Search

We experimentally demonstrate a novel, low-complexity Fourier Convolution-based Network (FConvNet) based equalizer for 112 Gb/s upstream PAM4-PON. At a BER of 0.005, FConvNet enhances the receiver sensitivity by 2 and 1 dB compared to a 51-tap Sato equalizer and benchmark machine learning algorithms respectively., Comment: 4 pages, 5 figures

In recent years, extensive research has been conducted to explore the utilization of machine learning algorithms in various direct-detected and self-coherent short-reach communication applications. These applications encompass a wide range of tasks, including bandwidth request prediction, signal quality monitoring, fault detection, traffic prediction, and digital signal processing (DSP)-based equalization. As a versatile approach, machine learning demonstrates the ability to address stochastic phenomena in optical systems networks where deterministic methods may fall short. However, when it comes to DSP equalization algorithms, their performance improvements are often marginal, and their complexity is prohibitively high, especially in cost-sensitive short-reach communications scenarios such as passive optical networks (PONs). They excel in capturing temporal dependencies, handling irregular or nonlinear patterns effectively, and accommodating variable time intervals. Within this extensive survey, we outline the application of machine learning techniques in short-reach communications, specifically emphasizing their utilization in high-bandwidth demanding PONs. Notably, we introduce a novel taxonomy for time-series methods employed in machine learning signal processing, providing a structured classification framework. Our taxonomy categorizes current time series methods into four distinct groups: traditional methods, Fourier convolution-based methods, transformer-based models, and time-series convolutional networks. Finally, we highlight prospective research directions within this rapidly evolving field and outline specific solutions to mitigate the complexity associated with hardware implementations. We aim to pave the way for more practical and efficient deployment of machine learning approaches in short-reach optical communication systems by addressing complexity concerns., Comment: 23 pages, 2 figure, 3 tables, Accepted as MDPI Photonics Journal Speical Issue Machine Learning Applied to Optical Communication Systems

A frequency-calibrated SCINet (FC-SCINet) equalizer is proposed for down-stream 100G PON with 28.7 dB path loss. At 5 km, FC-SCINet improves the BER by 88.87% compared to FFE and a 3-layer DNN with 10.57% lower complexity., Comment: 3 pages, 6 figures, accepted by Optical Fiber Communications Conference and Exhibition 2024

This paper addresses the transmission-aware transceiver allocation problem of flexible optical networks for a multi-period planning. The proposed approach aims at assigning the best configuration of bandwidth variable transceivers (BVTRX) considering the amplifier noise and nonlinear channel interferences using the incoherent Extended Gaussian Noise (EGN) model. The proposed solution improves the network throughput and spectrum utilization in the early planning periods and allocates lower number of BV-TRXs in later periods in comparison to algorithms presented recently. A heuristic approach to regenerator placement has also been applied achieving up to 25% transceiver and 50% spectrum utilization savings in comparison to configurations without regenerators.

Smart cities will be characterized by a variety of intelligent and networked services, each with specific requirements for the underlying network infrastructure. While smart city architectures and services have been studied extensively, little attention has been paid to the network technology. The KIGLIS research project, consisting of a consortium of companies, universities and research institutions, focuses on artificial intelligence for optimizing fiber-optic networks of a smart city, with a special focus on future mobility applications, such as automated driving. In this paper, we present early results on our process of collecting smart city requirements for communication networks, which will lead towards reference infrastructure and architecture solutions. Finally, we suggest directions in which artificial intelligence will improve smart city networks., Comment: Accepted for publication at ISC2 2021

We propose multi-dimensional short blocklength probabilistic shaping to increase the nonlinear tolerance gain in digital subcarrier multiplexing transmission systems and demonstrate an improvement in performance compared to lower dimensional formats., Comment: Accepted for SPPCom 2021

Due to their compatibility to existing telecom technology, continuous variable (CV) weak coherent state protocols are promising candidates for a broad deployment of quantum key distribution (QKD) technology. We demonstrate how an existing simulation framework for modelling classical optical systems can be utilized for simulations of weak-coherent CV-QKD links. The quantum uncertainties for the measured characteristics of coherent signals are modelled in the electrical domain by shot noise, while a coherent signal in the optical domain is described by its quadrature components. We simulate various degradation effects such as attenuation, laser RIN, Raman noise (from classical channels in the same fibre), and device imperfections and compare the outcome with analytical theory. Having complemented the physical simulation layer by the post-processing layer (reconciliation and privacy amplification), we are able to estimate secure key rates from simulations, greatly boosting the development speed of practical CV-QKD schemes and implementations.

The reconciliation step of continuous-variable quantum key distribution protocols usually involves forward error correction codes. Matching the code rate and the signal-to-noise ratio (SNR) of the quantum channel is required to achieve the high reconciliation efficiencies that are crucial for long distance links. Puncturing and shortening is a way to adapt the code rate to the SNR at the cost of a slightly reduced reconciliation efficiencies. Instead of adapting the code rate to the SNR, we propose to add a controlled amount of artificial noise to the measured data, so that the resulting SNR could be reduced to match the given code rate. We show that our method can compete with puncturing and shortening and even outperform it in high-loss, high-excess noise scenarios.