Subject: Cutting-Edge Advancements in Wireless Communications

Hi Elman,

General Overview

This newsletter explores cutting-edge advancements in wireless communications, focusing on novel antenna technologies, signal processing techniques, and the application of machine learning. Several papers delve into the burgeoning field of reconfigurable intelligent surfaces (RIS), highlighting their potential to revolutionize wireless systems.

Sheemar et al. (2025a) introduce the concept of multi-user holographic UAV communications, proposing joint optimization of hybrid holographic beamforming and 3D UAV positioning to maximize sum rate. Building on this, Sheemar et al. (2025b) tackle the problem of joint hybrid holographic beamforming and user scheduling under per-user QoS constraints, developing an iterative optimization framework that manages beamforming patterns and user scheduling while ensuring minimum QoS requirements. Salem et al. (2025) introduce the concept of fluid RIS (FRIS), where each element has both position and phase shift reconfigurability. They demonstrate significant rate improvements over conventional RIS in both single-user and multi-user scenarios using particle swarm optimization (PSO) for element positioning. Peng et al. (2025) investigate joint size and placement optimization for hybrid active-passive IRS (HIRS), considering various deployment schemes and characterizing system capacity scaling laws. Finally, Zhi et al. (2025) explore holographic MIMO (HMIMO) in multi-cell systems, developing low-complexity algorithms for precoder design and HMIMO element tuning under practical power constraints.

Beyond RIS, several papers explore other innovative antenna technologies and their applications. Amhaz et al. (2025) investigate movable antenna (MA) technology in CoMP-RSMA systems, proposing a gradient-based meta-learning algorithm for joint optimization of beamforming, stream allocation, and MA positioning. Vummadisetty et al. (2025) present a metasurface-enhanced patch antenna array for Ku-band satellite reception, targeting low-cost user terminals. Vasquez-Peralvo et al. (2025) address uneven power distribution in multi-beam satellite systems, proposing a genetic algorithm (GA) to optimize beamforming coefficients while ensuring uniform antenna element activation. Mu et al. (2025) introduce a pinching-antenna system (PASS) for multicast communications, optimizing PA positions for performance maximization. Zhao et al. (2025) further develop this concept, proposing waveguide division multiple access (WDMA) for PASS and developing algorithms for joint power allocation and pinching beamforming.

Machine learning plays a prominent role in several contributions, addressing challenges in channel estimation, signal processing, and various applications. Liu et al. (2025) introduce SpikACom, a neuromorphic computing framework based on spiking neural networks (SNNs) for green communications. Faisal et al. (2025) propose a conditional generative adversarial network (CGAN) for channel estimation in RIS-assisted ISAC systems. Cano et al. (2025) introduce AIRIS2, a deep learning algorithm for predicting rain events and enabling efficient gateway switching in satellite systems. Wang et al. (2025) propose a sensing-assisted channel estimation scheme for OFDM ISAC systems, developing a robust LMMSE estimation algorithm. Kozlenko et al. (2025) investigate software-defined demodulation of M-FSK using a dense neural network for weak signal communications.

Several papers focus on specific applications and signal processing techniques. Yapici et al. (2025) present a novel noise modulation scheme for simultaneous energy harvesting and communication. Michelena et al. (2025) investigate the recovery of mixture parameters of spike signals. Segman et al. (2025) propose a structure-aware matrix pencil method for detecting and estimating parameters of complex exponentials. Peng et al. (2025) investigate latency-aware resource allocation for integrated communications, computation, and sensing. Hou et al. (2025) present a prototype design for a 220 GHz RIS-aided multi-user THz communication system. Wang et al. (2025) propose a 2-bit wideband 5G mmWave RIS. Gavras et al. (2025) develop a CRB minimizing approach for near-field localization with dynamic metasurface antennas.

Paper Highlights

Movable Antennas: A Paradigm Shift in Wireless Networks

A Tutorial on Movable Antennas for Wireless Networks by Lipeng Zhu, Wenyan Ma, Weidong Mei, Yong Zeng, Qingqing Wu, Boyu Ning, Zhenyu Xiao, Xiaodan Shao, Jun Zhang, Rui Zhang https://arxiv.org/abs/2502.17905

Caption: This illustration showcases the diverse applications of Movable Antennas (MAs) across various communication scenarios, including satellite, airborne, maritime, and terrestrial networks, as well as in integrated sensing and communication (ISAC) systems. Different MA architectures, such as rotatable, sliding, and foldable arrays, are depicted, highlighting their adaptability to different environments and use cases. The image emphasizes the transformative potential of MAs in optimizing wireless performance through dynamic antenna positioning and orientation.

Movable antennas (MAs) present a dynamic shift from traditional fixed antennas (FAs) in wireless communications and sensing. Their ability to adjust position and/or orientation optimizes channel conditions and enhances system performance. Central to MA systems is the field-response channel model, h(t, r) = f(r)^HΣg(t), which describes the channel's spatial variation as a function of transmitter (Tx) and receiver (Rx) antenna positions (t and r respectively). The model encompasses MIMO systems, wideband scenarios, near-field conditions, and even six-dimensional MA (6DMA) systems, incorporating 3D position and 3D orientation. The tutorial also covers various MA architectures, from mechanically movable elements to liquid-based and electronically reconfigurable antennas, each with its own advantages and limitations.

Two key design challenges for MA systems are antenna movement optimization and channel acquisition. The former aims to maximize utility functions like achievable rate or sensing accuracy by jointly optimizing antenna positions and communication/sensing resources. This is computationally complex due to the non-linear relationship between antenna position and channel gain. The tutorial presents algorithms for various system setups (SISO, MISO/SIMO, MIMO, multiuser) and applications. Channel acquisition, on the other hand, focuses on reconstructing the channel map between arbitrary antenna positions in the Tx and Rx regions. This involves model-based methods leveraging the field-response channel model and exploiting channel sparsity, or model-free methods based on measurements and interpolation/extrapolation techniques.

Experimental results from MA prototypes showcase significant performance improvements. In single-MA systems, measured signal power variations reached over 40 dB at 3.5 GHz and 23 dB at 27.5 GHz, demonstrating potential for signal strength enhancement and interference mitigation. MA-aided MIMO systems achieve higher capacity than FPA systems, especially at high SNR. MA-aided sensing also shows improved angular resolution and target AoA estimation accuracy. The tutorial extends MA technology to broader applications, including space-air-ground integrated networks, edge computing, security, and wireless power transfer, also exploring its synergy with intelligent reflecting surfaces (IRSs). Future research directions include robust MA position optimization under imperfect CSI, near-field channel optimization, movement overhead management, and efficient channel acquisition algorithms for complex setups. New standardization frameworks and protocols are crucial for MA integration in future networks.

Optimizing BD-RIS Architectures for Multiuser MIMO

Beyond Diagonal RIS in Multiuser MIMO: Graph Theoretic Modeling and Optimal Architectures with Low Complexity by Zheyu Wu, Bruno Clerckx https://arxiv.org/abs/2502.16509

Caption: Transmit power vs. circuit complexity for different BD-RIS architectures, demonstrating the superior performance-complexity trade-off of band- and stem-connected RIS.

Reconfigurable intelligent surfaces (RIS) are transforming wireless communications through wave-domain beamforming. Beyond Diagonal RIS (BD-RIS) introduces tunable impedances connecting RIS elements, offering greater flexibility than conventional diagonal RIS. However, the increased interconnections in fully-connected BD-RIS result in a prohibitive O(N₁²) circuit complexity. This paper focuses on finding optimal BD-RIS architectures that match the performance of fully-connected RIS while minimizing complexity in multiuser MIMO systems.

The authors use graph theory to model BD-RIS architectures, representing interconnections as edges in a graph, with the adjacency matrix capturing the topological connectivity. A general utility optimization problem encompassing various metrics like sum-rate maximization, transmit power minimization, and energy efficiency maximization is formulated.

A key result is a sufficient condition on the adjacency matrix guaranteeing that the corresponding BD-RIS architecture achieves the same optimal performance as fully-connected RIS. If a permutation matrix P exists such that A_G = PĀ_GP^T, where Ā_G meets a specific condition related to the degree of freedom (DoF) D of the multiuser MIMO channel (D = min{∑^K_k=1N_k, N_T}), then the BD-RIS architecture is optimal. The circuit complexity of this architecture class becomes O(N₁min{D, N₁/2}). As D is typically much smaller than N₁, this represents a significant complexity reduction.

The paper introduces two novel architectures: band-connected RIS and stem-connected RIS. When the band/stem width is 2min{D, N₁/2} - 1, both architectures fall within the optimal class. Simulations validate these findings, showing that the proposed architectures achieve the same performance as fully-connected RIS with significantly lower complexity, highlighting their superior performance-complexity trade-off compared to group-connected RIS and conventional RIS in multiuser MIMO scenarios.

Holographic UAVs Take Flight: Joint Beamforming and 3D Positioning for Enhanced Aerial Communications

Joint Beamforming and 3D Location Optimization for Multi-User Holographic UAV Communications by Chandan Kumar Sheemar, Asad Mahmood, Christo Kurisummoottil Thomas, George C. Alexandropoulos, Jorge Querol, Symeon Chatzinotas, Walid Saad https://arxiv.org/abs/2502.17428

Caption: This figure illustrates a UAV equipped with a Reconfigurable Holographic Surface (RHS) transmitting to multiple ground users. The RHS, controlled by digital beamforming, RF chains, and a waveguide, dynamically adjusts the holographic beamforming patterns to maximize the sum rate of the network. This hybrid holographic transceiver enables the UAV to optimize its 3D position and beamforming simultaneously for enhanced aerial communications.

This paper introduces a novel approach to multi-user holographic UAV communications, focusing on the joint optimization of hybrid holographic beamforming and 3D UAV positioning. Instead of conventional phased array antennas, the UAV utilizes a reconfigurable holographic surface (RHS), allowing dynamic adjustments to both position and beamforming to maximize the network's sum rate. This addresses limitations of phased arrays like bulk, complexity, and power consumption, while capitalizing on RHS advantages such as compact size, energy efficiency, and beamforming flexibility.

The proposed framework tackles a complex optimization problem constrained by the UAV's operational area, altitude limits, transmit power, and holographic beamforming weights. An iterative alternating optimization algorithm decomposes the task into three subproblems: digital beamforming optimization, holographic beamforming optimization, and 3D UAV position optimization. Zero-forcing is employed for digital beamforming, while gradient ascent optimizes holographic patterns and 3D UAV position. The 3D position optimization incorporates a derived gradient considering variations in elevation and azimuth angles relative to each user and path loss, ensuring precise UAV movement updates.

Simulations demonstrate significant performance gains compared to a benchmark with random holographic beamforming amplitudes. The proposed method consistently outperforms the benchmark across various SNR levels, with the performance gap widening at higher SNRs. Larger holographic surface sizes also yield higher sum rates, emphasizing the importance of RHS size in enhancing holographic beamforming resolution. Analysis of UAV trajectories reveals the algorithm's adaptability to user distributions, dynamically adjusting the UAV's position to optimize coverage and maximize the sum rate.

The impact of SNR on system performance is also explored. At low SNRs, the system faces interference management challenges, especially with smaller RHS sizes, but the proposed algorithm exhibits resilience. At high SNRs, improved interference suppression benefits both methods, with the proposed algorithm maintaining its superior performance. These findings underscore the crucial role of RHS size in enhancing holographic beamforming and improving system performance, particularly in challenging multi-user and low-SNR environments.

Conclusion

This newsletter highlights significant advancements in wireless communication technologies, with a particular focus on the transformative potential of reconfigurable intelligent surfaces (RIS), holographic MIMO (HMIMO), and movable antennas (MA). The convergence of these antenna technologies with sophisticated signal processing and machine learning techniques paves the way for higher data rates, improved spectral efficiency, and enhanced robustness in future wireless networks. The development of novel algorithms and prototype implementations, as exemplified by the holographic UAV communication system and the exploration of optimal BD-RIS architectures, demonstrates the practical viability of these advancements. Furthermore, the application of these technologies to address specific challenges in areas like satellite communications, THz communications, and integrated sensing and communication (ISAC) underscores their versatility and potential to drive innovation across diverse domains. The ongoing research in these areas promises to reshape the landscape of wireless communications in the years to come.