no code implementations • 5 Mar 2024 • Charlotte Muth, Benedikt Geiger, Daniel Gil Gaviria, Laurent Schmalen
To enhance the training behavior, we decouple the loss functions from the respective SNR values and the number of sensing snapshots, using bounds of the sensing performance.
no code implementations • 23 Feb 2024 • Jonas Ney, Patrick Matalla, Vincent Lauinger, Laurent Schmalen, Sebastian Randel, Norbert Wehn
In this work, we present a high-throughput field programmable gate array (FPGA) demonstrator of an artificial neural network (ANN)-based equalizer.
no code implementations • 23 Jan 2024 • Luca Schmid, Tomer Raviv, Nir Shlezinger, Laurent Schmalen
We investigate the application of the factor graph framework for blind joint channel estimation and symbol detection on time-variant linear inter-symbol interference channels.
no code implementations • 17 Jan 2024 • Vincent Lauinger, Patrick Matalla, Jonas Ney, Norbert Wehn, Sebastian Randel, Laurent Schmalen
We demonstrate and evaluate a fully-blind digital signal processing (DSP) chain for 100G passive optical networks (PONs), and analyze different equalizer topologies based on neural networks with low hardware complexity.
no code implementations • 26 Dec 2023 • Jinxiang Song, Vincent Lauinger, Christian Häger, Jochen Schröder, Alexandre Graell i Amat, Laurent Schmalen, Henk Wymeersch
We propose a novel frequency-domain blind equalization scheme for coherent optical communications.
1 code implementation • 20 Dec 2023 • Alexander von Bank, Eike-Manuel Edelmann, Laurent Schmalen
We propose an energy-efficient equalizer for IM/DD systems based on spiking neural networks.
no code implementations • 8 Dec 2023 • Daniel Gil Gaviria, Marcus Müller, Felix Artmann, Laurent Schmalen
The use of modern software-defined radio (SDR) devices enables the implementation of efficient communication systems in numerous scenarios.
no code implementations • 2 Jun 2023 • Luca Schmid, Joshua Brenk, Laurent Schmalen
Message passing on factor graphs is a powerful framework for probabilistic inference, which finds important applications in various scientific domains.
no code implementations • 24 May 2023 • Lucas Giroto de Oliveira, David Brunner, Axel Diewald, Charlotte Muth, Laurent Schmalen, Thomas Zwick, Benjamin Nuss
This article introduces a bistatic joint radar-communication (RadCom) system based on orthogonal frequency-division multiplexing (OFDM).
1 code implementation • 27 Apr 2023 • Alexander von Bank, Eike-Manuel Edelmann, Laurent Schmalen
A spiking neural network (SNN) equalizer with a decision feedback structure is applied to an IM/DD link with various parameters.
no code implementations • 14 Apr 2023 • Jonas Ney, Vincent Lauinger, Laurent Schmalen, Norbert Wehn
In recent years, communication engineers put strong emphasis on artificial neural network (ANN)-based algorithms with the aim of increasing the flexibility and autonomy of the system and its components.
no code implementations • 22 Feb 2023 • Jinxiang Song, Vincent Lauinger, Yibo Wu, Christian Häger, Jochen Schröder, Alexandre Graell i Amat, Laurent Schmalen, Henk Wymeersch
Furthermore, we show that for the linear channel, the proposed scheme exhibits better convergence properties than the \ac{MMSE}-based, the \ac{CMA}-based, and the \ac{VAE}-based equalizers in terms of both convergence speed and robustness to variations in training batch size and learning rate.
1 code implementation • 23 Jan 2023 • Charlotte Muth, Laurent Schmalen
Foremost, we develop a suitable encoding scheme for the training of the AE and for targeting a fixed false alarm rate of the target detection during training.
no code implementations • 16 Jan 2023 • Vincent Lauinger, Fred Buchali, Laurent Schmalen
We evaluate the start-up of blind equalizers at critical working points, analyze the advantages and obstacles of commonly-used algorithms, and demonstrate how the recently-proposed variational autoencoder (VAE) based equalizers can improve bootstrapping.
1 code implementation • 7 Dec 2022 • Andrej Rode, Benedikt Geiger, Shrinivas Chimmalgi, Laurent Schmalen
As the demand for higher data throughput in coherent optical communication systems increases, we need to find ways to increase capacity in existing and future optical communication links.
no code implementations • 5 Dec 2022 • Andrej Rode, Laurent Schmalen
We show that separating the in-phase and quadrature component in optimized, machine-learning based demappers of optical communications systems with geometric constellation shaping reduces the required computational complexity whilst retaining their good performance.
1 code implementation • 21 Nov 2022 • Lukas Rapp, Luca Schmid, Andrej Rode, Laurent Schmalen
We propose a novel method to optimize the structure of factor graphs for graph-based inference.
no code implementations • 9 Nov 2022 • Eike-Manuel Bansbach, Alexander von Bank, Laurent Schmalen
In the past years, artificial neural networks (ANNs) have become the de-facto standard to solve tasks in communications engineering that are difficult to solve with traditional methods.
no code implementations • 15 Sep 2022 • Vincent Lauinger, Manuel Hoffmann, Jonas Ney, Norbert Wehn, Laurent Schmalen
The proposed approach is independent of the equalizer topology and enables the application of powerful neural network based equalizers.
no code implementations • 31 Aug 2022 • Eike-Manuel Bansbach, Yigit Kiyak, Laurent Schmalen
For orthogonal multiple access (OMA) systems, the number of served user equipments (UEs) is limited to the number of available orthogonal resources.
no code implementations • 24 Jun 2022 • Andrej Rode, Laurent Schmalen
We present a novel method to investigate the effects of varying channel parameters on geometrically shaped constellations for communication systems employing the blind phase search algorithm.
1 code implementation • 25 Apr 2022 • Vincent Lauinger, Fred Buchali, Laurent Schmalen
We investigate the potential of adaptive blind equalizers based on variational inference for carrier recovery in optical communications.
1 code implementation • 30 Mar 2022 • Luca Schmid, Laurent Schmalen
In this paper, we develop and evaluate efficient strategies to improve the performance of the factor graph-based symbol detection by means of neural enhancement.
no code implementations • 7 Mar 2022 • Luca Schmid, Laurent Schmalen
We study the application of the factor graph framework for symbol detection on linear inter-symbol interference channels.
no code implementations • 6 Dec 2021 • Andrej Rode, Benedikt Geiger, Laurent Schmalen
We perform geometric constellation shaping with optimized bit labeling using a binary autoencoder including a differential blind phase search (BPS).
no code implementations • 27 Aug 2021 • Eike-Manuel Bansbach, Victor Eliachevitch, Laurent Schmalen
In particular, varying requirements lead to a non-convex optimization problem when maximizing the systems data rate while preserving fairness between UEs.
no code implementations • 18 May 2020 • Boris Karanov, Mathieu Chagnon, Vahid Aref, Domanic Lavery, Polina Bayvel, Laurent Schmalen
We investigate end-to-end optimized optical transmission systems based on feedforward or bidirectional recurrent neural networks (BRNN) and deep learning.
no code implementations • 18 May 2020 • Boris Karanov, Mathieu Chagnon, Vahid Aref, Filipe Ferreira, Domanic Lavery, Polina Bayvel, Laurent Schmalen
The investigation of digital signal processing (DSP) optimized on experimental data is extended to pulse amplitude modulation with receivers performing sliding window sequence estimation using a feed-forward or a recurrent neural network as well as classical nonlinear Volterra equalization.
no code implementations • 21 Jan 2020 • Andreas Buchberger, Christian Häger, Henry D. Pfister, Laurent Schmalen, Alexandre Graell i Amat
In this paper, we introduce a method to tailor an overcomplete parity-check matrix to (neural) BP decoding using machine learning.
no code implementations • 11 Dec 2019 • Boris Karanov, Mathieu Chagnon, Vahid Aref, Domaniç Lavery, Polina Bayvel, Laurent Schmalen
We perform an experimental end-to-end transceiver optimization via deep learning using a generative adversarial network to approximate the test-bed channel.
no code implementations • 2 Oct 2019 • Boris Karanov, Gabriele Liga, Vahid Aref, Domaniç Lavery, Polina Bayvel, Laurent Schmalen
In this paper, we apply deep learning for communication over dispersive channels with power detection, as encountered in low-cost optical intensity modulation/direct detection (IM/DD) links.
Information Theory Signal Processing Information Theory
1 code implementation • 7 Mar 2019 • Ahmed Elkelesh, Moustafa Ebada, Sebastian Cammerer, Laurent Schmalen, Stephan ten Brink
Moreover, GenAlg can be used to design LDPC codes with the aim of reducing decoding latency and complexity, leading to coding gains of up to $0. 325$ dB and $0. 8$ dB at BLER of $10^{-5}$ for both AWGN and Rayleigh fading channels, respectively, when compared to state-of-the-art short LDPC codes.
Information Theory Information Theory
no code implementations • 24 Jan 2019 • Boris Karanov, Domaniç Lavery, Polina Bayvel, Laurent Schmalen
Our novel SBRNN design aims at tailoring the end-to-end deep learning-based systems for communication over nonlinear channels with memory, such as the optical IM/DD fiber channel.
no code implementations • 11 Apr 2018 • Boris Karanov, Mathieu Chagnon, Félix Thouin, Tobias A. Eriksson, Henning Bülow, Domaniç Lavery, Polina Bayvel, Laurent Schmalen
In this paper, we implement an optical fiber communication system as an end-to-end deep neural network, including the complete chain of transmitter, channel model, and receiver.
no code implementations • 2 Apr 2018 • Mohamad Dia, Vahid Aref, Laurent Schmalen
In this work, we formulate the fixed-length distribution matching as a Bayesian inference problem.