Leveraging the Doppler Effect for Channel Charting

Channel Charting is a dimensionality reduction technique that reconstructs a map of the radio environment from similarity relationships found in channel state information. Distances in the channel chart are often computed based on some dissimilarity metric, which can be derived from angular-domain information, channel impulse responses, measured phase differences or simply timestamps. Using such information implicitly makes strong assumptions about the level of phase and time synchronization between base station antennas or assumes approximately constant transmitter velocity. Many practical systems, however, may not provide phase and time synchronization and single-antenna base stations may not even have angular-domain information. We propose a Doppler effect-based loss function for Channel Charting that only requires frequency synchronization between spatially distributed base station antennas, which is a much weaker assumption. We use a dataset measured in an indoor environment to demonstrate that the proposed method is practically feasible with just four base station antennas, that it produces a channel chart that is suitable for localization in the global coordinate frame and that it outperforms other state-of-the-art methods under the given limitations.

GAN-based Massive MIMO Channel Model Trained on Measured Data

Wireless channel models are a commonly used tool for the development of wireless telecommunication systems and standards. The currently prevailing geometry-based stochastic channel models (GSCMs) were manually specified for certain environments in a manual process requiring extensive domain knowledge, on the basis of channel measurement campaigns. By taking into account the stochastic distribution of certain channel properties like Rician k-factor, path loss or delay spread, they model the distribution of channel realizations. Instead of this manual process, a generative machine learning model like a generative adversarial network (GAN) may be used to automatically learn the distribution of channel statistics. Subsequently, the GAN's generator may be viewed as a channel model that can replace conventional stochastic or raytracer-based models. We propose a GAN architecture for a massive MIMO channel model, and train it on measurement data produced by a distributed massive MIMO channel sounder.

CRAP Part II: Clutter Removal with Continuous Acquisitions Under Phase Noise

The mitigation of clutter is an important research branch in Integrated Sensing and Communication (ISAC), one of the emerging technologies of future cellular networks. In this work, we extend our previously introduced method Clutter Removal with Acquisitions Under Phase Noise (CRAP) by means to track clutter over time. This is necessary in scenarios that require high reliability but can change dynamically, like safety applications in factory floors. To that end, exponential smoothing is leveraged to process new measurements and previous clutter information in a unique matrix using the singular value decomposition, allowing adaptation to changing environments in an efficient way.We further propose a singular value threshold based on the Marchenko-Pastur distribution to select the meaningful clutter components. Results from both simulations and measurements show that continuously updating the clutter components with new acquisitions according to our proposed algorithm Smoothed CRAP (SCRAP) enables coping with dynamic clutter environments and facilitates the detection of sensing targets.

Antenna Array Design for Mono-Static ISAC

Mono-static sensing operations in Integrated Sensing and Communications (ISAC) require joint beamforming operations between transmitter and receiver, according to all the considerations already done in the radar literature about coarray theory. In contrast to pure radar systems, ISAC requires to fulfill communications tasks and to retain the corresponding design constraints for at least one half-duplex array. This shifts the available degrees of freedom to the design of the second half-duplex array, that completes the mono-static sensing setup of the ISAC system. Therefore, it is necessary to translate the analysis from the radar literature for the design of sparse arrays to the new ISAC paradigm in order to provision such systems. Accordingly, we propose a model to evaluate the angular capabilities of an ISAC setup, constrained to the shape of the communications array and its topology requirements. Our analysis is validated by simulation experiments, confirming the value of our model in providing system designers with a tool to drastically improve the trade-off between angular capabilities for sensing and the cost of the deployed hardware. Finally, we discuss possible enhancements to the cellular standards to fully leverage the angular capabilities of such mono-static ISAC systems.

Graph Neural Network-based Joint Equalization and Decoding

This paper proposes to use graph neural networks (GNNs) for equalization, that can also be used to perform joint equalization and decoding (JED). For equalization, the GNN is build upon the factor graph representations of the channel, while for JED, the factor graph is expanded by the Tanner graph of the parity-check matrix (PCM) of the channel code, sharing the variable nodes (VNs). A particularly advantageous property of the GNN is the robustness against cycles in the factor graphs which is the main problem for belief propagation (BP)-based equalization. As a result of having a fully deep learning-based receiver, joint optimization instead of individual optimization of the components is enabled, so-called end-to-end learning. Furthermore, we propose a parallel flooding schedule that further reduces the latency, which turns out to improve also the error correcting performance. The proposed approach is analyzed and compared to state-of-the-art baselines in terms of error correcting capability and latency. At a fixed low latency, the flooding GNN for JED demonstrates a gain of 2.25 dB in bit error rate (BER) compared to an iterative Bahl--Cock--Jelinek--Raviv (BCJR)-BP baseline.

Deep Learning Based Adaptive Joint mmWave Beam Alignment

The challenging propagation environment, combined with the hardware limitations of mmWave systems, gives rise to the need for accurate initial access beam alignment strategies with low latency and high achievable beamforming gain. Much of the recent work in this area either focuses on one-sided beam alignment, or, joint beam alignment methods where both sides of the link perform a sequence of fixed channel probing steps. Codebook-based non-adaptive beam alignment schemes have the potential to allow multiple user equipment (UE) to perform initial access beam alignment in parallel whereas adaptive schemes are favourable in achievable beamforming gain. This work introduces a novel deep learning based joint beam alignment scheme that aims to combine the benefits of adaptive, codebook-free beam alignment at the UE side with the advantages of a codebook-sweep based scheme at the base station. The proposed end-to-end trainable scheme is compatible with current cellular standard signaling and can be readily integrated into the standard without requiring significant changes to it. Extensive simulations demonstrate superior performance of the proposed approach over purely codebook-based ones.

Augmenting Channel Charting with Classical Wireless Source Localization Techniques

Channel Charting aims to construct a map of the radio environment by leveraging similarity relationships found in high-dimensional channel state information. Although resulting channel charts usually accurately represent local neighborhood relationships, even under conditions with strong multipath propagation, they often fall short in capturing global geometric features. On the other hand, classical model-based localization methods, such as triangulation and multilateration, can easily localize signal sources in the global coordinate frame. However, these methods rely heavily on the assumption of line-of-sight channels and distributed antenna deployments. Based on measured data, we compare classical source localization techniques to channel charts with respect to localization performance. We suggest and evaluate methods to enhance Channel Charting with model-based localization approaches: One approach involves using information derived from classical localization methods to map channel chart locations to physical positions after conventional training of the forward charting function. Foremost, though, we suggest to incorporate information from model-based approaches during the training of the forward charting function in what we call "augmented Channel Charting". We demonstrate that Channel Charting can outperform classical localization methods on the considered dataset.

Learning Radio Environments by Differentiable Ray Tracing

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.

Angle-Delay Profile-Based and Timestamp-Aided Dissimilarity Metrics for Channel Charting

Channel charting is a self-supervised learning technique whose objective is to reconstruct a map of the radio environment, called channel chart, by taking advantage of similarity relationships in high-dimensional channel state information. We provide an overview of processing steps and evaluation methods for channel charting and propose a novel dissimilarity metric that takes into account angular-domain information as well as a novel deep learning-based metric. Furthermore, we suggest a method to fuse dissimilarity metrics such that both the time at which channels were measured as well as similarities in channel state information can be taken into consideration while learning a channel chart. By applying both classical and deep learning-based manifold learning to a dataset containing sub-6GHz distributed massive MIMO channel measurements, we show that our metrics outperform previously proposed dissimilarity measures. The results indicate that the new metrics improve channel charting performance, even under non-line-of-sight conditions.

Multi-Target Localization in Multi-Static Integrated Sensing and Communication Deployments

In future wireless communication networks, existing active localization will gradually evolve into more sophisticated (passive) sensing functionalities. One main enabler for this process is the merging of information collected from the network's nodes, sensing the environment in a multi-static deployment. The current literature considers single sensing node systems and/or single target scenarios, mainly focusing on specific issues pertaining to hardware impairments or algorithmic challenges. In contrast, in this work we propose an ensemble of techniques for processing the information gathered from multiple sensing nodes, jointly observing an environment with multiple targets. A scattering model is used within a flexibly configurable framework to highlight the challenges and issues with algorithms used in this distributed sensing task. We validate our approach by supporting it with detailed link budget evaluations, considering practical millimeter-wave systems' capabilities. Our numerical evaluations are performed in an indoor scenario, sweeping a variety of parameter to analyze the KPIs sensitivity with respect to each of them. The proposed algorithms to fuse information by multiple nodes show significant gains in terms of targets' localization performance, with up to 35\% for the probability of detection, compared to the baseline with a mono-static setup.