Projects

Customizable WiFi/WiGig Networks based on Machine Learning

Customizable WiFi/WiGig Networks based on Machine Learning
As the Wi-Fi technology becomes pervasive in reality, its usage pattern is also turning highly diversified in operation settings and application scenarios. This leads to new design requirements for WiFi/WiGig networking solutions in terms of various combinations in throughput, latency and energy efficiency. We believe that both the user demand pull and the technology push call for customizable WiFi/WiGig network solutions. We seek to employ machine learning based approaches (e.g., neural networks) to develop solutions. Our experiments show that they can provide implicit correlations between wireless settings, performance, and channel conditions.

Publications

  1. An Experience Driven Design for IEEE 802.11ac Rate Adaptation based on Reinforcement Learning
    Syuan-Cheng Chen, Chi-Yu Li, Chui-Hao Chiu
    IEEE INFOCOM'21.

  2. Practical Machine Learning-based Rate Adaptation Solution for Wi-Fi NICs: IEEE 802.11ac as a Case Study
    Chi-Yu Li, Syuan-Cheng Chen, Chien-Ting Kuo, Chui-Hao Chiu
    IEEE Transactions on Vehicular Technology (TVT), Vol. 69, Issue 9, pp. 10274-10277, Sept. 2020.

  3. Deep Reinforcement Learning Based Rate Adaptation in 802.11ac Wireless Networks
    Syuan-Cheng Chen, Chi-Yu Li, Chui-Hao Chiu, Yu-Shuo Chang
    Telecom'20.
    Best Paper Award

  4. An Energy Efficiency Perspective on Rate Adaptation for 802.11n NIC
    Chi-Yu Li, Chunyi Peng, Peng Cheng, Songwu Lu, Xinbing Wang, Fengyuan Ren, Tao Wang
    IEEE Transactions on Mobile Computing (TMC), vol. 15, no. 6, p1333-1347, 2016.

  5. Latency-Aware Rate Adaptation in 802.11n Home Networks
    Chi-Yu Li, Chunyi Peng, Guan-Hua Tu, Songwu Lu, Xinbing Wang, Ranveer Chandra
    IEEE INFOCOM'15

Security in 4G/5G Networks: Attacks and Countermeasures

Security in 4G/5G Networks: Attacks and Countermeasures
To support various advanced services, the mobile Internet technology is evolving from a simple, working design to a resilient, high-performance networking solution suite. For example, during the evolution from 2G/3G to 4G, the security impact is not fully understood. My research shows that unprecedented malicious attacks can be launched on both mobile devices and the network infrastructure in 3G/4G systems. It is because security loopholes usually lie in the newly offered services (e.g., IoT and D2D) or those legacy ones (e.g., VoLTE) that require new interactions.

Publications

  1. The Untold Secrets of WiFi-Calling Services: Vulnerabilities, Attacks, and Countermeasures
    Tian Xie, Guan-Hua Tu, Bangjie Yin, Chi-Yu Li, Chunyi Peng, Mi Zhang, Hui Liu, Xiaoming Liu
    IEEE Transactions on Mobile Computing (TMC), Vol. 20, Issue 11, pp. 3131-3147, Nov. 2021.

  2. How Can IoT Services Pose New Security Threats in Operational Cellular Networks?
    Tian Xie, Guan-Hua Tu, Chi-Yu Li, Chunyi Peng
    IEEE Transactions on Mobile Computing (TMC), Vol. 20, Issue 8, pp. 2592-2606, Aug. 2021.

  3. Ghost Calls from Operational 4G Call Systems: IMS Vulnerability, Call DoS Attack, and Countermeasure
    Yu-Han Lu, Chi-Yu Li, Yao-Yu Li, Sandy Hsin-Yu Hsiao, Tian Xie, Guan-Hua Tu, Wei-Xun Chen
    ACM MOBICOM'20.

  4. The Dark Side of Operational Wi-Fi Calling Services
    Tian Xie, Guan-Hua Tu, Chi-Yu Li, Chunyi Peng, Jaiwei Li, Mi Zhang
    IEEE CNS'18.
    Best Paper Award, Google Security Reward

  5. New Security Threats Caused by IMS-based SMS Service in 4G LTE Networks
    Guan-Hua Tu*, Chi-Yu Li* (*:Co-Primary), Chunyi Peng, Yuanjie Li, Songwu Lu
    ACM CCS'16
    Facebook Security Reward

Edge Computing Solutions in 4G/5G Networks

Edge Computing Solutions in 4G/5G Networks
Mobile/Multi-access edge computing (MEC) is one key technology to achieve low-latency performance in cellular networks. It has been determined as a key feature in future 5G networks by both ETSI and 3GPP standardization organizations. It seeks to provide a cloud computing platform at the network edge to be closer to mobile users than conventional cloud systems. Due to emerging low-latency demands, several MEC deployment solutions are being in development. We seek to design an MEC platform that can be easily deployed in 4G LTE networks, as well as may be used as a reference design for future 5G networks.

Publications

  1. Prioritized Traffic Shaping for Low-latency MEC Flows in MEC-enabled Cellular Networks
    Po-Hao Huang, Fu-Cheng Hsieh, Wen-Jen Hsieh, Chi-Yu Li, Ying-Dar Lin
    IEEE CCNC'22.

  2. Transparent AAA Security Design for Low-latency MEC-integrated Cellular Networks
    Chi-Yu Li, Ying-Dar Lin, Yuan-Cheng Lai, Hsu-Tung Chien, Yu-Sheng Huang, Po-Hao Huang, Hsueh-Yang Liu
    IEEE Transactions on Vehicular Technology (TVT), Vol. 69, Issue 3, pp. 3231-3243, March 2020.

  3. A Transparent MEC Testbed for 4G/5G Cellular Networks
    Yu-Sheng Huang, Po-Hao Huang, Yu-Cheng Lai, Hsu-Tung Chien, Chi-Yu Li
    Mobile Computing Workshop'19.
    Best Demo Award

  4. Mobile Edge Computing Platform Deployment in 4G LTE Networks: A Middlebox Approach
    Chi-Yu Li, Hsueh-Yang Liu, Po-Hao Huang, Hsu-Tung Chien, Guan-Hua Tu, Pei-Yuan Hong, Ying-Dar Lin
    USENIX HotEdge'18.

C-V2X Safety Warning System

C-V2X Safety Warning System
The cellular-connected vehicle is forecasted to be a mainstream reality. Its key technology, C-V2X (Cellular Vehicle-to-Everything), enables heterogeneous connectivity and on-device intelligence on the vehicle. One of the visions is to improve road safety. A major focus can be the safety of vulnerable road users (VRU), which include pedestrians, cyclists and motorcyclists. In recent years, they take nearly half (49%) of the people who die on the world's roads. It calls for VRU safety warning services built on top of the C-V2X.

Publications

  1. V2PSense: Enabling Cellular-based V2P Collision Warning Service Through Mobile Sensing
    Chi-Yu Li, Giovanni Salinas, Po-Hao Huang, Guan-Hua Tu, Guo-Huang Hsu, Tien-Yuan Hsieh
    IEEE ICC'18.

Wireless Networking on UAVs

Wireless Networking on UAVs
Our developed UAV platform is the first one equipped with both 4G LTE and WiGig network technologies in the world. It addresses the limitations of current UAV platforms, which rely on only the WiFi technology for wireless communication. The limitations include short control and flight ranges with only several hundred meters, and slow speeds for the communication with servers on the ground. To address the limitations, we integrate 4G LTE and WiGig network technologies into our customized UAV platform. We further aim to design solutions to enable UAV-based mobile APs, which can support Internet access for mobile users on the ground.

Publications

  1. UAV-FAP: User Fairness-Driven Access Point on UAV for Wi-Fi Networks
    Yung-Chuan Wu, Hong-Rong Chang, Meng-Shou Wu, Chi-Yu Li, Kuochen Wang
    IEEE CCNC'22.

  2. WBF-PS: WiGig Beam Fingerprinting for UAV Positioning System in GPS-denied Environments
    Pei-Yuan Hong, Chi-Yu Li, Hong-Rong Chang, YuanHao Hsueh, Kuochen Wang
    IEEE INFOCOM'20.

  3. 3D On-Demand Flying Mobile Communication for Millimeter Wave Heterogeneous Networks
    Kai-Ten Feng, Li-Hsiang Shen, Chi-Yu Li, Po-Tsang Huang, Sau-Hsuan Wu, Li-Chun Wang, Yi-Bing Lin, Mau-Chung Frank Chang
    IEEE Network, Vol. 34, Issue 5, pp. 198-204, September/October 2020.

  4. Communications and Networking Technologies for Intelligent Drone Cruisers
    Li-Chun Wang, Chuan-Chi Lai, Hong-Han Shuai, Hsin-Piao Lin, Chi-Yu Li, Teng-Hu Cheng, Chiun-Hsun Chen
    IEEE Globecom'19 Workshop on SGINs.