Recurrent Neural Networks (RNNs) are powerful models for sequential tasks, but their training can be computationally intensive. On the other hand, Echo State Networks (ESNs), a specific RNN architecture, simplify this by configuring a fixed, random reservoir of recurrent weights. Instead of setting random weights as in the ESN case, in this work, we investigate the recurrent weight configuration problem of the fully-fledged RNN for the task of Linear Time-Invariant (LTI) system approximation. Our investigation focuses on a specific RNN architecture with a network of recurrent neurons limited to self-loops and linear activation. We demonstrate that as the recurrent weights of this RNN are trained on large datasets, their distribution converges to a near-identical match of an optimal distribution that can be analytically derived using the available domain knowledge of the LTI system. This insight establishes that domain-informed weight configuration is a highly efficient alternative to data-driven training. Building upon this, we propose a novel deterministic algorithm to set the recurrent weights, which significantly improves approximation accuracy. Numerical results show our domain-informed RNN weight configuration achieves up to a four-order-of-magnitude performance gain over conventional ESNs.

Deep learning is making a profound impact in the physical layer of wireless communications. Despite exhibiting outstanding empirical performance in tasks such as MIMO receive processing, the reasons behind the demonstrated superior performance improvement remain largely unclear. In this work, we advance the field of Explainable AI (xAI) in the physical layer of wireless communications utilizing signal processing principles. Specifically, we focus on the task of MIMO-OFDM receive processing (e.g., symbol detection) using reservoir computing (RC), a framework within recurrent neural networks (RNNs), which outperforms both conventional and other learning-based MIMO detectors. Our analysis provides a signal processing-based, first-principles understanding of the corresponding operation of the RC. Building on this fundamental understanding, we are able to systematically incorporate the domain knowledge of wireless systems (e.g., channel statistics) into the design of the underlying RNN by directly configuring the untrained RNN weights for MIMO-OFDM symbol detection. The introduced RNN weight configuration has been validated through extensive simulations demonstrating significant performance improvements. This establishes a foundation for explainable RC-based architectures in MIMO-OFDM receive processing and provides a roadmap for incorporating domain knowledge into the design of neural networks for NextG systems.

The increased flexibility and density of spectrum access in 5G New Radio (NR) has made jamming detection and classification a critical research area. To detect coexisting jamming and subtle interference, we introduce a Bayesian Inference-assisted machine learning (ML) methodology. Our methodology uses cross-layer Key Performance Indicator data collected on a Non-Standalone (NSA) 5G NR testbed to leverage supervised learning models, further assessed, calibrated, and revealed using Bayesian Network Model (BNM)-based inference. The models can operate on both instantaneous and sequential time-series data samples, achieving an Area under Curve above 0.954 for instantaneous models and above 0.988 for sequential models including the echo state network (ESN) from the Reservoir Computing (RC) family, across various jamming scenarios. The 180 ms instantaneous detection time allows for continuous tracking of the dynamic jamming condition due to UE mobility. Our approach serves as a validation method and a resilience enhancement tool for ML-based jamming detection while also enabling root cause identification for observed performance degradation. The introduced BNM-based inference proof-of-concept is successful in addressing 72.2% of the erroneous predictions of the RC-based sequential detection model caused by insufficient training data samples, thereby demonstrating its near real-time applicability in 5G NR and Beyond-5G networks.

In this paper, we derive a theoretical upper bound on the generalization error of reservoir computing (RC), a special category of recurrent neural networks (RNNs). The specific RC implementation considered in this paper is the echo state network (ESN), and an upper bound on its generalization error is derived via the empirical Rademacher complexity (ERC) approach. While recent work in deriving risk bounds for RC frameworks makes use of a non-standard ERC measure and a direct application of its definition, our work uses the standard ERC measure and tools allowing fair comparison with conventional RNNs. The derived result shows that the generalization error bound obtained for ESNs is tighter than the existing bound for vanilla RNNs, suggesting easier generalization for ESNs. With the ESN applied to symbol detection in MIMO-OFDM (Multiple Input Multiple Output-Orthogonal Frequency Division Multiplexing) systems, we show how the derived generalization error bound can guide underlying system design. Specifically, the derived bound together with the empirically characterized training loss is utilized to identify the optimum reservoir size in neurons for the ESN-based symbol detector. Finally, we corroborate our theoretical findings with results from simulations that employ 3GPP standards-compliant wireless channels, signifying the practical relevance of our work.

Scalable video caching is a promising technique to alleviate backbone traffic in sixth generation (6G) networks, and to serve users with video quality that adapts to varying channel conditions. In this paper, we develop a layer-based scalable video caching technique with non-orthogonal transmission by taking advantage of the layer feature in the scalable video. In addition, the impact of different serving base station selection algorithms is investigated. Our results indicate that both the caching placement design and transmission scheme design dominate the caching performance. To evaluate the interplay of these two policies, a tractable metric of Caching Aided Data Rate (CADR) is characterized and maximized by jointly optimizing the aforementioned two policies. Together with extensive Monte Carlo simulations, numerical results are also evaluated in this paper, demonstrating that the proposed Layer-based video Caching scheme with Non-Orthogonal Transmission (LCNOT) can achieve higher CADR performance than other baseline schemes.

Jamming and intrusion detection are some of the most important research domains in 5G that aim to maintain use-case reliability, prevent degradation of user experience, and avoid severe infrastructure failure or denial of service in mission-critical applications. This paper introduces an anonymous jamming detection model for 5G and beyond based on critical signal parameters collected from the radio access and core network’s protocol stacks on a 5G testbed. The introduced system leverages both supervised and unsupervised learning to detect jamming with high-accuracy in real time, and allows for robust detection of unknown jamming types. Based on the given types of jamming, supervised instantaneous detection models reach an Area Under the Curve (AUC) within a range of 0.964 to 1 as compared to temporal-based long short-term memory (LSTM) models that reach AUC within a range of 0.923 to 1. The need for data annotation effort and the required knowledge of a vocabulary of known jamming limits the usage of the introduced supervised learning-based approach. To mitigate this issue, an unsupervised auto-encoder-based anomaly detection is also presented. The introduced unsupervised approach has an AUC of 0.987 with training samples collected without any jamming or interference and shows resistance to adversarial training samples within certain percentage. To retain transparency and allow domain knowledge injection, a Bayesian network model based causation analysis is further introduced.

In this paper, we derive an upper bound on the generalization error of reservoir computing (RC), a special category of recurrent neural networks (RNNs). The particular RC implementation considered in this paper is the echo state network (ESN), and the generalization bound is derived via the empirical Rademacher complexity (ERC) approach. While recent work in deriving risk bounds for RC frameworks makes use of a non-standard empirical Rademacher Complexity (ERC) measure and a direct application of its definition, our work uses the standard ERC measure and tools which allow easier extension to deep RC structures and a fair comparison with other RNN structures including the vanilla RNNs. Finally, we train an ESN for the task of symbol detection, a key component in the receive processing of 5G systems and show with an example how the derived generalization bound can guide the underlying system design. To be specific, we utilize the derived generalization error together with the characterized training loss to analytically identify the optimum number of neurons in the reservoir for the symbol detection task. Simulation results using 3GPP specified channels corroborate our theoretical findings, illustrating the significance and practical relevance of our work.

In this paper, we investigate online learning-based MIMO-OFDM symbol detection strategies focusing on a special recurrent neural network (RNN) - reservoir computing (RC). We first introduce the Time-Frequency RC to take advantage of the structural information inherent in OFDM signals. Using the time domain RC and the time-frequency RC as building blocks, we provide two extensions of the shallow RC to RCNet: 1) Stacking multiple time domain RCs; 2) Stacking multiple time-frequency RCs into a deep structure. The combination of RNN dynamics, the time-frequency structure of MIMO-OFDM signals, and the deep network enables RCNet to handle the interference and nonlinear distortion of MIMO-OFDM signals to outperform existing methods. Unlike most existing NN-based detection strategies, RCNet is also shown to provide a good generalization performance even with a limited online training set (i.e, similar amount of reference signals/training as standard model-based approaches). Numerical experiments demonstrate that the introduced RCNet can offer a faster learning convergence and as much as 20% gain in bit error rate over a shallow RC structure by compensating for the nonlinear distortion of the MIMO-OFDM signal, such as due to power amplifier compression in the transmitter or due to finite quantization resolution in the receiver.

With the rise in the population of the world's cities, understanding the dynamics of commuters' transportation patterns has become crucial in the planning and management of urban facilities and services. In this study, we analyze how commuter patterns change during different time instances such as between weekdays and weekends. To this end, we propose two data mining techniques, namely Common Orthogonal Basis Extraction (COBE), and Joint and Individual Variation Explained (JIVE) for Integrated Analysis of Multiple Data Types and apply them to smart card data available for passengers in Singapore. We also discuss the issues of model selection and interpretability of these methods. The joint and individual patterns can help transportation companies optimize their resources in light of changes in commuter mobility behavior.