HZHaoxiang Zhang

Haoxiang Zhang

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Haoxiang Zhang has a total of 42 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • A CGAN-LSTM-Based Framework for Time-Varying Non-Stationary Channel Modeling

    by Keying Guo, Ruisi He, Mi Yang, Yuxin Zhang, Bo Ai, Haoxiang Zhang, Jiahui Han, Ruifeng Chen

    Time-varying non-stationary channels exhibit complex dynamics and temporal evolution, which pose critical challenges for accurate modeling and performance evaluation of wireless communication systems. Different from most existing approaches that mainly focus on static channel generation or short-term prediction, our core objective is not precise future prediction, but the generation of synthetic channel sequences that statistically replicate the temporal non-stationary characteristics of real-world channels. To this end, we propose a hybrid deep learning framework that integrates conditional generative adversarial networks (CGAN) with long short-term memory (LSTM) networks. A temporal stationarity-constrained mechanism is further introduced to enforce correlations across consecutive snapshots, ensuring realistic channel evolution and preventing non-physical fluctuations. The framework is validated on ray tracing-based datasets under various non-stationary scenarios. Experimental results show that the generated channels closely match ground-truth simulations in terms of key statistical features.

  • Robust Transmission Design for RIS-Assisted High-Speed Train Communication Coverage Enhancement With Imperfect Cascaded Channels

    by Changzhu Liu, Ruisi He, Haoxiang Zhang, Jiahui Han, Ruifeng Chen, Bo Ai, Zhangdui Zhong

    Reconfigurable intelligent surface (RIS) has recently been gained attention as an effective technique improving the coverage and performance of communication systems by creating additional communication links. Deployment of RIS is crucial for overcoming signal coverage limitations, especially in high-speed train (HST) scenarios. Considerable research has been performed assuming perfect channel state information (CSI). However, due to the rapidly time-varying fading channels and feedback delays, achieving perfect CSI at the base station (BS) is not feasible in the HST scenarios. To tackle this problem, this paper investigates a robust design strategy for RIS-aided HST communication coverage enhancement, particularly focusing on cascaded BS-RIS-user channels at BS (CBRUB). The study explores the optimization problem under two types distinct of models: centered on minimizing transmit power subject to worst-case rate constraints within the bounded CSI error (BCSIE) model, and the other focusing on outage probability (OP) constraints under the statistical CSI error (SCSIE) model. We use the S-procedure to approximate the non-convex (NC) constraints, converting the worst-case rate constraints into linear matrix inequalities. Additionally, the Bernstein-type inequality is applied to transform the OP constraints into second-order cone constraints and linear inequalities. The simulation analysis results show that CBRUB errors have a significant effect on system performance compared to direct CSI errors.


  • Deep Learning-Based Dynamic Environment Reconstruction for Vehicular ISAC Scenarios

    by Junzhe Song, Ruisi He, Mi Yang, Zhengyu Zhang, Bingcheng Liu, Jiahui Han, Haoxiang Zhang, Bo Ai

    Integrated Sensing and Communication (ISAC) technology plays a critical role in future intelligent transportation systems, by enabling vehicles to perceive and reconstruct the surrounding environment through reuse of wireless signals, thereby reducing or even eliminating the need for additional sensors such as LiDAR or radar. However, existing ISAC-based reconstruction methods often lack the ability to track dynamic scenes with sufficient accuracy and temporal consistency, limiting the real-world applicability. To address this limitation, we propose a deep learning based framework for vehicular environment reconstruction by using ISAC channels. We first establish a joint channel–environment dataset based on multi-modal measurements from real-world urban street scenarios. Then, a multi-stage deep learning network is developed to reconstruct the environment. Specifically, a scene decoder identifies the environmental semantic context such as buildings, trees, and so on; a semantic center decoder predicts coarse spatial layouts by localizing dominant object centers; a point cloud decoder recovers fine-grained geometry and structure of surrounding environments. Experimental results demonstrate that the proposed method achieves high-quality global reconstruction of dynamic environments, with a Chamfer Distance of 0.29 and F-score@0.5m of 0.87. In addition, complexity analysis demonstrates the efficiency and practical applicability of the method in real-time scenarios. This work provides a pathway toward low-cost environment reconstruction based on ISAC for future intelligent transportation.


  • Point Cloud-Based Environmental Material Classification for Wireless Channel Ray-Tracing Simulations

    by Zhuoyin Li, Ruisi He, Mi Yang, Ziyi Qi, Zhong Zhang, Haoxiang Zhang, Jiahui Han, Bo Ai, Zhangdui Zhong

    Ray-tracing (RT) channel simulation has been widely used for simulating and analyzing propagation of electromagnetic waves in complex environments. Accuracy of RT simulations depends on environment construction, including both scene structure and material classification. However, existing RT studies rely on manual material segmentation and presuppose idealized material parameters, which highly overlooks critical challenges of material recognition from real-world data. This prevents RT from being applied in complex scenarios and leads to inaccurate simulation results. To address this issue, we use point cloud measurements to capture real-world environment information and propose a dual-branch network based on PointNet model to automatically classify environmental materials by integrating point cloud data and LiDAR-derived feature parameters. The proposed network significantly enhances material classification accuracy within complex scenes, thereby delivering more precise and computationally efficient input data for RT simulations. Furthermore, we analyze influence of material recognition accuracy on simulation parameters, such as path loss and delay spread. The results demonstrate that the proposed network achieves high classification performance and meets accuracy requirements of RT, thereby contributing to more realistic and reliable predictions for wireless systems. This approach lays a crucial foundation for development of environment-aware models for 6G networks, enabling more effective simulation of outdoor communication environments.

  • Impact of Point Cloud Reconstruction Detail on mmWave Ray-Tracing in Indoor Environments

    by Zhong Zhang, Ruisi He, Mi Yang, Ziyi Qi, Zhuoyin Li, Bo Ai, Haoxiang Zhang, Jiahui Han

    Ray tracing (RT) is a key tool for establishing accurate mappings between physical environments and propagation channels. However, due to the complexity of indoor scatterers (ISs) and the difficulty of fully capturing modeling details, there is no unified specification for scenario modeling, leaving the impact of modeling detail on RT simulations unclear. This work proposes an indoor modeling method based on LiDAR point clouds, which achieves automated mesh reconstruction via minimum bounding box estimation and generates scenario models at different levels of detail (LODs). Comparative analysis shows that modeling detail significantly affects the accuracy of multipath components (MPCs). A simple-detail model fails to accurately capture MPCs, whereas a medium-detail model exhibits higher simulation accuracy, and the full-detail model achieves the closest agreement with the ground truth model, although further accuracy gains diminish as complexity increases. Furthermore, ISs detail has a greater influence on RT results than room boundaries (RBs). To balance simulation accuracy and computational complexity, ISs should include at least external contour features, with opening structures preferred. In contrast, RB modeling can be simplified to a medium-detail model with only external contours. The sensitivity of different channel parameters to LOD also depends on the propagation scenario: in LOS scenarios, DS and PL are highly sensitive to LOD, whereas AS is relatively insensitive; and in NLOS scenarios, DS and AS are sensitive, while PL sensitivity significantly decreases.


  • Geometry-Based Channel Modeling for Bistatic Integrated Sensing and Communication Vehicular Networks

    by Yunwei Jin, Ruisi He, Bo Ai, Yuan Yuan, Haoxiang Zhang, Bincheng Liu

    Integrated Sensing and Communication (ISAC) has recently attracted significant attention. Vehicular networks (VNs) require communication links to effectively leverage dynamic information of surrounding vehicles. This necessity renders ISAC particularly suitable for vehicular applications, as it simultaneously provides robust communication and high-precision sensing capabilities. In this letter, we investigate channel modeling for bistatic ISAC in VNs, where the multi-functional vehicular transmitter is designed to communicate with the receiver while operating in a bistatic configuration with the sensing unit. To enhance generality of the model, we use an elliptical geometry structure to describe scatterers distribution, and derive channel impulse response for both the communication and sensing channels. The proposed model can be adapted to various bistatic sensing scenarios by adjusting the parameters. Subsequently, we analyze space-time-frequency correlation function. Numerical results indicate that the sensing channel has longer coherence time, while it exhibits greater sensitivity to variations in terminal velocity.

  • Ray-Tracing Calibration Based on Improved Particle Swarm Optimization in 5.9 GHz Outdoor Scenario

    by Zhuoyin Li, Ruisi He, Mi Yang, Ziyi Qi, Zhong Zhang, Bo Ai, Haoxiang Zhang, Jiahui Han, Jianhua Fan

    Please check and confirm whether the author affiliations in the first footnote are correct as set."?>Ray-tracing (RT) channel simulation has been widely used for simulating and analyzing propagation of electromagnetic waves in complex environments. The electrical properties of materials play a critical role in RT simulation accuracy and calibration of material is usually required. However, the traditional calibration methods for electrical parameters, such as the genetic algorithm and simulated annealing algorithm, suffer from slow convergence and limited parallelism, making them less suitable for complex scenarios with diverse material categories. This article presents a novel approach for automatic parameter optimization based on particle swarm optimization (PSO) algorithm. The proposed method leverages fast convergence advantage of the PSO algorithm. It incorporates a linearly-varying strategy to gradually reduce inertia weight, and enhances algorithm local search capability. Additionally, random perturbation factors are employed to reassign particles exceeding boundaries, preventing them from being trapped in local optima. Measurements and calibration of 5.9 GHz outdoor channels demonstrate the method, which achieves fairly high accuracy.


  • Deep Learning-Based Vehicular Millimeter-Wave Channel Prediction Using Visual Information

    by Shirui Wang, Ruisi He, Mi Yang, Bo Ai, Xuejian Zhang, Zhangdui Zhong, Haoxiang Zhang, Jiahui Han, Bingcheng Liu

    Millimeter-wave channel prediction requires detailed and accurate environmental information due to challenges of non-line-of-sight transmission and severe reflection and scattering. Image information conveniently reflects real-time changes of physical environment and provides a promising solution for millimeter-wave channel modeling. However, research on millimeter-wave channel prediction based on RGB images is still in its exploratory stage. Meanwhile, RGB images introduce new influencing factors, such as visual angle changes, leading to inaccurate prediction. This article proposes a millimeter-wave channel prediction model based on propagation knowledge and visual information, which enhances original RGB image features while combining manual features from specific domain prior knowledge in propagation. Experimental results show that the proposed model outperforms traditional RGB image-based prediction for a multi-scenario, multiview fused dataset, achieving a symmetric mean absolute percentage error as low as 4.16%. In addition, the proposed model demonstrates enhanced stability and interpretability.


  • Joint Angle-Delay Sparse Structured Compressed Sensing for ISAC Channel Estimation

    by Zeqian He, Ruisi He, Bo Ai, Haoxiang Zhang, Zhangdui Zhong

    In the upcoming 6 G communications, radar sensing will be integrated with communication to support a wide range of application scenarios. However, in complex Integrated Sensing and Communication (ISAC) channels, radar targets and communication scatterers may overlap, leading to similar sparse distributions of the sensing and communication channels in both the angle and delay domains. Additionally, the increasing use of compressed sensing (CS) for estimating communication parameters has given rise to numerous CS algorithms based on the expectation maximization (EM) method. However, due to the unique characteristics of the channel in specific scenarios, designing the channel sparse structure has become a crucial aspect of EM-base algorithm. Therefore, there is an urgent need to develop an appropriate channel sparse structure and design a CS algorithm specifically for the ISAC scenario. In this paper, we address the problem of channel estimation in a time-division duplex multi-input multi-output ISAC system using a bi-static model. We introduce a Joint Angle-Delay Sparse Structure (JADSS) to capture the sparsity characteristics of the ISAC channel in both the angle and delay domains. Furthermore, we design a message passing estimator within the framework of an EM-based double-loop algorithm to estimate the ISAC channel. Through simulations, we demonstrate that the proposed CS algorithm utilizing JADSS achieves competitive error performance.


  • Geometry-Based Stochastic MIMO Channel Model for Near-Field and Far-Field Scenarios of Integrated Sensing and Communications

    by Yunwei Jin, Ruisi He, Bo Ai, Yuan Yuan, Yong Niu, Haoxiang Zhang

    Integrated sensing and communication (ISAC) has become a promising technology in current sixth-generation (6G) wireless communications due to its ability to support both communication and sensing. Existing works regarding ISAC channels mainly focus on far-field propagation conditions based on plane wave assumption. However, with the deployment of ultra-massive multiple-input multiple-output (MIMO), the far-field assumption may no longer hold, as the communication and sensing propagation distances distances may be comparable to the array aperture size. To address this issue, we propose a three-dimensional MIMO channel model that simultaneously considers near-field and far-field propagation conditions in ISAC environments. Additionally, we distinguish the communication and sensing propagation environments by employing different effective scatterer distributions. In the model, the sensing channel is divided into target sensing and environmental sensing components, whereas the communication channel is divided into line-of-sight and non-line-of-sight components. The weighted sum of each component in communication and sensing channels results in accurate channel representations. Based on the proposed model, we derive and thoroughly investigate space-time-frequency correlation function, normalized absolute error function, and channel capacity. A key observation is that the channel correlation strongly depends on scatterer distribution. In certain conditions, the sensing channel can effectively capture the behavior of the communication channel. Error function results indicate that the proposed channel model achieves fairly high accuracy in ultra-massive MIMO scenarios. These findings provide essential support for the development of ISAC systems in future 6G wireless communications.