Karim Abed-Meraim

This paper proposes a novel semi-blind structured subspace-based channel estimation framework for wideband reconfigurable intelligent surface (RIS)-assisted millimeter-wave (mmWave) multi-user systems with direct transmission paths. The proposed method exploits the inherent Toeplitz structure of wideband channels induced by the convolutional nature of wireless signal propagation and incorporates a limited number of pilot symbols to construct a semi-blind estimation criterion. Specifically, a hybrid cost function is formulated by integrating a signal subspace constraint derived from the received signal covariance matrix with a least-squares pilot-based component using limited pilot observations. This formulation enables the proposed approach to effectively exploit both the statistical properties of the received signals and the limited pilot information, thereby improving estimation reliability while significantly reducing pilot overhead. The structural properties of the channel and the Kronecker product representation of the system model are leveraged to provide a closed-form solution through the minimization of the resulting quadratic optimization problem. Simulation results demonstrate that the proposed method achieves high channel estimation accuracy with substantially fewer pilot symbols compared to conventional pilot-based approaches, highlighting its potential for efficient and practical implementation in wideband RIS-assisted mmWave communication systems.

This paper proposes a semi-blind approach for near-field channel estimation and 3D source localization for Extremely Large Antenna Arrays. It considers Uniform Planar Arrays and Cross Arrays as particular cases. Unlike conventional Least-Squares estimators relying solely on predefined pilots, the proposed method exploits unknown data symbols within a subspace framework. Since it does not require any prior knowledge of the propagation model, our proposal achieves improved accuracy and robustness. To reduce computational complexity, a Divide-and-Conquer strategy is adopted, enabling fast and parallelizable subspace estimation. In a multipath single-user scenario with Inter-Symbol Interference, localization parameters (angles and range) are recovered from the first channel tap via a low-complexity least-squares fitting, assuming a line-of-sight path. The extension to Cross Arrays provides a simple specialization of the Uniform Plannar Array approach while drastically reducing the computational load. Simulation results confirm that the proposed method achieves accurate channel estimation and 3D localization with substantial complexity reduction as compared to the state of the art.

The performance gains offered by reconfigurable intelligent surfaces (RIS) in wireless networks critically depend on the availability and accuracy of channel information. Channel estimation becomes particularly challenging when the RIS comprises passive elements without sensing or processing capabilities, as the channels cannot be measured directly and must instead be inferred indirectly by an external, non-colocated controller. In this work, we propose a deep learning-based approach for composite channel estimation in RIS-assisted communication systems. Unlike traditional methods that estimate the direct and reflected channels separately, our approach integrates the effects of both paths into a unified model and employs deep learning to capture the complex nonlinear interactions introduced by RIS phase shifts. This method reduces pilot overhead and enhances the efficiency of channel estimation. Simulation results reveal that the proposed method significantly outperforms other existing methods in RIS-assisted communication systems.

The next generation of wireless communication systems is poised to integrate advanced technologies such as holographic multiple input, multiple output (HMIMO) systems and stacked intelligent metasurfaces (SIMs). These innovations aim to provide dynamic electromagnetic wave control while reducing hardware complexity. In SIM-assisted systems, the number of meta-atoms per SIM layer often exceeds the number of antennas at the base station (BS), making accurate channel estimation a significant challenge. Traditional model-based channel estimation methods often struggle with the high dimensionality and dynamic variability of such environments. To address this hurdle, we propose a deep learning (DL)-based uplink channel estimation framework that learns complex, nonlinear relationships between the received pilots and the channel responses. The proposed method significantly outperforms classical estimators in terms of accuracy and robustness. The simulation results validate its effectiveness under various propagation scenarios.

Maximizing the performance of reconfigurable intelligent surfaces (RIS) in wireless communication systems critically depends on accurate channel estimation. This challenge becomes more pronounced in the presence of double RISs, particularly in severely obstructed environments, where the number of channel coefficients increases substantially and more pilot overhead is required. To address these limitations, we propose a deep learningbased two-stage dual-reflection cascaded channel estimation for multi users multiple-input multiple-output (MIMO) system. In the first stage, the dual-reflection channel of a reference user is estimated and in the second stage, the channels of the remaining users are estimated using the reference estimate. The proposed approach operates the first RIS in a known fixed mode, while the second RIS in a dynamically adjustable mode, reducing the number of pilots and channel coefficients to be estimated. Simulation results demonstrate that the proposed deep learning method significantly outperforms existing methods as far as estimation accuracy is concerned especially in low SNR environments.