Over-the-air computation (AirComp) enables swift wireless data aggregation by leveraging the superposition property of multiple-access channels (MAC), making it essential for the seamless integration of communication and computing in future networks. Meanwhile, fluid antenna systems (FAS) offer dynamic spatial degrees of freedom (DoF) by reconfiguring antenna positions, thus enhancing adaptability under varying channel conditions. This paper investigates the integration of FAS into a communication and AirComp coexistence framework. We aim to jointly optimize the transceiver beamforming vectors and the antenna positioning vector (APV) to minimize the computation distortion while ensuring reliable cellular communication performance. To tackle this highly non-convex problem, we develop an efficient joint learning-optimization framework. Specifically, we propose a neural network (NN) framework with a dedicated surrogate loss function design to infer optimal APV based on multi-path channel conditions, while an alternating optimization (AO) method is developed to find a locally optimal solution of transceivers by iteratively optimizing each variables with the others being fixed. Besides, to provide analytical tractability and benchmark insight, the APV design problem is relaxed and transformed into a tractable quadratically constrained quadratic program (QCQP) by successive convex approximation (SCA) as a special case under line-of-sight (LoS) channels, which reveals the performance bounds and convergence properties of the system. Numerical results show that proposed method significantly improves the system performance compared with traditional fixed-position antenna (FPA) as well as various benchmark schemes with remarkable generalization capabilities across diverse channel conditions.

Extremely large aperture array (ELAA) represents a paradigm shift in wireless antenna systems, poised to redefine the capabilities of future networks, particularly in the era of the sixth generation (6G) networks. The integration of multiple-input multiple-output (MIMO) technology with an unprecedented scale of antenna arrays enables transformative potential for enhancing spectral efficiency, increasing coverage, and enabling new applications. This paper provides a comprehensive overview of ELAA, delving into fundamental concepts, state-of-the-art technologies, and practical applications. We commence by presenting the fundamentals of ELAA technology, focusing on the various architectural categories and two fundamental properties of communication in the near-field region of ELAA, namely spherical wave propagation and channel spatial non-stationarity. We then illustrate the phenomenon of finite beam depth in the near field before presenting general distance boundaries based on various criteria and conducting a comprehensive performance analysis of this technology. Subsequently, in light of the distinctive electromagnetic characteristics of ELAA, we will examine the practical challenges that have emerged, including channel estimation, beamforming design, and the practical hardware issues that have arisen. Subsequently, we examine the diverse applications of ELAA across various domains, emphasizing its transformative potential in fields such as physical layer security, communication and sensing, and wireless power and information transfer. Finally, the paper concludes by outlining several promising avenues for future research and exploration within the realm of ELAA. These avenues are identified as areas ripe for investigation and innovation.

Over-the-air computation (AirComp) has emerged as a promising technology for fast wireless data aggregation by harnessing the superposition property of wireless multiple-access channels. This letter investigates a fluid antenna (FA) array-enhanced AirComp system, employing the new degrees of freedom introduced by antenna movements. Specifically, we jointly optimize the transceiver design and antenna position vector (APV) to minimize the mean squared error (MSE) between target and estimated function values. To tackle the resulting highly non-convex problem, we adopt an alternating optimization technique to decompose it into three subproblems. These subproblems are then iteratively solved until convergence, leading to a locally optimal solution. Numerical results show that FA arrays with the proposed transceiver and APV design significantly outperform the traditional fixed-position antenna arrays in terms of MSE.

Over-the-air computation (AirComp) stands as a pioneering technology, offering swift wireless data aggregation by leveraging the superposition property inherent in wireless multiple-access channels. In future communication networks, the coexistence of communication and computing is inevitable, given the widespread adoption of wireless data aggregation. Meanwhile, fluid antenna (FA) has recently been considered as a promising technology to further enhance the system performance by exploiting the new degrees of freedom (DoFs) via antennas position movement. In this paper, a FA enhanced integrated communication and computation framework is proposed. Specifically, we undertake joint optimization of transceiver design and antenna position vector (APV) to minimize AirComp mean squared error (MSE) while maintaining the achievable rate of communication above a specified threshold. Numerical results shows superiority of FA arrays over conventional fixed-position antenna arrays (FPA).