Fanypn Liu

Dr. Fanypn Liu

orcid

Research Focus & Metrics

Academic Career Length

14Years (2013-Present)

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

Representative Publications

View All Publications
8 June 2022 in Symmetry
Verified
Journal Article
Open access
PDF file
Hsiu-Min ChuangHsiu-Min Chuang
Hsiu-Min Chuang
FLFanypn Liu
Fanypn Liu
CTChung-Hsien Tsai
Chung-Hsien Tsai
Recent developments have made software-defined networking (SDN) a popular technology for solving the inherent problems of conventional distributed networks. The key benefit of SDN is the decoupling between the control plane and the data plane, which makes the network more flexible and easier to manage. SDN is a new generation network architecture; however, its configuration settings are centralized, making it vulnerable to hackers. Our study investigated the feasibility of applying artificial intelligence technology to detect abnormal attacks in an SDN environment based on the current unit network architecture; therefore, the concept of symmetry includes the sustainability of SDN applications and robust performance of machine learning (ML) models in the case of various malicious attacks. In this study, we focus on the early detection of abnormal attacks in an SDN environment. On detection of malicious traffic in SDN topology, the AI module in the topology is applied to detect and act against the attack source through machine learning algorithms, making the network architecture more flexible. Under multiple abnormal attacks, we propose a hierarchical multi-class (HMC) architecture to effectively address the imbalanced dataset problem and improve the performance of minority classes. The experimental results show that the decision tree, random forest, bagging, AdaBoost, and deep learning models exhibit the best performance for distributed denial-of-service (DDoS) attacks. In addition, for the imbalanced dataset problem of multiclass classification, our proposed HMC architecture performs better than previous single classifiers. We also simulated the SDN topology and scenario verification. In summary, we concatenated the AI module to enhance the security and effectiveness of SDN networks in a practical manner.
1

10 September 2021 in Symmetry
Verified
Journal Article
Open access
PDF file
FLFanypn Liu
Fanypn Liu
Wireless sensor networks (WSNs) are the cornerstone of the current Internet of Things era. They have limited resources and features, a smaller packet size than other types of networks, and dynamic multi-hop transmission. WSNs can monitor a particular area of interest and are used in many different applications. For example, during the COVID-19 pandemic, WSNs have been used to measure social distancing/contact tracing among people. However, the major challenge faced by WSN protocols is limited battery energy. Therefore, the whole WSN area is divided into odd clusters using k-means++ clustering to make a majority rule decision to reduce the amount of additional data sent to the base station (or sink) and achieve node energy-saving efficiency. This study proposes an energy-efficient binarized data aggregation (EEBDA) scheme, by which, through a threshold value, the collected sensing data are asserted with binary values. Subsequently, the corresponding cluster head (CH), according to the Hamming weight and the final majority decision, is calculated and sent to the base station (BS). The EEBDA is based on each cluster and divides the entire WSN area into four quadrants. All CHs construct a data-relay transmission link in the same quadrant; the binary value is transferred from the CHs to the sink. The EEBDA adopts a CH rotation scheme to aggregate the data based on the majority results in the cluster. The simulation results demonstrate that the EEBDA can reduce redundant data transmissions, average the energy consumption of nodes in the cluster, and obtain a better network lifetime when compared to the LEACH, LEACH-C, and DEEC algorithms.

24 August 2016 in International Journal of Intelligent Systems
Verified
Journal Article
Closed access
PDF file
SCSheng-Tao Chen
Sheng-Tao Chen
TYTing-Jung Yu
Ting-Jung Yu
LCLiang-Chu Chen
Liang-Chu Chen
FLFanypn Liu
Fanypn Liu


23 July 2021
Verified
Conference Paper
Closed access
FLFanypn Liu
Fanypn Liu
STSsu-Jung Ting
Ssu-Jung Ting
YCYu-Ting Cheng
Yu-Ting Cheng
Wireless sensor networks (WSNs) are an emerging technology in the mobile ad hoc network as used for many different applications. It has several features, such as limited resources and features, smaller packet size, and dynamic multi-hop transmission. During the worldwide COVID-19 pandemic, WSNs have been used to measure social distancing/contact tracking among people. WSNs are also used to monitor the environment as a part of military applications. However, the major challenge that WSN protocols face lies in the limited battery energy for sensor networks. To effectively transfer data onto a base station (or “sink”) and achieve the efficiency of node energy saving, we group neighboring nodes into the same clusters with a K-means++ clustering and reduce the additional data sent to the sink. Therefore, this study proposes an energy-efficient data aggregation mechanism (EDAM) with a spatial correlation among sensor nodes to avoid high correlation among data and reduce node redundancy data transmission. The EDAM depends on the threshold of the first-order radio model's distance to divide the whole WSN area into several clusters. All cluster heads (CHs) construct a data relay transmission link in a WSN. The EDAM is based on the center of the whole sensing area and divides the WSN area into four quadrants. All data is transferred from all sensors to a CH in a quadrant. Then, the CH in each quadrant forwards the data it receives to its next quadrant. The EDAM adopts a quadrant rotation mechanism to aggregate the data in the four quadrants to the base station. The simulation results show that the EDAM reduces redundant data transmissions, averages the power consumption of nodes in the cluster, and obtains a better overall network lifetime than the LEACH, LEACH-C, and DEEC algorithms.