Extremely large-scale antenna arrays (XL-arrays) and ultra-high frequencies are two fundamental technologies for future sixth-generation (6G) wireless networks, providing enhanced system capacity and substantial bandwidth expansion. To fully leverage these technological advancements, conventional far-field models must be replaced by more accurate near-field spherical-wave propagation models. This paper investigates a near-field communication system comprising a hybrid analog-digital beamforming base station (BS) and multiple mobile users, aiming to maximize system sum-rate through optimized codebook design, beam selection, and digital precoding. To accommodate dynamic user distributions, we propose two model-agnostic meta-learning (MAML)-based frameworks that enable prompt adaptation by learning well-initialized models for fine tuning. The first framework integrates the MAML method with a deep neural network (DNN) to design near-field codebooks tailored to the user distributions, addressing the limitations of conventional uniform codebooks. The second framework employs a joint neural network (NN) for beam selection and digital precoding, combining deep reinforcement learning (DRL) and deep unfolding. The DRL NN formulates beam selection as a Markov Decision Process, while the deep-unfolding NN approximates optimal digital precoding through a lightweight iterative algorithm without matrix inversion. Simulation results show that the proposed frameworks significantly outperform conventional methods, achieving superior generalization and overall performance in dynamic near-field scenarios.

Simultaneously transmitting and reflecting reconfigurable intelligent surfaces (STAR-RIS) can enable full spatial coverage. In this paper, we investigate an artificial intelligence (AI)-empowered STAR-RIS-assisted multi-user communication system. We aim to maximize the system sum-rate by jointly optimizing the mode selection matrix of the STAR-RIS elements, the passive beamforming matrix of the STAR-RIS, and the beamforming matrix of the base station (BS). Due to the mixed-timescale structure, we propose a joint neural network (NN) design consisting of an advantage pointer-critic (APC) NN to optimize the discrete variables in the mode selection matrix, a fully connected NN, and a deep-unfolding NN to optimize the beamforming matrices. Specifically, the element mode selection problem is formulated as a Markov decision process (MDP), and an APC NN is carefully designed to solve it. The passive beamforming matrix of the STAR-RIS is optimized by employing a fully connected NN. Then, we apply an iterative weighted minimum mean-square error (WMMSE)-based deep-unfolding NN for the BS beamforming design. Simulation results verify that our jointly trained NN can outperform the conventional algorithms with reduced overhead.

Integrated sensing and communication (ISAC) has recently been considered as a promising approach to save spectrum resources and reduce hardware cost. Meanwhile, as information security becomes increasingly more critical issue, government agencies urgently need to legitimately monitor suspicious communications via proactive eavesdropping. Thus, in this paper, we investigate a wireless legitimate surveillance system with radar function. We seek to jointly optimize the receive and transmit beamforming vectors to maximize the eavesdropping success probability which is transformed into the difference of signal-to-interference-plus-noise ratios (SINRs) subject to the performance requirements of radar and surveillance. The formulated problem is challenging to solve. By employing the Rayleigh quotient and fully exploiting the structure of the problem, we apply the divide-and-conquer principle to divide the formulated problem into two subproblems for two different cases. For the first case, we aim at minimizing the total transmit power, and for the second case we focus on maximizing the jamming power. For both subproblems, with the aid of orthogonal decomposition, we obtain the optimal solution of the receive and transmit beamforming vectors in closed-form. Performance analysis and discussion of some insightful results are also carried out. Finally, extensive simulation results demonstrate the effectiveness of our proposed algorithm in terms of eavesdropping success probability.

The time of arrival (TOA)-based localization techniques, which need to estimate the delay of the line-of-sight (LoS) path, have been widely employed in location-aware networks. To achieve a high-accuracy delay estimation, a number of multiband-based algorithms have been proposed recently, which exploit the channel state information (CSI) measurements over multiple non-contiguous frequency bands. However, to the best of our knowledge, there still lacks an efficient scheme that fully exploits the multiband gains when the phase distortion factors caused by hardware imperfections are considered, due to that the associated multi-parameter estimation problem contains many local optimums and the existing algorithms can easily get stuck in a “bad” local optimum. To address these issues, we propose a novel two-stage global estimation (TSGE) scheme for multiband delay estimation. In the coarse stage, we exploit the group sparsity structure of the multiband channel and propose a Turbo Bayesian inference (Turbo-BI) algorithm to achieve a good initial delay estimation based on a coarse signal model, which is transformed from the original multiband signal model by absorbing the carrier frequency terms. The estimation problem derived from the coarse signal model contains fewer local optimums and thus a more stable estimation can be achieved than directly using the original signal model. Then in the refined stage, with the help of coarse estimation results to narrow down the search range, we perform a global delay estimation using a particle swarm optimization-least square (PSO-LS) algorithm based on a refined multiband signal model to exploit the multiband gains to further improve the estimation accuracy. Simulation results show that the proposed TSGE significantly outperforms the benchmarks with comparative computational complexity.

In location-aware networks, the time of arrival (TOA)-based localization techniques have been widely employed. To achieve high-accuracy delay estimation, a number of multiband-based algorithms have been proposed recently, which exploit the channel state information (CSI) measurements over multiple non-contiguous frequency bands. However, to our best knowledge, there still lacks an efficient scheme that fully exploits the multiband gains when phase distortion factors are considered, due to that the associated multi-parameter estimation problem contains many local optimums and the existing algorithms can easily get stuck in a “bad” local optimum. To address these issues, we propose a novel two-stage global estimation (TSGE) scheme for multiband delay estimation. In the coarse stage, we propose a weighted multiple signal classification (MUSIC) algorithm to achieve an initial delay estimation based on a coarse signal model. The estimation problem derived from the coarse signal model contains less local optimums and thus a more stable estimation can be achieved than directly using the original signal model. Then in the refined stage, with the help of coarse estimation results to narrow down the search range, we perform a global delay estimation using a particle swarm optimization (PSO) algorithm based on a refined multiband signal model to exploit the multiband gains to further improve the estimation accuracy. Simulation results show that the proposed TSGE significantly outperforms the benchmarks.