Wideband Illumination with Liquid Crystal Reconfigurable Intelligent Surfaces: Modeling, Design, and Experimental Tests

Liquid crystal (LC) is a promising hardware solution for implementing large RISs, as it is cost-effective, energy efficient, scalable, and capable of providing continuous phase shifts with low power consumption. However, the phase shift response of LC-based RISs is inherently frequency dependent. If unaddressed, this characteristic leads to performance degradation, particularly in wideband scenarios. This issue is especially critical in secure communication applications, where minor phase shift variations across elements can result in considerable information leakage. This paper addresses these frequency-induced variations by developing a physics-based model for an LC unit cell across varying frequencies and proposing a novel phase shift design framework that maximizes secure communication across all subcarriers. Given the large number of elements in millimeter wave (mmWave) LC-RISs, acquiring full channel state information (CSI) is often impractical. Therefore, we optimize the phase shifts based solely on the locations of the legitimate mobile users (MUs) and potential eavesdroppers. Rather than targeting a single user point, the RIS is designed to illuminate a broader area. This approach enhances communication reliability for the MUs and mitigates performance degradation caused by location estimation errors. To solve the problem, we introduce both a semi-definite programming (SDP)-based solution and a low complexity heuristic method. While the SDP-based approach yields superior performance, it incurs higher computational complexity. Conversely, the scalable method exhibits a much slower scaling of complexity, which makes it highly suitable for extremely large RISs. Simulation results demonstrate that both algorithms improve the secrecy rate compared to baseline methods. Finally, the proposed design is validated through experimental evaluations on an LC RIS setup.

Autoencoder-based Optimization of Multi-user Molecule Mixture Communication Systems

In this paper, we introduce an autoencoder (AE)-based scheme for end-to-end optimization of a multi-user molecule mixture communication system. In the proposed scheme, each transmitter leverages an encoder network that maps the user symbol to a molecule mixture. The mixtures then propagate through the channel to the receiver, which samples the channel using a non-linear, cross-reactive sensor array. A decoder network then estimates the symbol transmitted by each user based on the sensor observations. The proposed scheme achieves, for a given signal-to-noise ratio, lower symbol error rates than a baseline scheme from the literature in a single-user setting with full channel state information. We additionally demonstrate that the proposed AE-based scheme allows reliable communication when the channel is unknown or changing. Finally, we show that for multiple access the system can account for different user priorities. In summary, the proposed AE-based scheme enables end-to-end system optimization in complex scenarios unsuitable for analytical treatment and thereby brings molecular communication systems closer to real-world deployment.

Molecule Mixture Detection and Design for MC Systems with Non-linear, Cross-reactive Receiver Arrays

Air-based molecular communication (MC) has the potential to be one of the first MC systems to be deployed in real-world applications, enabled by commercially available sensors. However, these sensors usually exhibit non-linear and cross-reactive behavior, contrary to the idealizing assumption of linear and perfectly molecule type-specific sensing often made in the MC literature. To address this mismatch, we propose several detectors and transmission schemes for a molecule mixture communication system where the receiver (RX) employs non-linear, cross-reactive sensors. All proposed schemes are based on the first- and second-order moments of the symbol likelihoods that are fed through the non-linear RX using the Unscented Transform. In particular, we propose an approximate maximum likelihood (AML) symbol-by-symbol detector for inter-symbol-interference (ISI)-free transmission scenarios and a complementary mixture alphabet design algorithm which accounts for the RX characteristics. When significant ISI is present at high data rates, the AML detector can be adapted to exploit statistical ISI knowledge. Additionally, we propose a sequence detector which combines information from multiple symbol intervals. For settings where sequence detection is not possible due to extremely limited computational power at the RX, we propose an adaptive transmission scheme which can be combined with symbol-by-symbol detection. Using computer simulations, we validate all proposed detectors and algorithms based on the responses of commercially available sensors as well as artificially generated sensor data incorporating the characteristics of metal-oxide semiconductor sensors. By employing a general system model that accounts for transmitter noise, ISI, and general non-linear, cross-reactive RX arrays, this work enables reliable communication for a large class of MC systems.

Flexible RISs: Learning-based Array Manifold Estimation and Phase-shift Optimization

Reconfigurable intelligent surfaces (RISs) are envisioned as a key enabler for next-generation wireless networks, offering programmable control over propagation environments. While extensive research focuses on planar RIS architectures, practical deployments often involve non-planar surfaces, such as structural columns or curved facades, where standard planar beamforming models fail. Moreover, existing analytical solutions for curved RISs are often restricted to specific, pre-defined array manifold geometries. To address this limitation, this paper proposes a novel deep learning (DL) framework for optimizing the phase shifts of non-planar RISs. We first introduce a low-dimensional parametric model to capture arbitrary surface curvature effectively. Based on this, we design a neural network (NN) that utilizes a sparse set of received power measurements to estimate the surface geometry and derive the optimal phase configuration. Simulation results demonstrate that the proposed algorithm converges fast and significantly outperforms conventional planar beamforming designs, validating its robustness against arbitrary surface curvature. We also analyze the impact of the measurement location error on the algorithm's performance.

Transparent and Resilient Activity Recognition via Attention-Based Distributed Radar Sensing

Distributed radar sensors enable robust human activity recognition. However, scaling the number of coordinated nodes introduces challenges in feature extraction from large datasets, and transparent data fusion. We propose an end-to-end framework that operates directly on raw radar data. Each radar node employs a lightweight 2D Convolutional Neural Network (CNN) to extract local features. A self-attention fusion block then models inter-node relationships and performs adaptive information fusion. Local feature extraction reduces the input dimensionality by up to 480x. This significantly lowers communication overhead and latency. The attention mechanism provides inherent interpretability by quantifying the contribution of each radar node. A hybrid supervised contrastive loss further improves feature separability, especially for fine-grained and imbalanced activity classes. Experiments on real-world distributed Ultra Wide Band (UWB) radar data demonstrate that the proposed method reduces model complexity by 70.8\%, while achieving higher average accuracy than baseline approaches. Overall, the framework enables transparent, efficient, and low-overhead distributed radar sensing.

Learning-Based Phase Shift Optimization of Liquid Crystal RIS in Dynamic mmWave Networks

To enhance coverage and signal quality in millimeter-wave (mmWave) frequencies, reconfigurable intelligent surfaces (RISs) have emerged as a game-changing solution to manipulate the wireless environment. Traditional semiconductor-based RISs face scalability issues due to high power consumption. Meanwhile, liquid crystal-based RISs (LC-RISs) offer energy-efficient and cost-effective operation even for large arrays. However, this promise has a caveat. LC-RISs suffer from long reconfiguration times, on the order of tens of milliseconds, which limits their applicability in dynamic scenarios. To date, prior works have focused on hardware design aspects or static scenarios to address this limitation, but little attention has been paid to optimization solutions for dynamic settings. Our paper fills this gap by proposing a reinforcement learning-based optimization framework to dynamically control the phase shifts of LC-RISs and maximize the data rate of a moving user. Specifically, we propose a Deep Deterministic Policy Gradient (DDPG) algorithm that adapts the LC-RIS phase shifts without requiring perfect channel state information and balances the tradeoff between signal-to-noise ratio (SNR) and configuration time. We validate our approach through high-fidelity ray tracing simulations, leveraging measurement data from an LC-RIS prototype. Our results demonstrate the potential of our solution to bring adaptive control to dynamic LC-RIS-assisted mmWave systems.

RIS-Assisted NOMA with Partial CSI and Mutual Coupling: A Machine Learning Approach

Non-orthogonal multiple access (NOMA) is a promising multiple access technique. Its performance depends strongly on the wireless channel property, which can be enhanced by reconfigurable intelligent surfaces (RISs). In this paper, we jointly optimize base station (BS) precoding and RIS configuration with unsupervised machine learning (ML), which looks for the optimal solution autonomously. In particular, we propose a dedicated neural network (NN) architecture RISnet inspired by domain knowledge in communication. Compared to state-of-the-art, the proposed approach combines analytical optimal BS precoding and ML-enabled RIS, has a high scalability to control more than 1000 RIS elements, has a low requirement for channel state information (CSI) in input, and addresses the mutual coupling between RIS elements. Beyond the considered problem, this work is an early contribution to domain knowledge enabled ML, which exploit the domain expertise of communication systems to design better approaches than general ML methods.

A Scalable Machine Learning Approach Enabled RIS Optimization with Implicit Channel Estimation

The reconfigurable intelligent surface (RIS) is considered as a key enabler of the next-generation mobile radio systems. While attracting extensive interest from academia and industry due to its passive nature and low cost, scalability of RIS elements and requirement for channel state information (CSI) are two major difficulties for the RIS to become a reality. In this work, we introduce an unsupervised machine learning (ML) enabled optimization approach to configure the RIS. The dedicated neural network (NN) architecture RISnet is combined with an implicit channel estimation method. The RISnet learns to map from received pilot signals to RIS configuration directly without explicit channel estimation. Simulation results show that the proposed algorithm outperforms baselines significantly.

Temperature-Resilient LC-RIS Phase-Shift Design for Multi-user Downlink Communications

The reflecting antenna elements in most reconfigurable intelligent surfaces (RISs) use semiconductor-based (e.g., positive-intrinsic-negative (PIN) diodes and varactors) phase shifters. Although effective, a drawback of this technology is the high power consumption and cost, which become particularly prohibitive in millimeter-wave (mmWave)/sub-Terahertz range. With the advances in Liquid Crystals (LCs) in microwave engineering, we have observed a new trend in using LC for realizing phase shifter networks of RISs. LC-RISs are expected to significantly reduce the fabrication costs and power consumption. However, the nematic LC molecules are sensitive to temperature variations. Therefore, implementing LC-RIS in geographical regions with varying temperatures requires temperature-resilient designs. The mentioned temperature variation issue becomes more significant at higher temperatures as the phase shifter range reduces in warmer conditions, whereas it expands in cooler ones. In this paper, we study the impact of temperature on the operation of LC-RISs and develop a temperature-resilient phase shift design. Specifically, we formulate a max-min signal-to-interference-plus-noise ratio optimization for a multi-user downlink mmWave network that accounts for the impact of temperature in the LC-RIS phase shifts. The simulation results demonstrate a significant improvement for the considered set of parameters when using our algorithm compared to the baseline approach, which neglects the temperature effects.

Fast Reconfiguration of LC-RISs: Modeling and Algorithm Design

LC technology is a promising hardware solution for realizing extremely large RISs due to its advantages in cost-effectiveness, scalability, energy efficiency, and continuous phase shift tunability. However, the slow response time of the LC cells, especially in comparison to the silicon-based alternatives like radio frequency switches and PIN diodes, limits the performance. This limitation becomes particularly relevant in TDMA applications where RIS must sequentially serve users in different locations, as the phase-shifting response time of LC cells can constrain system performance. This paper addresses the slow phase-shifting limitation of LC by developing a physics-based model for the time response of an LC unit cell and proposing a novel phase-shift design framework to reduce the transition time. Specifically, exploiting the fact that LC-RIS at milimeter wave bands have a large electric aperture, we optimize the LC phase shifts based on user locations, eliminating the need for full channel state information and minimizing reconfiguration overhead. Moreover, instead of focusing on a single point, the RIS phase shifters are designed to optimize coverage over an area. This enhances communication reliability for mobile users and mitigates performance degradation due to user location estimation errors. The proposed design minimizes the transition time between configurations, a critical requirement for TDMA schemes. Our analysis reveals that the impact of RIS reconfiguration time on system throughput becomes particularly significant when TDMA intervals are comparable to the reconfiguration time. In such scenarios, optimizing the phase-shift design helps mitigate performance degradation while ensuring specific QoS requirements. Moreover, the proposed algorithm has been tested through experimental evaluations, which demonstrate that it also performs effectively in practice.