Human Action Recognition using WiFi Channel State Information (CSI) has emerged as an attractive alternative to vision-based methods due to its ubiquity, device-agnostic nature, and inherent privacy-preserving capabilities. However, the high cost of manual annotation and the limited scale of publicly available CSI datasets restrict the performance of supervised approaches. Self-supervised learning (SSL) offers a promising avenue, but existing contrastive paradigms rely on data augmentations that conflict with the physical semantics of radio signals and require large-batch training, making them poorly suited for CSI. To overcome these challenges, we introduce CIG-MAE—a Cross-Modal Information-Guided Masked Autoencoder—that reconstructs both the amplitude and phase of CSI using a symmetric dual-stream architecture with a high masking ratio. Specifically, we propose an Adaptive Information-Guided Masking strategy that dynamically allocates attention to time–frequency regions with high information density to improve learning efficiency, and we incorporate a Barlow Twins regularizer to align cross-modal representations without negative samples. Experiments on three public datasets show that CIG-MAE consistently outperforms SOTA SSL methods and even surpasses a fully supervised baseline, demonstrating superior data efficiency, robustness, and representation generalization.

Due to the increasing number of users and antennas in extremely large antenna arrays (ELAA) based integrated sensing and communication (ISAC) systems, the complexity of beamforming optimization becomes overwhelming, which impedes real-time and cost-efficient ISAC deployment in practice. Specifically, a general ISAC system where the base station (BS) communicates with multiple users and performs target detection is considered. Then, a sum communication rate maximization problem is formulated, subjected to the constraints of transmit power and the minimum sensing rates of users. To solve this problem, we develop a framework that leverages deep learning algorithms to provide a low complexity and transferable (LCT) solution for ISAC beamforming. The proposed LCT beamforming optimization framework includes three modules: 1) an unsupervised learning based feature extraction algorithm is proposed to extract fixed-size latent features while keeping its essential information from the variable channel state information (CSI); 2) a reinforcement learning (RL) based beampattern optimization algorithm is proposed to search the desired beampattern according to the extracted features; 3) a supervised learning based beamforming reconstruction algorithm is proposed to reconstruct the beamforming vector from beampattern given by the RL agent. Simulation results demonstrate that the proposed LCT framework outperforms the baseline RL algorithm by optimizing the intuitional beampattern rather than beamforming. Moreover, the LCT framework provides a solution for low-cost beamforming optimization in ISAC systems. The trained RL module can be transferred without retraining when the antenna or user number changes.

In this paper, we introduce a novel simultaneous transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) assisted non-orthogonal multiple access (NOMA) model designed for non-line-of-sight (NLoS) scenarios. To ensure energy self-sustainability, an unmanned aerial vehicle (UAV) is introduced for wireless energy transfer. In the proposed model, ground users (GUs) situated in communication-obstructed environments are supported by STAR-RIS to connect with the base station (BS). Energy harvested from the UAV is utilized to enable prolonged communication with 360° full spatial coverage. An optimization problem is formulated to maximize the system’s sum-rate and is decomposed into three subproblems: phase-shift optimization, power allocation, and time allocation. These subproblems are solved using semidefinite relaxation (SDR), Dinkelbach’s method, and game theory, respectively. A joint resource allocation algorithm based on the block coordinate descent (BCD) method is then proposed. Simulation results show that the proposed UAV-assisted STAR-RIS-NOMA scheme, combined with the BCD algorithm, achieves a 43.64% improvement in system capacity compared to existing approaches.

A novel mobile edge generation (MEG) framework is proposed to support latency-sensitive generation tasks in the social-aware device-to-device (D2D) underlaying cellular network. Within this framework, user devices (UDs) and base station (BS) are organized into socially cohesive communities based on content preference and spatial proximity, enabling cooperative generation tasks via both cellular and intra-community D2D communications. A joint seed-and-content based BS-D2D (JSCB) transmission protocol is proposed to dynamically orchestrate the transmission mode between seed acquisition with local generation and direct content sharing across multiple consecutive task rounds, incorporating the spillover mechanism for handling overdue transmissions. Based on this protocol, an average task delay minimization problem is formulated to jointly optimize the UD association between cellular and D2D communication, transmission mode, D2D pairing, and BS-side beamforming. To efficiently solve the hybrid and temporally coupled problem, a joint matching and proximal policy optimization (JMPPO) algorithm is developed, where the discrete and continuous actions are decoupled with specialized modules though a hierarchical deep reinforcement learning and matching design. Numerical results validate that 1) the JSCB protocol reduces delay through adaptive transmission scheduling and cellular/D2D coordination; 2) the JMPPO algorithm outperforms both learning-based and traditional baselines in terms of average delay under the spillover and hybrid action scenarios; 3) the proposed schemes demonstrate robustness across diverse network and system conditions.

Understanding students' academic profiles and skillsets is crucial for personalized guidance in higher education. In transnational education (TNE) programmes, large student cohorts and time zone differences often complicate the allocation process of final year projects. To address these challenges, a project recommendation framework was developed. Using latent semantic analysis (LSA), students' academic profiles are summarized into skillsets, which are then matched with project requirements. This framework was deployed in a TNE programme between Queen Mary University of London (QMUL) and Beijing University of Posts and Telecommunications (BUPT). Quantitative results show that 80% of students using the framework secured a project on the first day of the allocation process, compared to 64% in a previous cohort without the tool, effectively shortening the allocation timeline. Qualitative feedback indicates high student satisfaction, emphasizing the tool's ease of use and relevance, as well as its ability to help students identify projects aligned with their academic profiles and interests. These findings reflect the framework's potential to streamline project allocation, reduce administrative workload, and enhance student support in project allocation. Moreover, the student skillsets generated by the framework can support broader applications, including employability analysis, academic profiling, and strategic decision-making to enhance institutional processes and student outcomes.

The beamforming optimization for maximizing weighted sum-rate (WSR) in continuous aperture array (CAPA)-based multi-user communications is studied. In particular, the transmit beamformers of CAPA are modelled as continuous source current patterns, rendering the beamforming optimization problem as a non-convex integral-based functional programming problem. In contrast to the state-of-the-art Fourier-based method that requires numerous Fourier basis functions to approximate the functional programming, a low-complexity calculus of variations (CoV)-based method is proposed to solve the functional programming problem for WSR maximization directly, where the optimal form of the continuous source patterns is derived. Based on this optimal form, a low-complexity integral-free iterative algorithm is developed. Our numerical results validate the effectiveness of the proposed designs. It is revealed that compared to the state-of-the-art Fourier-based method, the proposed CoV based method not only improves WSR performance but also reduces computational complexity by up to hundreds of times for large CAPA apertures and high frequencies.

In this paper, we propose a novel reconfigurable intelligent surface (RIS) -assisted non-orthogonal multiple access (NOMA) model, where an unmanned aerial vehicle (UAV) is employed for energy transfer. In this framework, the ground users (GUs) utilize energy from the UAV for information transmission to enable long-duration communication self-sufficiency. A sum-rate maximization problem is proposed, which is decomposed into four sub-problems: phase-shift optimization, power allocation, time allocation, and UAV trajectory optimization. These are solved using the SDR algorithm, CVX toolbox, game theory, and PSO algorithm, respectively. Subsequently, a joint resource allocation algorithm based on the block coordinate descent (BCD) method is introduced. Simulation results show that the UAV-assisted RIS-NOMA scheme, along with the proposed BCD algorithm, can increase the system capacity by 48.6% compared to the TDMA scheme as well as the alternating direction multiplier method (ADMM) and whale optimization algorithm (WOA).

In this paper, a novel device-to-device (D2D)-assisted cooperative non-orthogonal multiple access (NOMA) model with a two-stage transmission scenario is proposed, which consists of 1) partial decoding and forwarding from the transmitter to relay nodes; 2) transmission from relay nodes to the receivers. A sum-rate maximization problem is formulated, which is decoupled into subchannel selection and two-stage channel link power allocation. A joint optimization algorithm based on game theory and successive convex approximation (JOAGS) is proposed, which can efficiently utilize network resources and increase spectrum efficiency. The algorithm proposed in this paper has been validated through simulation results, demonstrating its substantial capability to amplify system capacity, diminish the outage probability of the communication link, and extend the communication distance. The findings reveal that when compared to the existing scheme, the system’s sum-rate is augmented by 10.8%, and the outage probability registers a notable reduction of 23.6%.

Node identification is the first line of defense for the security of wireless Internet-of-Things (IoT), which prevents illegal devices from accessing the network and launching attacks. Hardware features originating from innate hardware manufacturing imperfections are considered promising fingerprints for identification; among which, the hardware clock feature has been put under the spotlight due to its practicality and ease of extraction. However, current extractions of hardware clock features over wireless networks rely on the transmissions of time information, which, per se, enable significant vulnerabilities such as spoofing and replay attacks. In this paper, we propose a covert method to extract the hardware clock features, which does not rely on the insecure time information transmissions that are adopted in most existing schemes. We also analyze the security of the proposed covert extraction. We further propose SNICK, a secure node identification scheme based on our tailored implementation of covert clock feature extraction and machine learning. We implement and evaluate the proposed approach on a real IoT testbed consisting of a Long Range (LoRa) gateway and heterogeneous end nodes. We conduct experiments to prove the security of the proposed scheme and evaluate the proposed scheme under three scenarios: short-term, long-term, and cross-environment. Experimental results of three scenarios demonstrate average identification accuracies of 98.53%, 85.9%, and 88.3%. We further reveal the identification performance under parameter and environmental variations.

The simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) can provide a fullcoverage agile radio environment. A unique STAR-RIS-assisted coordinated multi-point (CoMP) framework is investigated in this paper, where cell-center and cell-edge users are embellished by the reflection and transmission features of the STAR-RIS. Unlike previous works controlling the transmission and reflection phase-shift independently, we consider a more practical coupled phase-shift model. We formulate an online active and passive beamforming problem to maximize long-term energy efficiency (EE) with time-varying locations and channels. Moreover, we propose a hybrid learning framework combining model-free and model-based optimization techniques. For the model-free method, we invoke a risk-sensitive multi-agent deep reinforcement learning algorithm to accelerate the online optimization of the passive beamforming of all STAR-RISs. For the model-based method, we invoke fractional programming (FP) to optimize the coordinated zero-forcing beamformer among all base stations and to realize an exact reward evaluation for each action in the DRL algorithm. Compared to DRL algorithms that optimize passive and active beamforming together, we dramatically shrink the action and state spaces. Comprehensive numerical results explain that the STAR-RIS enhanced CoMP system accomplishes a more excellent EE than the benchmark STAR-RIS cases and that without STAR-RIS.

This paper investigates the performance of simultaneously transmitting and reflecting surface (STARS) assisted semi-grant-free non-orthogonal multiple access network with randomly distributed users. By deploying STARS, the transmit signals of grant-based user (GBU) and grant-free users (GFUs) can be exquisitely adjusted to reduce interference. We propose a maximum channel scheduling (MCS) protocol that allows a GFU to access GBU’s channel with the assistance of STARS. In particular, the impacts of perfect/imperfect successive interference cancellation (pSIC/ipSIC) on MCS protocol are taken into account. To characterize the performance of STARS aided MCS (STARS-MCS) network, we derive the expressions of outage probability for GBU and GFU with pSIC/ipSIC. By applying convolution theorem and Laplace transform, the asymptotic expressions of outage probability and diversity orders for GBU and GFU are attained. We further design a STARS-based power control (SPC) strategy to eliminate the outage probability error floor and improve the outage performance. Numerical results show that: 1) The performance of STARS-MCS outperforms the existing benchmarks in terms of outage probability and system throughput; 2) The SPC strategy can effectively improve the performance of the STARS-MCS network and eliminate the outage probability error floor at high signal-to-noise ratios; and 3) By adjusting reflection and transmission coefficients of STARS, the outage performance of GBU and GFU can be greatly enhanced.

The capstone project is a crucial element of a degree programme and plays a vital role in the growth of learners, as it enables them to enhance their problem-solving skills and improve their employability prospects. In addition to this, the project provides the learners with an opportunity to demonstrate and showcase their critical thinking abilities and creativity. However, due to the year-long independent nature of these projects, learners can disengage due to a lack of motivation or self-regulated skills throughout the project. To address this problem, we formulated a data-driven intervention approach that conducts learner engagement analytics to identify and support disengaged learners, ensuring they maximise the benefits of completing a capstone project. The motivation was also to provide these learners with the necessary resources and support to get them back on track. This approach was implemented in the capstone projects conducted by learners at Queen Mary University of London within the School of Electronic Engineering and Computer Science. Based on the data of the three cohorts in 2020–21, 2021–22 and 2022–23, our analysis shows that the proposed data-driven intervention approach for capstone projects can effectively identify less-engaged learners and targeted interventions are shown to improve the overall performance of these less-engaged learners on capstone projects.

A near-field sensing (NISE)-enabled user tracking scheme is proposed. Compared to conventional angle-only sensing in the far-field scenario, NISE offers the capability of joint angle and distance sensing. In the proposed user tracking scheme, a deep unfolding neural network (DUNN)-based method is employed to sense radial and transverse velocities, which can further facilitate predicting the user’s location in the next time instant. Specifically, the DUNN is training on a synthesized dataset to update trainable parameters in an offline manner. Then, the DUNN is implemented online to extract the radial and transverse velocities from the received echo signal. Moreover, an online fine-tuning module is attached to the DUNN to refine the output of the pre-trained DUNN. Finally, based on estimated velocities, the user position in the consecutive time instant can be predicted, thus enabling user tracking. Simulation results show that the proposed scheme can extract velocities from echo signals and track the user accurately.

True-time delayers (TTDs) are popular components for hybrid beamforming architectures to combat the spatial-wideband effect in wideband near-field communications. A se-rial and a hybrid serial-parallel TTD configuration are inves-tigated for hybrid beamforming architectures. Compared to the conventional parallel configuration, the serial configuration exhibits a cumulative time delay through multiple TTDs, which potentially alleviates the maximum delay requirements on the TTDs. However, independent control of individual TTDs becomes impossible in the serial configuration. In this context, a hybrid TTD configuration is proposed as a compromise solution. More-over, the wideband near-field beamforming design for different configurations is studied for maximizing the spectral efficiency in single-user systems. In particular, a closed-form solution for the beamforming design is derived. The preferred user locations and the required maximum time delay of each TTD configuration are characterized. Our numerical results confirm the effectiveness of the proposed designs.

A simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) enhanced simultaneous wireless information and power transfer (SWIPT) system is investigated, in which a multi-antenna access point (AP) provides communication service and power supply for multiple single-antenna information decoding receivers (IDRs) and energy harvesting receivers (EHRs), respectively. To fully investigate the potential of STAR-RIS, three types of STAR-RIS protocols are employed in the SWIPT systems, namely the energy splitting (ES), the mode switching (MS), and the time switching (TS) protocols. The weighted sum power maximization problem is formulated for each STAR-RIS protocol to maximize the sum power received at the EHRs by jointly optimizing the beamforming at the AP and STAR-RIS. The non-convex problem for each STAR-RIS protocol is handled by the proposed low-complexity Gaussian randomization-based and high-precision penalty-based joint optimization algorithms to provide a more comprehensive range of algorithmic choices. Furthermore, we demonstrated that the AP only needs to send the information beam to achieve the best power supply and communication service in adopting any STAR-RIS protocol. Finally, numerical results demonstrate that: 1) the proposed schemes can achieve higher sum received power compared to the benchmarks; 2) the sum power received at the EHRs of the three STAR-RIS protocols can be ranked as ES > TS > MS; and 3) the performance gap in sum power received between penalty-based and Gaussian random algorithms depends on the number of STAR-RIS elements.

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 wideband simultaneously transmitting and reflecting surface (STARS) aided terahertz (THz) communication system is proposed. The spatial wideband effect at the base station (BS) and STARS leads to significant performance degradation due to the beam split issue. To address this, true time delayers (TTDs) are introduced into the conventional hybrid beamforming structure for facilitating wideband beamforming. The hybrid beamforming at the BS and the passive beamforming at the STARS are jointly designed to maximize the spectral efficiency of the proposed system. A double-loop iterative algorithm based on penalty dual decomposition is proposed to solve the resulting optimization problem. Finally, the numerical results confirm the effectiveness of exploiting STARS in wideband THz systems.

A near-field wideband multi-user communication system is studied. To eliminate the near-field beam split caused by the wideband spatial effect, a hybrid beamforming architecture based on true-time delayers (TTDs) is exploited. A heuristic two-stage approach is proposed for optimizing the analog and digital beamformers of the TTD-based hybrid beamforming architecture to facilitate near-field wideband beamfocusing. In particular, in the first stage, a closed-form analog beamformer design based on a piecewise-near-field approximation is proposed to maximize the array gain at users. Next, in the second stage, the digital beamformers are optimized by exploiting the successive convex approximation. Finally, our numerical results demonstrate that the proposed approach can effectively eliminate the near-field beam split and outperforms the existing approach in terms of spectral efficiency.

An adaptive next generation multiple access (NGMA) downlink scheme is provided, where non-orthogonal multiple access (NOMA) and space division multiple access (SDMA) users are served with the same orthogonal time and frequency resource to address the energy constraints and massive connectivity issues of Internet-of-Things networks. Based on this scheme, the long-term power-constrained sum rate maximization problem is investigated, where beamforming, power allocation, and user clustering are jointly optimized, subject to a long-term total power constraint. To solve the formulated problem, a spatial correlation-based user clustering approach is proposed and a resource allocation algorithm is designed based on the trust region policy optimization (TRPO) algorithm, which demonstrates stable convergence under large learning rates. Numerical results verify that the sum rate of the proposed NGMA scheme outperforms the conventional NOMA and SDMA schemes. Moreover, the spatial correlation-based clustering algorithm achieves an increasing sum rate gain compared to the channel correlation-based baseline algorithm as the spatial correlation in the channel model increases.

A reconfigurable intelligent surface (RIS)-assisted downlink nonorthogonal multiple access (NOMA) Internet of Things (IoT) network is proposed, where a Quality-of-Service (QoS)-based NOMA clustering scheme is conceived to effectively utilize the limited wireless resources among IoT devices. A throughput maximization problem is formulated by jointly optimizing the phase shifts of the RIS and the power allocation of the base station (BS) from the short-term and long-term perspectives. We aim to investigate and compare the performance of deep learning (DL) and deep reinforcement learning (DRL) algorithms for solving the formulated problems. In particular, the DL method utilizes model-agnostic-metalearning (MAML) to enhance the generalization capability of the neural network and to accelerate the convergence rate. For the DRL method, the deep deterministic policy gradient (DDPG) algorithm is employed to incorporate continuous phase-shift variables. It shows that the DL method only focuses on the maximization of the instantaneous throughput, whereas the DRL method can coordinate the power consumption over different time slots to maximize the long-term throughput. Numerical results demonstrate that: 1) the proposed QoS-based NOMA clustering scheme achieves higher IoT throughput than the conventional channel-based scheme; 2) the implementation of RISs induces approximately 5%–25% throughput gain as the number of RIS elements increases from 8 to 64; 3) DL and DRL achieve a similar throughput performance for the short-term optimization, while DRL is superior for the long-term optimization, especially when the total transmit power is limited.

A novel simultaneously transmitting and reflecting reconfigurable intelligent surfaces (STAR-RIS) assisted collabo-rative multi-cell network is proposed. Cell-center and cell-edge users are enhanced by the STAR-RIS's reflection and transmission functions, respectively. We propose an online joint active and passive beamforming framework to maximize this system's long-term energy efficiency (EE) under time-varying channels and users' requirements. We first invoke fractional programming (FP) to optimize the coordinated zero-forcing beamforming among all base stations and to construct an interference-free transmission during each time slot. Then, a parallel deep reinforcement learning (DRL) algorithm is proposed to facilitate the online optimization of the passive beamforming of all STAR-RISs. Finally, the beamforming calculated by the FP algorithm is utilized as a part of the reward function of the DRL. As a result, the size of the action and state space is reduced, and the dynamic joint optimization is realized. Extensive numerical results reveal that: 1) the proposed algorithm induces a low computation complexity compared with conventional DRL algorithms, and 2) the STAR-RIS enhanced system can achieve higher EE than systems without RIS or with conventional reflection/transmission-only RISs.

A novel reconfigurable intelligent surfaces (RISs)-based transmission framework is proposed for downlink non-orthogonal multiple access (NOMA) networks. We propose a quality-of-service (QoS)-based clustering scheme to improve the resource efficiency and formulate a sum rate maximization problem by jointly optimizing the phase shift of the RIS and the power allocation at the base station (BS). A model-agnostic meta-learning (MAML)-based learning algorithm is proposed to solve the joint optimization problem with a fast convergence rate and low model complexity. Extensive simulation results demonstrate that the proposed QoS-based NOMA network achieves significantly higher transmission throughput compared to the conventional orthogonal multiple access (OMA) network. It can also be observed that substantial throughput gain can be achieved by integrating RISs in NOMA and OMA networks. Moreover, simulation results of the proposed QoS-based clustering method demonstrate observable throughput gain against the conventional channel condition-based schemes.

Grant-free non-orthogonal multiple access (NOMA) is considered as one of the supporting technology for massive connectivity for future networks. In the grant-free NOMA systems with a massive number of users, user activity detection is of great importance. Existing multi-user detection (MUD) techniques rely on complicated update steps which may cause latency in signal detection. In this paper, we propose a generative neural network-based MUD (GenMUD) framework to utilize low-complexity neural networks, which are trained to reconstruct signals in a small fixed number of steps. By exploiting the uncorrelated user behaviours, we design a network architecture to achieve higher recovery accuracy with a low computational cost. Experimental results show significant performance gains in detection accuracy compared to conventional solutions under different channel conditions and user sparsity levels. We also provide a sparsity estimator through extensive experiments. Simulation results of the sparsity estimator showed high estimation accuracy, strong robustness to channel variations and neglectable impact on support detection accuracy.