Huagang Tong

Huagang Tong

Auto-generated profile, based on publicly available information

Huagang Tong

College of Economic and Management, Nanjing Tech University, Nanjing, China

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
8 January 2025 in IEEE Transactions on Computational Social Systems
Verified
Journal Article
Closed access
ZXZhiwen Xiao
Zhiwen Xiao
HTHuagang Tong
Huagang Tong
This article proposes a federated contrastive learning with feature-based distillation (FCLFD) framework tailored for human activity recognition (HAR). The FCLFD system integrates a central server with multiple mobile users to address a diverse range of HAR challenges. The framework encompasses two pivotal elements: a contrastive student--teacher (CST) architecture with feature-based distillation and an average weight scheme (AWS). The CST framework facilitates the transfer of comprehensive knowledge from a teacher model to a student model through feature-based distillation and contrastive learning, with both models sharing an identical architecture. Each participating user periodically uploads the weights of its student model to the central server, where the AWS deployed on the server calculates the average weights based on contributions from all connected users. The aggregated weights are then redistributed to each user, who updates their teacher model accordingly. Experimental evaluations demonstrate that when 50 users are connected, the proposed FCLFD scheme obtains the highest F 1 values of 89.01 and 94.19, outperforming several state-of-the-art federated learning algorithms on the wireless sensor data mining (WISDM) and PAMAP2 datasets.

1 September 2020 in Applied Soft Computing
Verified
Journal Article
Closed access
HTHuagang Tong
Huagang Tong
JZJianjun Zhu
Jianjun Zhu

3 November 2025 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
ZXZhiwen Xiao
Zhiwen Xiao
QWQian Wan
Qian Wan
HTHuagang Tong
Huagang Tong
HXHuanlai Xing
Huanlai Xing
Electrocardiogram (ECG) signals inherently encode both localized fluctuations and long-range temporal dynamics, serving as essential indicators for accurate cardiovascular diagnosis. With the rapid expansion of wearable and portable ECG acquisition devices—key enablers in instrumentation and measurement (IM)—the need for robust, scalable, and noise-resilient analytical frameworks has intensified. Yet, existing deep learning approaches often struggle to jointly capture fine-grained transient patterns and global temporal structures due to the high dimensionality and complexity of ECG data. To address these challenges, we propose AttMambaECG, an advanced attentional knowledge-based state-space model designed to enhance feature representation across multiple temporal scales. The architecture comprises four hierarchical attentional state space module (SSM) blocks, each integrating an SSM to capture sequential dynamics and a multi-head self-attention mechanism to extract diverse contextual representations. An intra-attention fusion layer further enhances temporal feature coherence, while the Gaussian error linear unit (GeLU) activation improves non-linear expressivity. This design enables AttMambaECG to achieve high-fidelity modeling across varying time scales and perturbation conditions, aligning with the stringent precision demands of IM. Comprehensive evaluations on three benchmark ECG datasets demonstrate that AttMambaECG consistently outperforms 22 representative machine learning and deep learning baselines. It achieves substantial gains in accuracy, recall, precision, and F1-score—even under severe noise conditions. Moreover, real-world deployment experiments on Raspberry Pi 4 and PYNQ-Z2 confirm its feasibility for efficient and low-latency inference on edge platforms, highlighting its practical value for mobile and wearable ECG monitoring applications.

12 May 2026 in IEEE Transactions on Knowledge and Data Engineering
Verified
Journal Article
Closed access
ZXZhiwen Xiao
Zhiwen Xiao
QWQian Wan
Qian Wan
Weiping DingWeiping Ding
Weiping Ding
FSFuhong Song
Fuhong Song
HTHuagang Tong
Huagang Tong
Multivariate time series classification (MTSC) requires a model capable of capturing both localized temporal patterns and long-range dependencies while effectively modeling complex inter-variable relationships. Existing convolutional neural network (CNN)-based capsule models suffer from limited receptive fields, constraining their ability to model long-range dependencies, while transformer-based capsule models rely solely on self-attention, which, despite its effectiveness in capturing global features, struggles with preserving local structures and efficiently processing long sequences. To overcome these limitations, we propose KACapMamba, a Knowledge-Aware State-Space Capsule Network, which integrates three attentive Mamba blocks with a routing layer to achieve hierarchical temporal modeling. Unlike conventional transformer-based methods that primarily depend on self-attention for global dependency modeling, each attentive Mamba block in KACapMamba fuses 1-dimensional CNNs, self-attention, state-space module (SSM), and mutual cross-attention, enabling a more structured and adaptive feature representation. Self-attention ensures effective long-range dependency modeling, while SSM provides a recurrent-state mechanism, inherently better suited for sequential processing compared to purely attention-based architectures, thereby enhancing temporal continuity and long-term pattern retention. Notably, mutual cross-attention addresses the limitations of traditional fusion strategies such as element-wise addition or multiplication, which lack the capacity to selectively enhance relevant features. By dynamically reweighting interactions between features, mutual cross-attention enables more expressive, context-aware representations, leading to improved feature disentanglement and inter-variable modeling. Additionally, the routing layer further enhances hierarchical feature disentanglement by refining capsule activations, reinforcing structural coherence and feature selectivity. Experiments conducted across the UEA benchmark archive demonstrate that KACapMamba consistently achieves the highest ‘win’/‘tie’/‘lose’/‘best’ ratios when evaluated against 10 leading transformer and Mamba architectures under both $Accuracy$ and $F_{1}$ metrics. Moreover, in comparison with 22 state-of-the-art MTSC models, it again secures the most favorable performance profile, demonstrating a clear and statistically supported advantage across both evaluation measures.

9 June 2026 in ACM Transactions on Embedded Computing Systems
Verified
Journal Article
Open access
ZXZhiwen Xiao
Zhiwen Xiao
QWQian Wan
Qian Wan
HTHuagang Tong
Huagang Tong
Wearable human activity recognition (HAR) focuses on classifying human activities from multi-sensor data collected by wearable devices and has become increasingly important in pervasive computing. However, existing methods face several challenges: 1) aggregating heterogeneous local models while preserving user-specific data distributions, 2) achieving personalized adaptation of global models to diverse behavioral patterns, and 3) capturing both local and global temporal dependencies inherent in sensor time-series data. To address these challenges, we propose PFLMamba, a personalized federated learning framework integrating an attention-enhanced state-space model (ASSM) for hierarchical temporal feature extraction. PFLMamba employs a server-side personalized attention aggregation mechanism to tailor global models for individual clients, while ASSM captures both local and long-range temporal patterns on the client side. Extensive evaluations on the WISDM and PAMAP2 datasets demonstrate that PFLMamba achieves F 1 scores of 91.83 and 96.01, respectively, outperforming state-of-the-art federated learning baselines such as FCLFD, EFDLS, and FKD. PFLMamba’s effectiveness is further validated on multi-user and heterogeneous-sensor datasets, namely UCI-HAR and UNIMIB-SHAR, confirming its generalization across diverse populations and device types. Beyond predictive accuracy, PFLMamba exhibits a favorable trade-off between efficiency and performance, with lower client-side training overhead than teacher-student-based frameworks and competitive throughput relative to lightweight alternatives. Further experiments on resource-constrained edge devices (i.e., Raspberry Pi 4 and PYNQ-Z2) validate its practical feasibility, highlighting low-latency inference and moderate energy consumption. These results establish PFLMamba as a robust and efficient solution for personalized wearable HAR.

2 January 2025 in Kybernetes
Verified
Journal Article
Closed access
HTHuagang Tong
Huagang Tong
CNChaoge Nie
Chaoge Nie
SWSong Wang
Song Wang
Purpose: The purpose of this study is to propose a customer-oriented method for locating charging stations based on queue theory. Design/methodology/approach: We first integrate queuing theory with customer satisfaction to analyze the composition of charging stations. Next, we examine both functional and spatial coordination among these stations. Finally, we develop an enhanced genetic algorithm incorporating flying levy and self-adaptive mechanisms. Findings: Our findings indicate that the customer-oriented approach significantly boosts customer satisfaction. Additionally, improved coordination enhances the feasibility of the charging stations and analyzing the investment payback period contributes to increased profitability. Originality/value: This research uniquely applies queue theory to the selection of charging stations, explores the dynamics of coordination and investigates the payback period of investments.