This paper investigates a novel pinching antenna assisted multi-target integrated sensing and communication (ISAC) system. We propose a joint optimization of transmit and receive PA positions to enable the simultaneous service of multiple sensing targets and a downlink communication user. The objective of the formulated optimization problem is to maximize the minimum sensing signal-to-interference-plus-noise ratio (SINR), which addresses inter-target interference and ensures fairness across multiple targets. The design jointly considers both the transmit and receive antenna positions as well as the receive beamformer. An efficient alternating optimization framework is introduced to handle this non-convex problem. The receive beamformer is obtained in closed-form using the minimum variance distortionless response method, while the antenna position optimization is solved via the differential evolution (DE) algorithm. The DE search process is guided by a customized fitness function that incorporates both communication quality-of-service requirements and physical constraints. Numerical results demonstrate that the proposed joint optimization scheme substantially outperforms transmitter-only optimization benchmarks, receiver-only optimization, and fixed antenna placements. Furthermore, comparative studies show that the DE algorithm achieves a better trade-off between performance and complexity, it not only surpasses particle swarm optimization in terms of performance, but also outperforms element-wise search when the antenna number is small, while maintaining lower computational complexity than both alternatives.

A model-driven deep learning framework is proposed for channel estimation in Rydberg atomic quantum receivers (RAQRs) based on the measurement of holographic snapshots. Specifically, we develop a Transformer-based unrolling architecture, termed URformer, to solve the non-linear biased phase retrieval problem, which is derived by unrolling a stabilized variant of the expectation-maximization Gerchberg-Saxton (EM-GS) algorithm. Each layer of the proposed URformer incorporates three trainable modules: 1) a learnable filter network that replaces the fixed Bessel kernel in the classic EM-GS algorithm; 2) a trainable gating mechanism that adaptively combines classic updates to ensure training stability; and 3) an efficient channel Transformer module that learns to correct residual errors by capturing non-local channel dependencies. Numerical results demonstrate that the proposed URformer significantly outperforms classic iterative algorithms and conventional black-box neural networks with less pilot overhead.

The non-orthogonal superimposed pilot (NOSIP) scheme significantly improves the spectrum efficiency of semantic communication (SemCom) systems by effectively reducing pilot overhead. However, existing SemCom systems based on NOSIP still face challenges, including strong model-channel coupling and limited adaptability to heterogeneous channels. To address these issues, this paper proposes a flexible, channel-adaptive digital SemCom (D-SemCom) architecture based on the NOSIP scheme. Specifically, we design a lightweight semantic codec, termed ShiftViT, and a semantic receiver, termed ShiftRx, which employ time- and frequency-domain shift mechanisms to decouple pilot and data and suppress multi-user interference, thereby enabling image transmission under complex channel conditions. Furthermore, a lightweight channel adaptation algorithm based on first-order meta-learning is proposed to facilitate rapid adaptation and mitigate the strong coupling between semantic models and channel environments. Numerical results demonstrate that the proposed D-SemCom system achieves approximately 25.14% and 1.16% goodput improvements over traditional receivers and existing state-of-the-art methods, respectively, while reducing the computational complexity in terms of FLOPs by approximately 40.37%. In addition, the proposed channel adaptation algorithm is shown to rapidly adapt to diverse channel scenarios within the D-SemCom system.

We investigate a multi-layer data transmission scheme with superimposed pilots (SIPs) to enhance the throughput of multiple-input multiple-output orthogonal frequency-division multiplexing systems. However, in multi-layer data transmission scenarios, signal coupling between different antennas and layers causes severe interference issues, posing significant challenges for receiver design. To address this issue, we propose a deep learning-based receiver architecture, named SANet, which leverages the parallel processing capabilities of the multi-head self-attention (MHSA) mechanism. Specifically, each head of the MHSA mechanism is used to extract local features from each layer of the received signal, enabling the separation and reception of multi-layer bitstream information. Additionally, a flexible and diverse data augmentation strategy is designed to enhance the generalization capability of the deep receiver. Numerical results show that, compared to traditional schemes, the proposed SANet with orthogonal pilots can improve throughput by 7.01%, while the proposed SANet with SIPs can improve throughput by 37.15%.

Autonomous aerial vehicles (AAVs) have emerged as the potential aerial base stations (BSs) to improve terrestrial communications. However, the limited onboard energy and antenna power of a AAV restrict its communication range and transmission capability. To address these limitations, this work employs collaborative beamforming through a AAV-enabled virtual antenna array to improve transmission performance from the AAV to terrestrial mobile users, under interference from non-associated BSs and dynamic channel conditions. Specifically, we introduce a memory-based random walk model to more accurately depict the mobility patterns of terrestrial mobile users. Following this, we formulate a multi-objective optimization problem (MOP) focused on maximizing the transmission rate while minimizing the flight energy consumption of the AAV swarm. Given the NP-hard nature of the formulated MOP and the highly dynamic environment, we transform this problem into a multi-objective Markov decision process and propose an improved evolutionary multi-objective reinforcement learning algorithm. Specifically, this algorithm introduces an evolutionary learning approach to obtain the approximate Pareto set for the formulated MOP. Moreover, the algorithm incorporates a long short-term memory network and hyper-sphere-based task selection method to discern the movement patterns of terrestrial mobile users and improve the diversity of the obtained Pareto set. Simulation results demonstrate that the proposed method effectively generates a diverse range of non-dominated policies and outperforms existing methods. Additional simulations demonstrate the scalability and robustness of the proposed CB-based method under different system parameters and various unexpected circumstances.

A joint cascaded channel estimation scheme is proposed for simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) systems with hardware imperfections. In particular, the practical hybrid near- and far-field electromagnetic radiation with spatial non-stationarity is investigated. By exploiting the cascaded channel correlations between different users and between different STAR-RIS elements, a multi-task learning (MTL)-based channel estimation framework is proposed. This framework is capable of estimating the cascaded channels for transmission and reflection simultaneously based on noisy observations of the mixture channel. Following the design guideline of the proposed MTL framework, an efficient multi-task network (MTN) is developed to reconstruct the high-dimensional channels with limited pilot overhead. In the proposed MTN architecture, a mixed convolution and multilayer perception module is exploited to capture the effective hybrid-field channel features. This module integrates the locality bias modeling of the channel-wise convolution and the long-range dependency modeling of MLP, which finely learns both local spatial correlations and specific spatial non-stationarity of the hybrid-field cascaded channels. Numerical results show that the proposed MTN achieves superior channel estimation accuracy with less training overhead compared with the existing state-of-the-art benchmarks, in terms of required pilots, computations, and network parameters.

A near-field wideband beamforming scheme is investigated for reconfigurable intelligent surface (RIS) assisted multiple-input multiple-output (MIMO) systems, in which a deep learning-based end-to-end (E2E) optimization framework is proposed to maximize the system spectral efficiency. To deal with the near-field double beam split effect, the base station is equipped with frequency-dependent hybrid precoding architecture by introducing sub-connected true time delay (TTD) units, while two specific RIS architectures, namely true time delay-based RIS (TTD-RIS) and virtual subarray-based RIS (SA-RIS), are exploited to realize the frequency-dependent passive beamforming at the RIS. Furthermore, the efficient E2E beamforming models without explicit channel state information are proposed, which jointly exploits the uplink channel training module and the downlink wideband beamforming module. In the proposed network architecture of the E2E models, the classical communication signal processing methods, i.e., polarized filtering and sparsity transform, are leveraged to develop a signal-guided beamforming network. Numerical results show that the proposed E2E models have superior beamforming performance and robustness to conventional beamforming benchmarks. Furthermore, the tradeoff between the beamforming gain and the hardware complexity is investigated for different frequency-dependent RIS architectures, in which the TTD-RIS can achieve better spectral efficiency than the SA-RIS while requiring additional energy consumption and hardware cost.

A multi-scale attention based channel estimation framework is proposed for reconfigurable intelligent surface (RIS) aided massive multiple-input multiple-output systems, in which hardware imperfections and time-varying characteristics of the cascaded channel are investigated. By exploiting the spatial correlations of different scales in the RIS reflection element domain, we construct a Laplacian pyramid attention network (LPAN) to realize the high-dimensional cascaded channel reconstruction with limited pilot overhead. In LPAN, we leverage the multi-scale supervision learning to progressively capture the spatial correlations of the cascaded channel, where the attention mechanism based dual-branch architecture is designed. To balance network performance and complexity of LPAN, we further propose a lightweight LPAN-L architecture. In LPAN-L, the partial standard convolutional layers are decomposed into the group convolution, dilated convolution and point-wise convolution, which forms a sparse convolutional filter set to extract the channel feature with less computation cost. Furthermore, we leverage parameter sharing and recursion strategy to reduce the space complexity. Moreover, a selective fine-tuning strategy is developed to realize the domain adaption. Simulation results show that the proposed LPAN can achieve higher estimation accuracy than the existing estimation schemes, while the LPAN-L architecture with a close performance to LPAN efficiently reduces the network complexity1.

A joint cascaded channel estimation framework is proposed for simultaneously transmitting and reflecting recon-figurable intelligent surfaces (STAR-RIS) systems with hardware imperfection, in which practical the hybrid-field electromagnetic wave radiation with spatial non-stationarity is investigated. By exploiting the cascaded channel correlations in user domain and STAR-RIS element domain, we propose a multitask network (MTN) with multi-expert branches to simultaneously reconstruct the high-dimensional transmitting and reflecting channels from the observed mixture channel with noise. In the proposed MTN architecture, a learnable shrinkage module is exploited to constrict the communication noise, and self-attention mechanism-based Transformer layers are utilized to extract the nonlocal feature of the non-stationary cascaded channel. Numerical results show that the proposed MTN achieves superior channel estimation accuracy with less training overhead compared with existing state-of-the-art benchmarks, in terms of required pilots, computations, and network parameters.

Motivated by the development of single image super-resolution (SR) reconstruction in computer version, classic SR networks have been widely applied to the channel estimation of wireless communication system. To capture the spatial correlations in the reflection element-domain of reconfigurable intelligent surface (RIS), we propose a multi-scale supervised learning-based Laplacian pyramid wide residual network (LapWRes) to achieve the progressive reconstruction of cascaded channel in a coarse-to-fine fashion. The LapWRes can be divided vertically into feature extraction branch (FEB) and channel reconstruction branch (CRB), while it can also be viewed horizontally as multiple channel reconstruction modules (RMs) at different scales. In the FEB, the wide activation residual blocks are stacked to extract the high-frequency information of cascaded channel. In the CRB, the high-frequency and low-frequency information of cascaded channel is fused by utilizing the residual learning. Simulation results show that the LapWRes can achieve better estimation accuracy than other channel estimation schemes and faster convergence than existing SR network-based channel estimation models.