudin, Zhixin Liu

UCL PhD student and friend of TRANSNET, Alex Bennett, will be presenting his work on line-by-line phase noise characterisation for a 2.5GHz repetition-rate frequency comb at the OECC poster sessions. Low-noise optical frequency combs are promising light sources for optical and radio wireless communications, with low phase noise tones a primary requirement in applications. While previous studies have investigated the phase noise of the one comb tone in the centre of the spectrum, the evolvement of phase noise across a wide spectral range has not been systematically studied. Alex presents the first report of line-by-line measurement of a solid-state comb of its kind.

 


 

OECC/PSC 2025 Website