arXiv subjects

Huiling Zhu

Publications and source records attributed to Huiling Zhu.

At least 19 recordsLinked to original sources

arXiv2026

TeRoR: Decoupled Temporal Rotation with Relational Circular Region for Temporal Knowledge Graph Embedding

In recent years, with the emergence of Temporal Knowledge Graphs (TKGs), research on learning entity and relation representations in TKGs has attracted increasing attention, giving rise to a large number of TKG embedding methods. TeRo is a simple and efficient temporal knowledge graph embedding approach. However, TeRo does not do well in modeling the mapping properties of various relations, such as one-to-many, many-to-one, and many-to-many. Meanwhile, it also has limitations in the expression of temporal information. To address these issues, we propose a novel TKG embedding method named TeRoR. This method divides the temporal evolution of entity embeddings, and conducts independent rotation transformations on head and tail entities in the complex vector space to strengthen temporal information modeling capacity. In terms of relational characteristics, we train a radius to constrain the rotated and translated head entities within a circular region centered on the tail entity, which effectively captures the diverse mapping properties of relations. Experimental results demonstrate that TeRoR achieves competitive performance against state-of-the-art models on four distinct TKG datasets.

cs.LG↗

arXiv2026

RelBall: Relation Ball with Quaternion Rotation for Knowledge Graph Completion

Real-world knowledge graphs are often incomplete, lacking many valid facts. Knowledge Graph Completion (KGC) aims to predict missing links using known triples, thereby enhancing graph coverage. A key challenge is modeling diverse relational patterns such as symmetry, antisymmetry, inversion, composition and semantic hierarchy. Existing models such as RotatE can capture symmetric, antisymmetric, inverse, and commutative composition patterns, yet struggle with non-commutative composition. Rotate3D addresses this by introducing non-commutativity via three-dimensional rotations, but still fails to capture the semantic hierarchies prevalent in knowledge graphs. Moreover, both models cannot effectively model one-to-many relations. To overcome these limitations, we propose RelBall, which extends Rotate3D with two innovations. First, our model introduces modulus transformation to model hierarchies, driving abstract concepts toward smaller moduli and concrete instances toward larger ones. Second, it introduces a tail-centric relation ball to model one-to-one, one-to-many, many-to-one, and many-to-many relations. RelBall offers the following advantages: (1) coverage of all relational patterns, including the ones mentioned above; (2) an interpretable hierarchical representation where the modulus directly reflect semantic levels; (3) support for one-to-one, one-to-many, many-to-one, and many-to-many relations. Experiments on multiple datasets demonstrate RelBall's competitive link prediction performance against various baselines.

cs.AI↗

arXiv2026

R$^{2}$Net: 2D Deep Residual Learning with Height Embedding for 3D Radio Map Estimation

Acquiring channel knowledge is required by many applications. For instance, handover in cellular networks is mainly decided based on the knowledge of pathloss. In contrast to traditional statistical distance-determined models that might provide misleading pathloss estimates, researchers started to explore deep learning methods recently to accurately estimate the radio map that characterizes the spatial distribution of pathloss according to the specific physical wireless propagation environment. However, existing works mainly focused on 2D radio map estimation by assuming that all receivers are at the same height. In fact, radio maps could be significantly different at different receiver heights, highlighting the importance of 3D radio map estimation. In this paper, we first propose a method to embed height information into 2D images, and then propose a general 2D radio residual network (R$^{2}$Net) for 3D radio map estimation. Since pathloss exhibits different characteristics in indoor and outdoor scenarios, we specifically propose R$^{2}$Net-In for indoor scenarios and R$^{2}$Net-Out for outdoor scenarios to better capture penetration loss and diffraction loss, respectively. Extensive experimental results show that our R$^{2}$Net significantly outperforms the state-of-the-art benchmarks in terms of estimation accuracy, computational and storage costs, and inference speed. In addition, due to the lack of publicly available 3D radio map datasets, a 3D indoor radio map dataset (3DiRM3200) is created, which took more than $1,000$ labour hours. The dataset and codes will be available at https://github.com/lighttime2023/3DiRM3200.git.

eess.SP↗

arXiv2026

CORE: Cyclic Orthotope Relation Embedding for Knowledge Graph Completion

Knowledge graph completion (KGC) aims to automatically infer missing facts in multi-relational data by mapping entities and relations into continuous representation spaces. Recent region-based embedding models have shown great promise in capturing complex logical patterns by representing relations as geometric regions. However, these models inevitably suffer from absolute boundary constraints during optimization. Conversely, without such constraints, relation regions expand indefinitely. To address the limitation, we propose \textbf{CORE} (Cyclic Orthotope Relation Embedding), a novel KGC model that embeds entities and relations onto a boundary-less torus manifold.CORE represents relations as cyclic orthotopes on the torus manifold, allowing regions to seamlessly wrap around spatial boundaries to ensure smooth gradient conduction. Furthermore, an adaptive width regularization is introduced to prevent unconditional region expansion. Theoretical analysis proves that CORE can capture various complex relation patterns such as subsumption and intersection. Extensive experiments on four benchmark datasets demonstrate that CORE achieves highly competitive performance, significantly improving link prediction accuracy in dense semantic environments.

cs.LG↗

arXiv2025

Data Heterogeneity-Aware Client Selection for Federated Learning in Wireless Networks

Federated Learning (FL) enables mobile edge devices, functioning as clients, to collaboratively train a decentralized model while ensuring local data privacy. However, the efficiency of FL in wireless networks is limited not only by constraints on communication and computational resources but also by significant data heterogeneity among clients, particularly in large-scale networks. This paper first presents a theoretical analysis of the impact of client data heterogeneity on global model generalization error, which can result in repeated training cycles, increased energy consumption, and prolonged latency. Based on the theoretical insights, an optimization problem is formulated to jointly minimize learning latency and energy consumption while constraining generalization error. A joint client selection and resource allocation (CSRA) approach is then proposed, employing a series of convex optimization and relaxation techniques. Extensive simulation results demonstrate that the proposed CSRA scheme yields higher test accuracy, reduced learning latency, and lower energy consumption compared to baseline methods that do not account for data heterogeneity.

cs.DC↗

arXiv2025

Knowledge Graph Embeddings with Representing Relations as Annular Sectors

Knowledge graphs (KGs), structured as multi-relational data of entities and relations, are vital for tasks like data analysis and recommendation systems. Knowledge graph completion (KGC), or link prediction, addresses incompleteness of KGs by inferring missing triples (h, r, t). It is vital for downstream applications. Region-based embedding models usually embed entities as points and relations as geometric regions to accomplish the task. Despite progress, these models often overlook semantic hierarchies inherent in entities. To solve this problem, we propose SectorE, a novel embedding model in polar coordinates. Relations are modeled as annular sectors, combining modulus and phase to capture inference patterns and relation attributes. Entities are embedded as points within these sectors, intuitively encoding hierarchical structure. Evaluated on FB15k-237, WN18RR, and YAGO3-10, SectorE achieves competitive performance against various kinds of models, demonstrating strengths in semantic modeling capability.

cs.LG↗

arXiv2024

Anti-Byzantine Attacks Enabled Vehicle Selection for Asynchronous Federated Learning in Vehicular Edge Computing

In vehicle edge computing (VEC), asynchronous federated learning (AFL) is used, where the edge receives a local model and updates the global model, effectively reducing the global aggregation latency.Due to different amounts of local data,computing capabilities and locations of the vehicles, renewing the global model with same weight is inappropriate.The above factors will affect the local calculation time and upload time of the local model, and the vehicle may also be affected by Byzantine attacks, leading to the deterioration of the vehicle data. However, based on deep reinforcement learning (DRL), we can consider these factors comprehensively to eliminate vehicles with poor performance as much as possible and exclude vehicles that have suffered Byzantine attacks before AFL. At the same time, when aggregating AFL, we can focus on those vehicles with better performance to improve the accuracy and safety of the system. In this paper, we proposed a vehicle selection scheme based on DRL in VEC. In this scheme, vehicle s mobility, channel conditions with temporal variations, computational resources with temporal variations, different data amount, transmission channel status of vehicles as well as Byzantine attacks were taken into account.Simulation results show that the proposed scheme effectively improves the safety and accuracy of the global model.

cs.LG↗

arXiv2024

Spectral Efficiency Maximization for Active RIS-aided Cell-Free Massive MIMO Systems with Imperfect CSI

A cell-free network merged with active reconfigurable reflecting surfaces (RIS) is investigated in this paper. Based on the imperfect channel state information (CSI), the aggregated channel from the user to the access point (AP) is initially estimated using the linear minimum mean square error (LMMSE) technique. The central processing unit (CPU) then detects uplink data from individual users through the utilization of the maximum ratio combining (MRC) approach, relying on the estimated channel. Then, a closed-form expression for uplink spectral efficiency (SE) is derived which demonstrates its reliance on statistical CSI (S-CSI) alone. The amplitude gain of each active RIS element is derived in a closed-form expression as a function of the number of active RIS elements, the number of users, and the size of each reflecting element. A soft actor-critic (SAC) algorithm is utilized to design the phase shift of the active RIS to maximize the uplink SE. Simulation results emphasize the robustness of the proposed SAC algorithm, showcasing its effectiveness in cell-free networks under the influence of imperfect CSI.

eess.SP↗

arXiv2024

Cooperative Edge Caching Based on Elastic Federated and Multi-Agent Deep Reinforcement Learning in Next-Generation Network

Edge caching is a promising solution for next-generation networks by empowering caching units in small-cell base stations (SBSs), which allows user equipments (UEs) to fetch users' requested contents that have been pre-cached in SBSs. It is crucial for SBSs to predict accurate popular contents through learning while protecting users' personal information. Traditional federated learning (FL) can protect users' privacy but the data discrepancies among UEs can lead to a degradation in model quality. Therefore, it is necessary to train personalized local models for each UE to predict popular contents accurately. In addition, the cached contents can be shared among adjacent SBSs in next-generation networks, thus caching predicted popular contents in different SBSs may affect the cost to fetch contents. Hence, it is critical to determine where the popular contents are cached cooperatively. To address these issues, we propose a cooperative edge caching scheme based on elastic federated and multi-agent deep reinforcement learning (CEFMR) to optimize the cost in the network. We first propose an elastic FL algorithm to train the personalized model for each UE, where adversarial autoencoder (AAE) model is adopted for training to improve the prediction accuracy, then {a popular} content prediction algorithm is proposed to predict the popular contents for each SBS based on the trained AAE model. Finally, we propose a multi-agent deep reinforcement learning (MADRL) based algorithm to decide where the predicted popular contents are collaboratively cached among SBSs. Our experimental results demonstrate the superiority of our proposed scheme to existing baseline caching schemes.

cs.LG↗

arXiv2024

Vehicle Selection for C-V2X Mode 4 Based Federated Edge Learning Systems

Federated learning (FL) is a promising technology for vehicular networks to protect vehicles' privacy in Internet of Vehicles (IoV). Vehicles with limited computation capacity may face a large computational burden associated with FL. Federated edge learning (FEEL) systems are introduced to solve such a problem. In FEEL systems, vehicles adopt the cellular-vehicle to everything (C-V2X) mode 4 to upload encrypted data to road side units' (RSUs)' cache queue. Then RSUs train the data transmitted by vehicles, update the locally model hyperparameters and send back results to vehicles, thus vehicles' computational burden can be released. However, each RSU has limited cache queue. To maintain the stability of cache queue and maximize the accuracy of model, it is essential to select appropriate vehicles to upload data. The vehicle selection method for FEEL systems faces challenges due to the random departure of data from the cache queue caused by the stochastic channel and the different system status of vehicles, such as remaining data amount, transmission delay, packet collision probability and survival ability. This paper proposes a vehicle selection method for FEEL systems that aims to maximize the accuracy of model while keeping the cache queue stable. Extensive simulation experiments demonstrate that our proposed method outperforms other baseline selection methods.

cs.NI↗

arXiv2023

Two-Timescale Design for RIS-aided Cell-free Massive MIMO Systems with Imperfect CSI

The objective of this paper is to evaluate the effectiveness of a two-timescale transmission design in cell-free massive multi-input multiple-output (MIMO) systems incorporating reconfigurable intelligent surfaces (RISs) under the assumption of imperfect channel state information (CSI). We examine the Rician channel model and formulate the passive beamforming for the RISs based on statistical channel state information (S-CSI). To that end, we put forth a linear minimum mean square error (LMMSE) estimator with the aim of estimating the aggregation of channels from the users to the APs within each channel coherence interval. Meanwhile, the active beamforming for the radio units (APs) is executed using the maximum ratio combining (MRC) approach, which utilizes the instantaneous aggregated channels, that result from the combination of the direct and reflected channels from the RISs. Subsequently, we derive the closed-form expressions of the achievable uplink spectral efficiency (SE), which is a function of S-CSI elements such as distance-dependent path loss, Rician factors as well as the number of RIS elements and AP antennas. We then optimize the phase shifts of the RISs to maximize the sum SE of the users, utilizing the soft actor-critic (SAC) which is a deep reinforcement learning (RL) method, and relying on the derived closed-form expressions. Numerical evaluations affirm that, despite the presence of imperfect CSI, the deployment of RIS in cell-free systems can lead to significant performance improvement.

eess.SP↗

arXiv2022

Statistical CSI-based Beamforming for RIS-Aided Multiuser MISO Systems using Deep Reinforcement Learning

The paper presents a joint beamforming algorithm using statistical channel state information (S-CSI) for reconfigurable intelligent surfaces (RIS) for multiuser MISO wireless communications. We used S-CSI, which is a long-term average of the cascaded channel as opposed to instantaneous CSI utilized in most existing works. Through this method, the overhead of channel estimation is dramatically reduced. We propose a proximal policy optimization (PPO) algorithm which is a well-known actor-critic based reinforcement learning (RL) algorithm to solve the optimization problem. To test the efficacy of this algorithm, simulation results are presented along with evaluations of key system parameters, including the Rician factor and RIS location, on the achievable sum rate of the users.

eess.SP↗

arXiv2022

Smart Interference Management xApp using Deep Reinforcement Learning

Interference continues to be a key limiting factor in cellular radio access network (RAN) deployments. Effective, data-driven, self-adapting radio resource management (RRM) solutions are essential for tackling interference, and thus achieving the desired performance levels particularly at the cell-edge. In future network architecture, RAN intelligent controller (RIC) running with near-real-time applications, called xApps, is considered as a potential component to enable RRM. In this paper, based on deep reinforcement learning (RL) xApp, a joint sub-band masking and power management is proposed for smart interference management. The sub-band resource masking problem is formulated as a Markov Decision Process (MDP) that can be solved employing deep RL to approximate the policy functions as well as to avoid extremely high computational and storage costs of conventional tabular-based approaches. The developed xApp is scalable in both storage and computation. Simulation results demonstrate advantages of the proposed approach over decentralized baselines in terms of the trade-off between cell-centre and cell-edge user rates, energy efficiency and computational efficiency.

eess.SP↗

arXiv2021

Self-Sustainable Reconfigurable Intelligent Surface Aided Simultaneous Terahertz Information and Power Transfer (STIPT)

This paper proposes a new simultaneous terahertz (THz) information and power transfer (STIPT) system, which is assisted by reconfigurable intelligent surface (RIS) for both the information data and power transmission. We aim to maximize the information users' (IUs') data rate while guaranteeing the energy users' (EUs') and RIS's power harvesting requirements. To solve the formulated non-convex problem, the block coordinate descent (BCD) based algorithm is adopted to alternately optimize the transmit precoding of IUs, RIS's reflecting coefficients, and RIS's coordinate. The Penalty Constrained Convex Approximation (PCCA) Algorithm is proposed to solve the intractable optimization problem of the RIS's coordinate, where the solution's feasibility is guaranteed by the introduced penalties. Simulation results confirm that the proposed BCD algorithm can significantly enhance the performance of STIPT by employing RIS.

eess.SP↗

arXiv2020

UAV-Assisted and Intelligent Reflecting Surfaces-Supported Terahertz Communications

In this paper, unmanned aerial vehicles (UAVs) and intelligent reflective surface (IRS) are utilized to support terahertz (THz) communications. To this end, the joint optimization of UAV's trajectory, the phase shift of IRS, the allocation of THz sub-bands, and the power control is investigated to maximize the minimum average achievable rate of all the users. An iteration algorithm based on successive Convex Approximation with the Rate constraint penalty (CAR) is developed to obtain UAV's trajectory, and the IRS phase shift is formulated as a closed-form expression with introduced pricing factors. Simulation results show that the proposed scheme significantly enhances the rate performance of the whole system.

eess.SP↗

arXiv2020

Sum Rate Maximization for Intelligent Reflecting Surface Assisted Terahertz Communications

In this paper, an intelligent reflecting surface (IRS) is deployed to assist the terahertz (THz) communications. The molecular absorption causes path loss peaks to appear in the THz frequency band, and the fading peak is greatly affected by the transmission distance. In this paper, we aim to maximize the sum rate with individual rate constraints, in which the IRS location, IRS phase shift, the allocation of sub-bands of the THz spectrum, and power control for UEs are jointly optimized. For the special case of a single user equipment (UE) with a single sub-band, the globally optimal solution is provided. For the general case with multiple UEs, the block coordinate searching (BCS) based algorithm is proposed to solve the non-convex problem. Simulation results show that the proposed scheme can significantly enhance system performance.

cs.IT↗

arXiv2020

Cost Minimization for Cooperative Computation Framework in MEC Networks

In this paper, a cooperative task computation framework exploits the computation resource in UEs to accomplish more tasks meanwhile minimizes the power consumption of UEs. The system cost includes the cost of UEs' power consumption and the penalty of unaccomplished tasks and the system cost is minimized by jointly optimizing binary offloading decisions, the computational frequencies, and the offloading transmit power. To solve the formulated mixed-integer non-linear programming problem, three efficient algorithms are proposed, i.e., integer constraints relaxation-based iterative algorithm (ICRBI), heuristic matching algorithm, and the decentralized algorithm. The ICRBI algorithm achieves the best performance at the cost of the highest complexity, while the heuristic matching algorithm significantly reduces the complexity while still providing reasonable performance. As the previous two algorithms are centralized, the decentralized algorithm is also provided to further reduce the complexity, and it is suitable for the scenarios that cannot provide the central controller. The simulation results are provided to validate the performance gain in terms of the total system cost obtained by the proposed cooperative computation framework.

cs.IT↗

arXiv2020

Latency Minimization for Task Offloading in Hierarchical Fog-Computing C-RAN Networks

Fog-computing network combines the cloud computing and fog access points (FAPs) equipped with mobile edge computing (MEC) servers together to support computation-intensive tasks for mobile users. However, as FAPs have limited computational capabilities and are solely assisted by a remote cloud center in the baseband processing unit (BBU) of the cloud radio access (C-RAN) network, the latency benefits of this fog-computing C-RAN network may be worn off when facing a large number of offloading requests. In this paper, we investigate the delay minimization problem for task offloading in a hierarchical fog-computing C-RAN network, which consists of three tiers of computational services: MEC server in radio units, MEC server in distributed units, and the cloud computing in central units. The receive beamforming vectors, task allocation, computing speed for offloaded tasks in each server and the transmission bandwidth split of fronthaul links are optimized by solving the formulated mixed integer programming problem. The simulation results validate the superiority of the proposed hierarchical fog-computing C-RAN network in terms of the delay performance.

cs.IT↗