Stephan ten Brink

Jakob Hoydis, Fayçal Aït Aoudia, Sebastian Cammerer et al.

Ray tracing (RT) is instrumental in 6G research in order to generate spatially-consistent and environment-specific channel impulse responses (CIRs). While acquiring accurate scene geometries is now relatively straightforward, determining material characteristics requires precise calibration using channel measurements. We therefore introduce a novel gradient-based calibration method, complemented by differentiable parametrizations of material properties, scattering and antenna patterns. Our method seamlessly integrates with differentiable ray tracers that enable the computation of derivatives of CIRs with respect to these parameters. Essentially, we approach field computation as a large computational graph wherein parameters are trainable akin to weights of a neural network (NN). We have validated our method using both synthetic data and real-world indoor channel measurements, employing a distributed multiple-input multiple-output (MIMO) channel sounder.

Marvin Geiselhart, Ahmed Elkelesh, Jannis Clausius et al.

Finding optimal message quantization is a key requirement for low complexity belief propagation (BP) decoding. To this end, we propose a floating-point surrogate model that imitates quantization effects as additions of uniform noise, whose amplitudes are trainable variables. We verify that the surrogate model closely matches the behavior of a fixed-point implementation and propose a hand-crafted loss function to realize a trade-off between complexity and error-rate performance. A deep learning-based method is then applied to optimize the message bitwidths. Moreover, we show that parameter sharing can both ensure implementation-friendly solutions and results in faster training convergence than independent parameters. We provide simulation results for 5G low-density parity-check (LDPC) codes and report an error-rate performance within 0.2 dB of floating-point decoding at an average message quantization bitwidth of 3.1 bits. In addition, we show that the learned bitwidths also generalize to other code rates and channels.

Daniel Tandler, Sebastian Dörner, Marc Gauger et al.

We investigate the applicability of deep reinforcement learning algorithms to the adaptive initial access beam alignment problem for mmWave communications using the state-of-the-art proximal policy optimization algorithm as an example. In comparison to recent unsupervised learning based approaches developed to tackle this problem, deep reinforcement learning has the potential to address a new and wider range of applications, since, in principle, no (differentiable) model of the channel and/or the whole system is required for training, and only agent-environment interactions are necessary to learn an algorithm (be it online or using a recorded dataset). We show that, although the chosen off-the-shelf deep reinforcement learning agent fails to perform well when trained on realistic problem sizes, introducing action space shaping in the form of beamforming modules vastly improves the performance, without sacrificing much generalizability. Using this add-on, the agent is able to deliver competitive performance to various state-of-the-art methods on simulated environments, even under realistic problem sizes. This demonstrates that through well-directed modification, deep reinforcement learning may have a chance to compete with other approaches in this area, opening up many straightforward extensions to other/similar scenarios.

Jonathan Mandelbaum, Paul Bezner, Holger Jäkel et al.

In the short block length regime, ensemble decoding schemes with their inherently parallel structure can improve error correction performance and reduce latency compared to stand-alone suboptimal decoders such as belief propagation (BP). In this work, we introduce affine subcode ensemble decoding (aSCED), which uses an ensemble of decoders operating on linear block codes and both linear and strictly affine subcodes. This generalizes the recently proposed subcode ensemble decoding (SCED), which is restricted to linear subcodes. We derive BP update rules for affine subcodes and show that aSCED simplifies ensemble design compared to SCED, multiple bases BP, and automorphism ensemble decoding. Monte-Carlo simulations of two low-density parity-check codes and two Bose-Chaudhuri-Hocquenghem (BCH) codes demonstrate improved error correction performance of aSCED over competing existing ensemble schemes. Notably, for one BCH code, when combining ensemble design with algorithms for constructing high-performance parity-check matrices, aSCED achieves near-maximum likelihood performance using only 64 BP decoding paths.

Jannis Clausius, Marvin Geiselhart, Stephan ten Brink

For improving short-length codes, we demonstrate that classic decoders can also be used with real-valued, neural encoders, i.e., deep-learning based codeword sequence generators. Here, the classical decoder can be a valuable tool to gain insights into these neural codes and shed light on weaknesses. Specifically, the turbo-autoencoder is a recently developed channel coding scheme where both encoder and decoder are replaced by neural networks. We first show that the limited receptive field of convolutional neural network (CNN)-based codes enables the application of the BCJR algorithm to optimally decode them with feasible computational complexity. These maximum a posteriori (MAP) component decoders then are used to form classical (iterative) turbo decoders for parallel or serially concatenated CNN encoders, offering a close-to-maximum likelihood (ML) decoding of the learned codes. To the best of our knowledge, this is the first time that a classical decoding algorithm is applied to a non-trivial, real-valued neural code. Furthermore, as the BCJR algorithm is fully differentiable, it is possible to train, or fine-tune, the neural encoder in an end-to-end fashion.

Paul Bezner, Erdem Eray Cil, Jannis Clausius et al.

The growing demand for higher data rates necessitates continuous innovations in wireless communication systems, particularly with the emergence of 6G. Channel coding plays a crucial role in this evolution. In 5G systems, rate-adaptive raptor-like quasi-cyclic irregular low-density parity-check codes are used for the data link, while polar codes with successive cancellation list decoding handle short messages on the synchronization channel. However, to meet the stringent requirements of future 6G systems, a versatile and unified coding scheme should be developed - one that offers competitive error-correcting performance alongside low complexity encoding and decoding schemes that enable energy-efficient hardware implementations. This white paper outlines the vision for such a unified coding scheme. We explore various 6G communication scenarios that pose new challenges to channel coding and provide a first analysis of potential solutions.

Sebastian Dörner, Sebastian Cammerer, Jakob Hoydis et al.

End-to-end learning of communications systems is a fascinating novel concept that has so far only been validated by simulations for block-based transmissions. It allows learning of transmitter and receiver implementations as deep neural networks (NNs) that are optimized for an arbitrary differentiable end-to-end performance metric, e.g., block error rate (BLER). In this paper, we demonstrate that over-the-air transmissions are possible: We build, train, and run a complete communications system solely composed of NNs using unsynchronized off-the-shelf software-defined radios (SDRs) and open-source deep learning (DL) software libraries. We extend the existing ideas towards continuous data transmission which eases their current restriction to short block lengths but also entails the issue of receiver synchronization. We overcome this problem by introducing a frame synchronization module based on another NN. A comparison of the BLER performance of the "learned" system with that of a practical baseline shows competitive performance close to 1 dB, even without extensive hyperparameter tuning. We identify several practical challenges of training such a system over actual channels, in particular the missing channel gradient, and propose a two-step learning procedure based on the idea of transfer learning that circumvents this issue.

Tim Janz, Simon Obermüller, Andreas Zunker et al.

Chase-like decoding algorithms are a popular choice for soft-input decoding of algebraic codes. In this paper, we evaluate the performance of different test pattern sets using three methods. For test pattern sets with a certain structure such as Chase-II test patterns and patterns up to a maximum logistic weight, we use a method that relies on order statistics. The performance of arbitrary sets of test patterns is evaluated by calculating covered space probabilities and via direct Monte Carlo simulation. Based on the idea of covering as many likely error patterns as possible, we propose an algorithm for the design of test pattern sets which performs up to 0.2\,dB better for high-rate BCH codes than commonly used test patterns.

Phillip Stephan, Florian Euchner, Stephan ten Brink

The ability of modern telecommunication systems to locate users and objects in the radio environment raises justified privacy concerns. To prevent unauthorized localization, single-antenna transmitters can obfuscate the signal by convolving it with a randomized sequence prior to transmission, which alters the channel state information (CSI) estimated at the receiver. However, this strategy is only effective against CSI-based localization systems deploying single-antenna receivers. Inspired by the concept of blind multichannel identification, we propose a simple CSI recovery method for multi-antenna receivers to extract channel features that ensure reliable user localization regardless of the transmitted signal. We comparatively evaluate the impact of signal obfuscation and the proposed recovery method on the localization performance of CSI fingerprinting, channel charting, and classical triangulation using real-world channel measurements. This work aims to demonstrate the necessity for further efforts to protect the location privacy of users from adversarial radio-based localization systems.

Jannis Clausius, Marvin Geiselhart, Stephan ten Brink

Isolated training with Gaussian priors (TGP) of the component autoencoders of turbo-autoencoder architectures enables faster, more consistent training and better generalization to arbitrary decoding iterations than training based on deep unfolding. We propose fitting the components via extrinsic information transfer (EXIT) charts to a desired behavior which enables scaling to larger message lengths ($k \approx 1000$) while retaining competitive performance. To the best of our knowledge, this is the first autoencoder that performs close to classical codes in this regime. Although the binary cross-entropy (BCE) loss function optimizes the bit error rate (BER) of the components, the design via EXIT charts enables to focus on the block error rate (BLER). In serially concatenated systems the component-wise TGP approach is well known for inner components with a fixed outer binary interface, e.g., a learned inner code or equalizer, with an outer binary error correcting code. In this paper we extend the component training to structures with an inner and outer autoencoder, where we propose a new 1-bit quantization strategy for the encoder outputs based on the underlying communication problem. Finally, we discuss the model complexity of the learned components during design time (training) and inference and show that the number of weights in the encoder can be reduced by 99.96 %.

Moustafa Ebada, Sebastian Cammerer, Ahmed Elkelesh et al.

In this work, we introduce a deep learning-based polar code construction algorithm. The core idea is to represent the information/frozen bit indices of a polar code as a binary vector which can be interpreted as trainable weights of a neural network (NN). For this, we demonstrate how this binary vector can be relaxed to a soft-valued vector, facilitating the learning process through gradient descent and enabling an efficient code construction. We further show how different polar code design constraints (e.g., code rate) can be taken into account by means of careful binary-to-soft and soft-to-binary conversions, along with rate-adjustment after each learning iteration. Besides its conceptual simplicity, this approach benefits from having the "decoder-in-the-loop", i.e., the nature of the decoder is inherently taken into consideration while learning (designing) the polar code. We show results for belief propagation (BP) decoding over both AWGN and Rayleigh fading channels with considerable performance gains over state-of-the-art construction schemes.

Mark Widmaier, Maximilian Arnold, Sebastian Dörner et al.

We showcase the practicability of an indoor positioning system (IPS) solely based on Neural Networks (NNs) and the channel state information (CSI) of a (Massive) multiple-input multiple-output (MIMO) communication system, i.e., only build on the basis of data that is already existent in today's systems. As such our IPS system promises both, a good accuracy without the need of any additional protocol/signaling overhead for the user localization task. In particular, we propose a tailored NN structure with an additional phase branch as feature extractor and (compared to previous results) a significantly reduced amount of trainable parameters, leading to a minimization of the amount of required training data. We provide actual measurements for indoor scenarios with up to 64 antennas covering a large area of 80m2. In the second part, several robustness investigations for real-measurements are conducted, i.e., once trained, we analyze the recall accuracy over a time-period of several days. Further, we analyze the impact of pedestrians walking in-between the measurements and show that finetuning and pre-training of the NN helps to mitigate effects of hardware drifts and alterations in the propagation environment over time. This reduces the amount of required training samples at equal precision and, thereby, decreases the effort of the costly training data acquisition

Daniel Tandler, Sebastian Dörner, Sebastian Cammerer et al.

In this work, we analyze the capabilities and practical limitations of neural networks (NNs) for sequence-based signal processing which can be seen as an omnipresent property in almost any modern communication systems. In particular, we train multiple state-of-the-art recurrent neural network (RNN) structures to learn how to decode convolutional codes allowing a clear benchmarking with the corresponding maximum likelihood (ML) Viterbi decoder. We examine the decoding performance for various kinds of NN architectures, beginning with classical types like feedforward layers and gated recurrent unit (GRU)-layers, up to more recently introduced architectures such as temporal convolutional networks (TCNs) and differentiable neural computers (DNCs) with external memory. As a key limitation, it turns out that the training complexity increases exponentially with the length of the encoding memory $ν$ and, thus, practically limits the achievable bit error rate (BER) performance. To overcome this limitation, we introduce a new training-method by gradually increasing the number of ones within the training sequences, i.e., we constrain the amount of possible training sequences in the beginning until first convergence. By consecutively adding more and more possible sequences to the training set, we finally achieve training success in cases that did not converge before via naive training. Further, we show that our network can learn to jointly detect and decode a quadrature phase shift keying (QPSK) modulated code with sub-optimal (anti-Gray) labeling in one-shot at a performance that would require iterations between demapper and decoder in classic detection schemes.

Ahmed Elkelesh, Moustafa Ebada, Sebastian Cammerer et al.

We propose a new framework for constructing polar codes (i.e., selecting the frozen bit positions) for arbitrary channels, and tailored to a given decoding algorithm, rather than based on the (not necessarily optimal) assumption of successive cancellation (SC) decoding. The proposed framework is based on the Genetic Algorithm (GenAlg), where populations (i.e., collections) of information sets evolve successively via evolutionary transformations based on their individual error-rate performance. These populations converge towards an information set that fits both the decoding behavior and the defined channel. Using our proposed algorithm over the additive white Gaussian noise (AWGN) channel, we construct a polar code of length 2048 with code rate 0.5, without the CRC-aid, tailored to plain successive cancellation list (SCL) decoding, achieving the same error-rate performance as the CRC-aided SCL decoding, and leading to a coding gain of 1 dB at BER of $10^{-6}$. Further, a belief propagation (BP)-tailored construction approaches the SCL error-rate performance without any modifications in the decoding algorithm itself. The performance gains can be attributed to the significant reduction in the total number of low-weight codewords. To demonstrate the flexibility, coding gains for the Rayleigh channel are shown under SCL and BP decoding. Besides improvements in error-rate performance, we show that, when required, the GenAlg can be also set up to reduce the decoding complexity, e.g., the SCL list size or the number of BP iterations can be reduced, while maintaining the same error-rate performance.

Ahmed Elkelesh, Moustafa Ebada, Sebastian Cammerer et al.

We propose a new polar code construction framework (i.e., selecting the frozen bit positions) for the additive white Gaussian noise (AWGN) channel, tailored to a given decoding algorithm, rather than based on the (not necessarily optimal) assumption of successive cancellation (SC) decoding. The proposed framework is based on the Genetic Algorithm (GenAlg), where populations (i.e., collections) of information sets evolve successively via evolutionary transformations based on their individual error-rate performance. These populations converge towards an information set that fits the decoding behavior. Using our proposed algorithm, we construct a polar code of length 2048 with code rate 0.5, without the CRC-aid, tailored to plain successive cancellation list (SCL) decoding, achieving the same error-rate performance as the CRC-aided SCL decoding, and leading to a coding gain of 1 dB at BER of $10^{-6}$. Further, a belief propagation (BP)-tailored polar code approaches the SCL error-rate performance without any modifications in the decoding algorithm itself.

Maximilian Arnold, Sebastian Dörner, Sebastian Cammerer et al.

A major obstacle for widespread deployment of frequency division duplex (FDD)-based Massive multiple-input multiple-output (MIMO) communications is the large signaling overhead for reporting full downlink (DL) channel state information (CSI) back to the basestation (BS), in order to enable closed-loop precoding. We completely remove this overhead by a deep-learning based channel extrapolation (or "prediction") approach and demonstrate that a neural network (NN) at the BS can infer the DL CSI centered around a frequency $f_\text{DL}$ by solely observing uplink (UL) CSI on a different, yet adjacent frequency band around $f_\text{UL}$; no more pilot/reporting overhead is needed than with a genuine time division duplex (TDD)-based system. The rationale is that scatterers and the large-scale propagation environment are sufficiently similar to allow a NN to learn about the physical connections and constraints between two neighboring frequency bands, and thus provide a well-operating system even when classic extrapolation methods, like the Wiener filter (used as a baseline for comparison throughout) fails. We study its performance for various state-of-the-art Massive MIMO channel models, and, even more so, evaluate the scheme using actual Massive MIMO channel measurements, rendering it to be practically feasible at negligible loss in spectral efficiency when compared to a genuine TDD-based system.