Aktuelle Publikationen

September 2021

Interconnection challenges on integrated terahertz photonic systems

Guillermo Carpintero, Norbert Keil, David DeFelipe, Björn Globisch, Lars Liebermeister, Muhsin Ali, Luis Enrique Garcia-Munoz, Sebastian Lauck, Michael Nagel, Alejandro Rivera-Lavado, Daniel Gallego, Dmitry Lyubchenko, Nikolaos Xenidis, Enrique Prados-Castro, Jose Maria Pindado-Buendia, Riccardo Rossetti

Current challenges for RF interconnects are presented, esp. for calibrated measures of the frequency response of components operating > 100 GHz. Photonics and electronics are combined to develop new paradigm in the millimetre and Terahertz...

September 2021

Curiously Effective Features For Image Quality Prediction

Sören Becker, Thomas Wiegand, Sebastian Bosse

We find feature extractors constructed from random noise to be sufficient to learn a linear regression model whose quality predictions reach high correlations with human visual quality ratings, on par with a model with learned features.

September 2021

An Efficient Multi-Link Channel Model for LiFi

Sreelal Maravanchery Mana, Volker Jungnickel, Jonas Hilt, Peter Hellwig, Sepideh Mohammadi Kouhini, Kerolos Gabra Kamel Gabra

The emergence of LiFi for indoor communications opens up new possibilities for wireless services in crowded multiuser scenarios. The deployment of LiFi in indoor scenarios is challenging due to the line-of-sight (LOS) blockage as well as...

August 2021

FantastIC4: A Hardware-Software Co-Design Approach for Efficiently Running 4bit-Compact Multilayer Perceptrons

Simon Wiedemann, Thomas Wiegand, Wojciech Samek, Friedel Gerfers, Daniel Becking, Suhas Shivapakashy, Pablo Wiedemann

With the growing demand for deploying Deep Learning models to the “edge”, it is paramount to develop techniques that allow to execute models within very tight and limited resource constraints. In this work we propose a software-hardware...

August 2021

Zero on Shape: A Generic 2D-3D Instance Similarity Metric learned from Synthetic Data

Maciej Janik, Peter Eisert, Anna Hilsmann, Niklas Gard

We present a network architecture which compares RGB images and untextured 3D models by the similarity of the represented shape. Our system is optimised for Zero-Shot retrieval, meaning it can recognise shapes never shown in training.

August 2021

Enabling S-C-L-Band Systems with Standard C-Band Modulator and Coherent Receiver using Nonlinear Predistortion

Robert Emmerich, Colja Schubert, Carsten Schmidt-Langhorst, Ronald Freund, Robert Elschner, Isaac Sackey, Matheus Ribeiro Sena

To counteract the forthcoming capacity crunch in optical networks by increasing the throughput over already existing fiber infrastructures, innovative approaches are required. Multiband systems are an interesting approach to address this issue....

August 2021

S-Band Transmission with Off-the-Shelf C-Band Components Enabled by Nonlinear Predistortion based on Coherent System Identification

Robert Emmerich, Colja Schubert, Carsten Schmidt-Langhorst, Ronald Freund, Robert Elschner, Isaac Sackey, Mustafa Sezer Erkilinc, Matheus Ribeiro Sena

Nonlinear Predistortion based on Coherent System Identification Text: In order to cope with the rapid traffic growth and as well reducing the cost-per-bit, reuse of the legacy optical fiber infrastructure is one of the main objectives for...

August 2021

Clustered Federated Learning: Model-Agnostic Distributed Multi-Task Optimization under Privacy Constraints

Felix Sattler, Klaus-Robert Müller, Wojciech Samek

Federated learning (FL) is the most widely adopted framework for collaborative training of deep learning models under privacy constraints. Albeit its popularity, it has been observed that FL yields suboptimal results if the clients’ data...

August 2021

Explain and Improve: LRP-Inference Fine Tuning for Image Captioning Models

Jiamei Sun, Wojciech Samek, Alexander Binder, Sebastian Lapuschkin

This paper analyzes the predictions of image captioning models with attention mechanisms beyond visualizing the attention itself. We compare the interpretability of attention heatmaps systematically against the explanations. We demonstrate that...

August 2021

Deep Convolutional Neural Network for Network-wide QoT Estimation

Pooyan Safari, Johannes Fischer, Mohammad Behnam Shariati, Geronimo Bergk

We propose a novel Deep Convolutional Neural Network formulation for network-wide QoT classification tasks and show its effectiveness for networks with significant topological differences. Our formulation achieves ~99% accuracy on large and...

Ergebnisse pro Seite10ǀ20ǀ30
Ergebnisse 11-20 von 86
<< < 1 2 3 4 5 6 7 8 > >>