Xiaoxia Huang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Xiaoxia Huang has a total of 82 co-authors. Here are the top co-authors by number of shared publications:
- Yuguang Fang
City University of Hong Kong
8 shared publications - Lanhua Li
School of Intelligent Systems Engineering, Shenzhen Campus of Sun Yat-sen University, Shenzhen, China
5 shared publications - Xiaoyang He
School of Electronics and Communication Engineering, Sun Yat-sen University, Shenzhen, China
5 shared publications - Yun Zhang
School of Electronics and Communication Engineering, Sun Yat-sen University, Shenzhen Campus, Shenzhen, China
5 shared publications Guolin Chen
School of Electronics and Communication Engineering, Sun Yat-sen University, Shenzhen, China
4 shared publications- Pan Li
Hangzhou Dianzi University, Hangzhou, China
3 shared publications
Latest Publications
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 36  (11) , 17155-17169
Perceptually Optimized Bit Allocation Between Attribute and Geometry for Video-Based Point Cloud Compression
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 36  (11) , 17155-17169
Perceptually Optimized Bit Allocation Between Attribute and Geometry for Video-Based Point Cloud Compression
- 1 July 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25  (7) , 10274-10289
OnMAXFlow: Link-Aware Online Maximum Flow for Hybrid Ambient Backscatter Wireless Networks
- 1 July 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25  (7) , 10274-10289
OnMAXFlow: Link-Aware Online Maximum Flow for Hybrid Ambient Backscatter Wireless Networks
- 30 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Networking
- Vol. 34, 4959-4973
CausalFi: Causality-Based Cross-Domain Human Activity Recognition With Wi-Fi
WiFi-based human activity recognition (HAR) has demonstrated significant potential in diverse intelligent applications. However, sensitive to environmental factors, cross-domain WiFi channel state information (CSI) poses significant challenges in the generalization of HAR models across different environments. In this paper, the structural causal model (SCM) is introduced to model causal relationships among activities, latent variables, and CSI data, laying a solid foundation toward developing a domain-invariant model for WiFi-based HAR tasks. In this paper, we propose CausalFi, a novel framework that leverages causal inference to mitigate the confounding effects of latent variables, significantly improving generalization performance. Integrating novel feature selection and importance sampling algorithms as the condition and intervention operations, CausalFi can effectively identify the stable action-relevant features from WiFi CSI for activity recognition. Furthermore, a novel counterfactual style augmentation approach is proposed to increase the stylistic diversity of the training data, reducing the risk of biased data distributions even with limited training samples in source domains. We implement a prototype of CausalFi using commercial ASUS RT-AC86U WiFi devices and conduct extensive cross-domain experiments to validate the effectiveness of the proposed approach. With an average recognition accuracy of 92.6%, CausalFi significantly outperforms state-of-the-art baselines in complex cross-domain environments, confirming the practicality of our framework for real-world WiFi-based HAR applications.
- 30 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Networking
- Vol. 34, 4959-4973
CausalFi: Causality-Based Cross-Domain Human Activity Recognition With Wi-Fi
WiFi-based human activity recognition (HAR) has demonstrated significant potential in diverse intelligent applications. However, sensitive to environmental factors, cross-domain WiFi channel state information (CSI) poses significant challenges in the generalization of HAR models across different environments. In this paper, the structural causal model (SCM) is introduced to model causal relationships among activities, latent variables, and CSI data, laying a solid foundation toward developing a domain-invariant model for WiFi-based HAR tasks. In this paper, we propose CausalFi, a novel framework that leverages causal inference to mitigate the confounding effects of latent variables, significantly improving generalization performance. Integrating novel feature selection and importance sampling algorithms as the condition and intervention operations, CausalFi can effectively identify the stable action-relevant features from WiFi CSI for activity recognition. Furthermore, a novel counterfactual style augmentation approach is proposed to increase the stylistic diversity of the training data, reducing the risk of biased data distributions even with limited training samples in source domains. We implement a prototype of CausalFi using commercial ASUS RT-AC86U WiFi devices and conduct extensive cross-domain experiments to validate the effectiveness of the proposed approach. With an average recognition accuracy of 92.6%, CausalFi significantly outperforms state-of-the-art baselines in complex cross-domain environments, confirming the practicality of our framework for real-world WiFi-based HAR applications.
- 30 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 35  (10577149) , 3451-3465
Rate-Reconfigurable Deep Point Cloud Compression With Perceptual Bit Allocation Optimization
Conventional end-to-end learning-based point cloud compression requires training multiple models to adapt to different target bit rates. Moreover, the rate difference between geometry and attribute components of point clouds is not well-considered. In this paper, we propose an end-to-end Rate-Reconfigurable Deep Point Cloud Compression (RR-DPCC) with on/off-line Perceptual Bit Allocation Optimization (PBAO-ON/ OFF) , which achieves arbitrary bit rate control with one trained deep model and high efficiency joint geometry and attribute coding. First, we propose the framework of the RR-DPCC using PBAO-ON/OFF, which includes Point Cloud Quality Assessment (PCQA) for perceptual quality measurement, PBAO-ON/OFF modules for bit allocation and RR-DPCC for high efficiency point cloud coding. Second, we propose a one-stream network of the RR-DPCC to encode the attribute and geometry of point clouds jointly. Moreover, in RR-DPCC, a bitrate reconfigurable module is proposed to encode multiple fine-grained bitrate points with one trained model and a rate allocation module is proposed to allocate bits between geometry and attribute. Third, we propose on/off-line PBAO algorithms to maximize the perceptual quality of the reconstructed point cloud, where the bits are properly allocated based on the importance of geometry and attribute. Meanwhile, rate-distortion models (R-α/β and D-α/β) are derived for high accuracy rate control and bit allocation. Experimental results show that the proposed RR-DPCC achieves fine-grained bitrate control and allocation through a single trained model. When combined the proposed RR-DPCC with PBAO-ON, it reduces -6.56% and -18.68% bit rate on average as comparing with the state-of-the-art V-PCC and Deep Joint Geometry and Attribute Compression (Deep-JGAC), respectively. When combined with the PBAO-OFF, it reduces - 4.90% and -15.34% bit rate reductions on average, and reduces 98.38%/22.05% and 53.75%/10.04% encoding/decoding time on average with respect to V-PCC and Deep-JGAC.
- 30 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 35  (10577149) , 3451-3465
Rate-Reconfigurable Deep Point Cloud Compression With Perceptual Bit Allocation Optimization
Conventional end-to-end learning-based point cloud compression requires training multiple models to adapt to different target bit rates. Moreover, the rate difference between geometry and attribute components of point clouds is not well-considered. In this paper, we propose an end-to-end Rate-Reconfigurable Deep Point Cloud Compression (RR-DPCC) with on/off-line Perceptual Bit Allocation Optimization (PBAO-ON/ OFF) , which achieves arbitrary bit rate control with one trained deep model and high efficiency joint geometry and attribute coding. First, we propose the framework of the RR-DPCC using PBAO-ON/OFF, which includes Point Cloud Quality Assessment (PCQA) for perceptual quality measurement, PBAO-ON/OFF modules for bit allocation and RR-DPCC for high efficiency point cloud coding. Second, we propose a one-stream network of the RR-DPCC to encode the attribute and geometry of point clouds jointly. Moreover, in RR-DPCC, a bitrate reconfigurable module is proposed to encode multiple fine-grained bitrate points with one trained model and a rate allocation module is proposed to allocate bits between geometry and attribute. Third, we propose on/off-line PBAO algorithms to maximize the perceptual quality of the reconstructed point cloud, where the bits are properly allocated based on the importance of geometry and attribute. Meanwhile, rate-distortion models (R-α/β and D-α/β) are derived for high accuracy rate control and bit allocation. Experimental results show that the proposed RR-DPCC achieves fine-grained bitrate control and allocation through a single trained model. When combined the proposed RR-DPCC with PBAO-ON, it reduces -6.56% and -18.68% bit rate on average as comparing with the state-of-the-art V-PCC and Deep Joint Geometry and Attribute Compression (Deep-JGAC), respectively. When combined with the PBAO-OFF, it reduces - 4.90% and -15.34% bit rate reductions on average, and reduces 98.38%/22.05% and 53.75%/10.04% encoding/decoding time on average with respect to V-PCC and Deep-JGAC.
- 4 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 25  (15361276) , 11738-11751
ACK-UCB: An Asynchronous Contextual Kernel-Based Bandit Approach for User Association in mmWave Vehicular Networks
Timely channel conditions are essential for vehicles to determine which base station (BS) to connect to, but acquiring them in mmWave vehicular networks is costly. Without additional channel estimations, the proposed asynchronous contextual kernelized upper confidence bound (ACK-UCB) algorithm estimates the current instantaneous transmission rates based on the historical transmission rates and contexts, such as the vehicle’s historical locations, velocities, and numbers of concurrent transmissions at the BS. ACK-UCB captures the nonlinear relationship between context and transmission rate, mapping the context into a reproducing kernel Hilbert space (RKHS), where a linear relationship becomes observable. To enhance estimation accuracy, a novel kernel function incorporating mmWave signal propagation characteristics is introduced in RKHS, allowing for a more precise evaluation of context similarity in relation to transmission rates. Furthermore, ACK-UCB encourages vehicles to share only reward distribution features after sufficient explorations, accelerating the learning process while keeping communication costs manageable. Numerical results show that ACK-UCB achieves 99.5%–100.5% network throughput and reduces 89%–91% communication cost of a benchmark algorithm that directly shares all local historical contexts and transmission rates, demonstrating the sharing efficiency of the ACK-UCB algorithm.
- 4 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 25  (15361276) , 11738-11751
ACK-UCB: An Asynchronous Contextual Kernel-Based Bandit Approach for User Association in mmWave Vehicular Networks
Timely channel conditions are essential for vehicles to determine which base station (BS) to connect to, but acquiring them in mmWave vehicular networks is costly. Without additional channel estimations, the proposed asynchronous contextual kernelized upper confidence bound (ACK-UCB) algorithm estimates the current instantaneous transmission rates based on the historical transmission rates and contexts, such as the vehicle’s historical locations, velocities, and numbers of concurrent transmissions at the BS. ACK-UCB captures the nonlinear relationship between context and transmission rate, mapping the context into a reproducing kernel Hilbert space (RKHS), where a linear relationship becomes observable. To enhance estimation accuracy, a novel kernel function incorporating mmWave signal propagation characteristics is introduced in RKHS, allowing for a more precise evaluation of context similarity in relation to transmission rates. Furthermore, ACK-UCB encourages vehicles to share only reward distribution features after sufficient explorations, accelerating the learning process while keeping communication costs manageable. Numerical results show that ACK-UCB achieves 99.5%–100.5% network throughput and reduces 89%–91% communication cost of a benchmark algorithm that directly shares all local historical contexts and transmission rates, demonstrating the sharing efficiency of the ACK-UCB algorithm.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cognitive Communications and Networking
- Vol. 12, 12498-12513
An Efficient Frequency-Domain Variational Autoencoder-Enhanced Conditional Diffusion Model for Sampling-Free Fine-Grained Radio Map Generation
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Cognitive Communications and Networking
- Vol. 12, 12498-12513
An Efficient Frequency-Domain Variational Autoencoder-Enhanced Conditional Diffusion Model for Sampling-Free Fine-Grained Radio Map Generation
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Network Science and Engineering
- Vol. 13, 1-18
Nash Game Modeling for Network Defense Via Node Hiding and Resource Allocation
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Network Science and Engineering
- Vol. 13, 1-18
Nash Game Modeling for Network Defense Via Node Hiding and Resource Allocation
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Communications
- Vol. 74, 723-737
Personalized RIS for Tagging Users: Parallel Communications Based on Blind One-Shot Channel and Delay Estimation
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Communications
- Vol. 74, 723-737
Personalized RIS for Tagging Users: Parallel Communications Based on Blind One-Shot Channel and Delay Estimation
- 11 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13  (2) , 2748-2759
IRS-Enhanced Parallel Offloading or Direct Offloading: An Online Semi-Distributed Optimization for Latency Sensitive MEC
The surge in task-intensive applications underscores the critical need for mobile edge computing (MEC) to deliver low-latency services that cater to the computational demands of user equipment (UE). Computing density is inherently heterogeneous and dynamic, driven by the varying demands of UEs across space and time. To address the challenges posed by heterogeneous computing density, we propose a novel hybrid offloading framework that enables resource-constrained devices to dynamically choose between direct offloading and intelligent reflecting surface (IRS)-enhanced parallel offloading, thereby minimizing overall latency, improving resource utilization and adapting to varying conditions. Additionally, for prompt computation distribution, we propose an online semi-distributed optimization based on a two-timescale approach. Combining long-term centralized planning with short-term distributed computation assignment, this framework enables low-complexity distributed task allocation while ensuring performance and stability. Furthermore, we develop an equivalence and relaxation approach to derive an upper bound for completion delay of servers, which greatly simplifies the computation complexity of completion delay of tasks. Our proposed online semi-distributed algorithm converges, with reduced delay 24.03% and 32.68% compared to existing IRS-enhanced parallel offloading scheme and direct offloading scheme.
- 11 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 13  (2) , 2748-2759
IRS-Enhanced Parallel Offloading or Direct Offloading: An Online Semi-Distributed Optimization for Latency Sensitive MEC
The surge in task-intensive applications underscores the critical need for mobile edge computing (MEC) to deliver low-latency services that cater to the computational demands of user equipment (UE). Computing density is inherently heterogeneous and dynamic, driven by the varying demands of UEs across space and time. To address the challenges posed by heterogeneous computing density, we propose a novel hybrid offloading framework that enables resource-constrained devices to dynamically choose between direct offloading and intelligent reflecting surface (IRS)-enhanced parallel offloading, thereby minimizing overall latency, improving resource utilization and adapting to varying conditions. Additionally, for prompt computation distribution, we propose an online semi-distributed optimization based on a two-timescale approach. Combining long-term centralized planning with short-term distributed computation assignment, this framework enables low-complexity distributed task allocation while ensuring performance and stability. Furthermore, we develop an equivalence and relaxation approach to derive an upper bound for completion delay of servers, which greatly simplifies the computation complexity of completion delay of tasks. Our proposed online semi-distributed algorithm converges, with reduced delay 24.03% and 32.68% compared to existing IRS-enhanced parallel offloading scheme and direct offloading scheme.
- Open Access
- 1 August 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
Multiscale Feature Importance-based Bit Allocation for End-to-End Feature Coding for Machines
Feature Coding for Machines (FCM) aims to compress intermediate features effectively for remote intelligent analytics, which is crucial for future intelligent visual applications. In this paper, we propose a Multiscale Feature Importance-based Bit Allocation (MFIBA) for end-to-end FCM. First, we find that the importance of features for machine vision tasks varies with the scales, object size, and image instances. Based on this finding, we propose a Multiscale Feature Importance Prediction (MFIP) module to predict the importance weight for each scale of features. Secondly, we propose a task loss-rate model to establish the relationship between the task accuracy losses of using compressed features and the bitrate of encoding these features. Finally, we develop a MFIBA for end-to-end FCM, which is able to assign coding bits of multiscale features more reasonably based on their importance. Experimental results demonstrate that when combined with a retained Efficient Learned Image Compression (ELIC), the proposed MFIBA achieves an average of 38.202% bitrate savings in object detection compared to the anchor ELIC. Moreover, the proposed MFIBA achieves an average of 17.212% and 36.492% feature bitrate savings for instance segmentation and keypoint detection, respectively. When the proposed MFIBA is applied to the LIC-TCM, it achieves an average of 18.103%, 19.866% and 19.597% bit rate savings on three machine vision tasks, respectively, which validates the proposed MFIBA has good generalizability and adaptability to different machine vision tasks and FCM base codecs.
- 1 August 2025
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Multimedia Computing, Communications, and Applications
Open AccessMultiscale Feature Importance-based Bit Allocation for End-to-End Feature Coding for Machines
Feature Coding for Machines (FCM) aims to compress intermediate features effectively for remote intelligent analytics, which is crucial for future intelligent visual applications. In this paper, we propose a Multiscale Feature Importance-based Bit Allocation (MFIBA) for end-to-end FCM. First, we find that the importance of features for machine vision tasks varies with the scales, object size, and image instances. Based on this finding, we propose a Multiscale Feature Importance Prediction (MFIP) module to predict the importance weight for each scale of features. Secondly, we propose a task loss-rate model to establish the relationship between the task accuracy losses of using compressed features and the bitrate of encoding these features. Finally, we develop a MFIBA for end-to-end FCM, which is able to assign coding bits of multiscale features more reasonably based on their importance. Experimental results demonstrate that when combined with a retained Efficient Learned Image Compression (ELIC), the proposed MFIBA achieves an average of 38.202% bitrate savings in object detection compared to the anchor ELIC. Moreover, the proposed MFIBA achieves an average of 17.212% and 36.492% feature bitrate savings for instance segmentation and keypoint detection, respectively. When the proposed MFIBA is applied to the LIC-TCM, it achieves an average of 18.103%, 19.866% and 19.597% bit rate savings on three machine vision tasks, respectively, which validates the proposed MFIBA has good generalizability and adaptability to different machine vision tasks and FCM base codecs.