FCFuxing Chen

Fuxing Chen

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

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


Fuxing Chen has a total of 26 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    A Systematic Comparison of Large Language Models Performance for Intrusion Detection

    by Minh-Thanh Bui, Matteo Boffa, Rodolfo Vieira Valentim, Jose Manuel Navarro, Fuxing Chen, Xiaosheng Bao, Zied Ben Houidi, Dario Rossi

    We explore the capabilities of Large Language Models (LLMs) to assist or substitute devices (i.e., firewalls) and humans (i.e., security experts) respectively in the detection and analysis of security incidents. We leverage transformer-based technologies, from relatively small to foundational sizes, to address the problem of correctly identifying the attack severity (and accessorily identifying and explaining the attack type). We contrast a broad range of LLM techniques (prompting, retrieval augmented generation, and fine-tuning of several models) using state-of-the-art machine learning models as a baseline. Using proprietary data from commercial deployment, our study provides an unbiased picture of the strengths and weaknesses of LLM for intrusion detection.


  • Open Access

    Change Point Detection in WLANs with Random AP Forests

    by Alexis Huet, Jonatan Krolikowski, Jose Manuel Navarro, Fuxing Chen, Dario Rossi

    Troubleshooting WiFi networks is knowingly difficult due to the variability of the wireless medium. Complementary to existing works that focus on detecting short-term fluctuations of radio signals (i.e., anomalies), we tackle the problem of reliably detecting long-term changes in statistical properties of WiFi networks. We propose a new method to reliably gain insights on such environmental changes, which we refer to as Random Access Point Forest (RAPF). RAPF identifies the changes from a forest of individual learners, each of them consisting of a random tree approximating the signal of a specific pair of APs. The biased selection of APs in a distributed manner along with the stochastic construction of each individual tree ensure its robustness to noise and biases. We conduct a measurement campaign on a real WLAN by collecting the path loss among pairs of APs in a network for which labels are available and perform an extensive comparison of our methodology against state-of-the-art change point methodologies, which conclusively shows RAPF to yield the most robust detection capabilities.

  • Open Access

    Replication: Contrastive Learning and Data Augmentation in Traffic Classification Using a Flowpic Input Representation

    by Alessandro Finamore, Chao Wang, Jonatan Krolikowski, Jose M. Navarro, Fuxing Chen, Dario Rossi


  • Open Access

    A Lightweight, Efficient and Explainable-by-Design Convolutional Neural Network for Internet Traffic Classification

    by Kevin Fauvel, Fuxing Chen, Dario Rossi

    Traffic classification, i.e., the identification of the type of applications flowing in a network, is a strategic task for numerous activities (e.g., intrusion detection, routing). This task faces some critical challenges that current deep learning approaches do not address. The design of current approaches do not take into consideration the fact that networking hardware (e.g., routers) often runs with limited computational resources. Further, they do not meet the need for faithful explainability highlighted by regulatory bodies. Finally, these traffic classifiers are evaluated on small datasets which fail to reflect the diversity of applications in real-world settings. Therefore, this paper introduces a new Lightweight, Efficient and eXplainable-by-design convolutional neural network (LEXNet) for Internet traffic classification, which relies on a new residual block (for lightweight and efficiency purposes) and prototype layer (for explainability). Based on a commercial-grade dataset, our evaluation shows that LEXNet succeeds to maintain the same accuracy as the best performing state-of-the-art neural network, while providing the additional features previously mentioned. Moreover, we illustrate the explainability feature of our approach, which stems from the communication of detected application prototypes to the end-user, and we highlight the faithfulness of LEXNet explanations through a comparison with post hoc methods.


  • Open Access

    A Lightweight, Efficient and Explainable-by-Design Convolutional Neural Network for Internet Traffic Classification

    by Kevin Fauvel, Fuxing Chen, Dario Rossi

    Traffic classification, i.e. the identification of the type of applications flowing in a network, is a strategic task for numerous activities (e.g., intrusion detection, routing). This task faces some critical challenges that current deep learning approaches do not address. The design of current approaches do not take into consideration the fact that networking hardware (e.g., routers) often runs with limited computational resources. Further, they do not meet the need for faithful explainability highlighted by regulatory bodies. Finally, these traffic classifiers are evaluated on small datasets which fail to reflect the diversity of applications in real-world settings. Therefore, this paper introduces a new Lightweight, Efficient and eXplainable-by-design convolutional neural network (LEXNet) for Internet traffic classification, which relies on a new residual block (for lightweight and efficiency purposes) and prototype layer (for explainability). Based on a commercial-grade dataset, our evaluation shows that LEXNet succeeds to maintain the same accuracy as the best performing state-of-the-art neural network, while providing the additional features previously mentioned. Moreover, we illustrate the explainability feature of our approach, which stems from the communication of detected application prototypes to the end-user, and we highlight the faithfulness of LEXNet explanations through a comparison with post hoc methods.

  • Open Access

    A Cross-Sectional Survey on Cigarette Smoking in the Chinese Navy

    by Fuxing Chen, Pengwei Hu, Wang Chang, Chulin Chen, Tianyuan Ding, Xiaorong Liu

    The military is a risk environment for cigarette smoking. This study determined the prevalence of, cessation attempts, knowledge, and attitudes regarding smoking, and factors related to it in the Chinese navy. A cross-sectional survey was conducted of military personnel on active duty in the Chinese navy, Guangdong province. Data were collected using a standard structured questionnaire and analyzed through SPSS. Multiple logistic regression was used to examine factors associated with smoking. The research protocols were reviewed and approved by the Institutional Review Board of Naval Medical University, and participants gave written informed consent. A total of 1,530 military personnel who had served for >3 months completed the survey, 859 (56.14%) of whom were current smokers. The main reasons for smoking were stress alleviation and social needs. A total of 377 smokers had previously attempted quitting. More than half considered the military environment supportive of smoking. Subjects’ knowledge regarding smoking mainly comprised negative respiratory-health effects. Smoking status was related to education level, military service duration, number of smoking family members, and smoking knowledge. Smoking prevalence in the Chinese navy is higher than that among Chinese civilians and the US and UK military personnel. Although many sailors are aware of the dangers of smoking to their health, they seldom get advice or help regarding smoking cessation. Most smokers believed that smoking cessation could improve their health. Therefore, the navy should formulate policies encouraging and facilitating military personnel’s smoking cessation.


  • Open Access

    Conversion between 2 military combat-related injury coding systems

    by Pengwei Hu, Fuxing Chen, Wang Chang, Tai Xie, Jinhui Liu, Zhiping An, Guoliang Chen, Xiaorong Liu

    Deployable medical systems(DEPMEDS) patient conditions (PCs) and Military Combat Injury Scale (MCIS) are 2 important military medical coding systems. However, both of them have defects when applied in military medical planning. Although each PC code contains information about treatment, intensity of care, treatment time, length of stay, and probability of disposition that is relevant to simulation, its description is too comprehensive and ambiguous to code historical military medical records. Therefore, conversion between PC and other medical coding systems applied in standard medical data is required when validity is required following simulation. The information linked to each PC code is based on subject matter expert opinion instead of standard medical data from the theater that need to be continuously updated depending on the results of medical data analysis. MCIS, a combat-related injury coding system, shows significant promise in coding real medical data, but it does not seek detailed information important for prediction and simulation unlike PCs. Therefore, MCIS cannot be used in planning tools directly. Thus, the effort to map MCIS to PCs is significant for medical logistic planning. We aim to identify whether conversion between PCs and MCIS is possible and to evaluate inter-coder reliability. Three senior coders assigned all possible MCIS codes to 187 combat-related PC codes. The data records were structured based on an earlier study. Inter-rater reliability was measured by using Cohen's k statistic and percent agreement. Low inter-rater reliability indicated the difficulty in conversion between PCs and MCIS. The injury descriptors of PCs should be expanded by referring to new standard medical data. The existing MCIS codes need to be modified to include more information on treatment brief, treatment time, length of stay, and other key information, and historical data statistics need to be developed.


  • Improving TCP responsiveness with connection history in data center networks

    by Dagang Li, Jiaxiang Wang, Fuxing Chen

    Data center networks are high-bandwidth, low-latency networks with clear-defined topology and busy internal data exchange traffic. TCP is widely used in data centers to transport data, however its default behavior during the flow initialization phase may be unsuitable for data center networks and affect the responsiveness of data transmission, especially when the flow is short. Considering that data center flows happen frequently but only between limited number of data center servers connected by a rather stable physical topology, in this paper we propose to use the RTT records from previous transmissions to help TCP calculate more suitable initialization parameters when a new flow starts, so as to avoid unnecessary lengthy timeouts that has heavy consequences on the overall performance, especially on short flows. Experiments show that our algorithms can indeed eliminate the long timeouts caused by early packet losses, and improve the responsiveness and stability of data transmission in a data center networking environment.