Aktuelle Publikationen

Explainability is supposed to strengthen trust in artificial intelligence, it is necessary to systematically review and compare explanation methods in order to confirm their correctness. We therefore built Quantus—a comprehensive, evaluation...

We numerically show modulation format identification in the optical domain using Silicon-on-Insulator-based Photonic-Integrated-Circuit (PIC) reservoir. Identification of 32 GBd single-polarization signals of OOK, PAM4, BPSK and QPSK is...

We compare three different neural network architectures for nonlinearity mitigation of 32 GBd OOK and QPSK signals after transmission over a dispersion-compensated link of 10-km SSMF and 10-km DCF. OSNR gains up to 2.2 dB were achieved using...

We experimentally study waveband-shift-free optical phase conjugation (OPC) of a 400-Gb/s dual-carrier superchannel and discuss the effectiveness of the OPC for multi-channel nonlinear mitigation. After joint reception of the 31.44 GBd-PDM-16QAM...

We experimentally present a low-complexity dispersion-engineered all-optical wavelength-converter using a photonic integrated-circuit based on SOI waveguide. We achieve a single-sided conversion bandwidth of ~35 nm from C- to S-band, and...

We demonstrate the transmission of constant-envelope orthogonal frequency division multiplexing (CE-OFDM) signals, based on electrical phase modulation, in a visible light communication (VLC) system. An increased tolerance to nonlinearity...

We propose a novel approach to perform QoT estimation relying on joint exploitation of machine learning and analytical formula that offers accurate estimation when applied to scenarios with heterogeneous span profiles and sparsely occupied links....

We demonstrate sophisticated laboratory automation and data pipeline capable of generating large, diverse, and high-quality public datasets. The demo covers the full workflow from setup reconfiguration to data monitoring and storage, represented...

We showcase a proof-of-concept demonstration of a ML pipeline for real-time traffic forecasting deployed on a passive optical access network using an XGS-PON compatible telemetry framework. The demonstration reveals the benefits of fine-granular...

We demonstrate on-off-keying optical wireless transmissions according to the IEEE P802.15.13 PM-PHY, in which we replace 8b10b line-coding by guided and non-guided data scramblers and compensate the remaining high-pass distortions through a...

We evaluate a high power on-off-keying transmitter for non-directed optical wireless communications based on VCSEL-arrays. Error-free transmission after FEC with a net data rate of 1.75 GBit/s is achieved across a distance of 2.5 m with a...

This paper presents optimization and deep learning procedures aiming at increasing power and spectral efficiency of visible light communications systems for two exemplary industrial scenarios. We propose a hybrid multi-objective optimization to...

Multimodal user interfaces can provide better solutions, as they combine various interaction modalities to enable more flexibility and naturalness in human-machine interactions (HMI). They allow to seamlessly adapt to user and application...

In this chapter we discuss the algorithmic and theoretical underpinnings of layer-wise relevance propagation (LRP), apply the method to a complex model trained for the task of visual question answering (VQA), and demonstrate that it produces...

Explainable Artificial Intelligence (XAI) is an emerging research field bringing transparency to highly complex and opaque machine learning (ML) models. This paper offers a comprehensive overview over techniques that apply XAI practically to...

Analog Multiplexers in Silicon-Germanium technology allow increasing the analog bandwidth of Digital-to-Analog-Converters and enable faster data transfers. We demonstrated 4-level pulse-amplitude modulation (PAM-4) signal transmission at 150 GBd...

Brain-age (BA) estimates based on deep learning are increasingly used as neuroimaging biomarker for brain health; however, the underlying neural features have remained unclear. We combined ensembles of convolutional neural networks with...

Histological sections of the lymphatic system are usually the basis of static (2D) morphological investigations. Here, we performed a dynamic (4D) analysis of human reactive lymphoid tissue using confocal fluorescent laser microscopy in...

We investigate the applicability of PAS for outdoor THz wireless links in simulations with weather-dependent loss models. Link performances are evaluated and optimal shaping entropies are determined to adjust error rates to a given FEC threshold....

We present a remote quality assurance use-case in distributed production sites that can be realized with the powerful capabilities of Artificial Intelligence (AI) combined with real-time video streaming systems and high-speed, low-latency...

Ultra-Broadband Wavelength Conversion is one of the key issues of future optical networks. The physical background of ultra-broadband optical wavelength conversion in a multi-modal Silicon waveguide and methods to optimize its functionality are...

One of the overarching goals of the ETSI Industry Specification Group (ISG) F5G on Fifth Generation Fixed Network is to establish a regular rhythm of evolution for the fixed telecommunications network. So far, F5G has published technical...

Medical and dental artificial intelligence (AI) require the trust of both users and recipients of the AI to enhance implementation, acceptability, reach, and maintenance. Standardization is one strategy to generate such trust, with quality...

Building performant and robust artificial intelligence (AI)–based applications for dentistry requires large and high-quality data sets. Collaborative efforts are limited as privacy constraints forbid direct sharing across the borders of these...

Network canonization has recently been introduced, restructuring a neural network model into a functionally identical equivalent to which established explanation methods can be applied optimally. In this work, we quantitatively verify the...

We demonstrate the generation, of a mmWave signal via the injection of an optical frequency comb (OFC) into an integrated tunable dual distributed Bragg reflector (DBR) laser as well as the fiber transmission and the processing of this signal by...

In this work, we aim to understand what type of pretraining works reliably in practice and what type of pretraining dataset is best suited to achieve good performance in small target dataset size scenarios. Considering diabetic retinopathy...

This paper presents an improved probability estimation scheme for the entropy coder of Incremental Neural Network Coding (INNC), which is currently under standardization in ISO/IEC MPEG. Major finding is that the probability of a significant...

In the evolution of optical networks, enhancement of spectral efficiency (SE) enhancement has been the most cost-efficient and thus the main driver for capacity enhancementincrease for decades. As a result, the development of optical transport...

This work proposes a parallel neural network based demosaicing procedure for single-camera-one-shot-for hyperspectral imaging trained on a new ground truth dataset captured in a controlled environment by a hyperspectral snapshot camera with a 4×4...