Research Projects
- On-Device AI with Collaborative Learning and Prediction over Wirelessly Connected Edge Devices
We consider distributed deep learning architecture for scalable learning and inference capability at resource-constrained edge devices. To avoid vulnerable data breach and massive data upload burden problems caused by the widely used centralized approach, we take a holistically new approach from both computing and networking perspectives in a distributed fashion. We aim to implement a lightweight collaborative learning architecture by recruiting available edge devices and forming an agile ad-hoc resource network for distributed learning.
|
- Energy-Efficient Data Delivery in Wireless Ad-Hoc/Sensor/Fog/UAV Networks
The goal of this research project is to design optimal routing algorithms by mining users' behavior/movement patterns, social relationship among users, group movement patterns, and network behavior. We apply various statistical modeling, machine learning, and algorithmic techniques to investigate the underlying structures from empirical data. We exploit them to implement efficient data delivery mechanisms practically feasible in real-world embedded networked systems with respect to energy efficiency and QoS delivery.
|
- Network Optimization in Heterogeneous Networks
As wireless handheld devices have embedded a variety of radio access interfaces such as 5G, LTE, Wi-Fi (including 802.11ac/ad), ZigBee, and NFC, Heterogeneous Network (HetNet) has recently attracted significant attention. We work on the problems of interoperation among different network interfaces and inter-Radio Access Technology (RAT) handover. We aim to improve user throughput, system throughput, and delay performance by fully utilizing network resources together with fairness. We investigate scheduling, resource allocation, inter-RAT handover, and interference management by taking optimization techniques and game theoretical approaches.
|
Publications
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
International Patents
- HyungJune Lee, Hyun Seok Kim, Ye Lim Youn, and Ik Joon Chang, "Ad-Hoc Network System Using Selective Data Compression Algorithm, and Data Transmission Method in Ad-Hoc Network," U.S. Patent US 14/443,024 (pending), filed May 14, 2015.
- HyungJune Lee and Justin Aimin Sang, "Method and System for Efficiently Scheduling Short Range Wireless Data Transmissions," U.S. Patent US8743860 B2, filed June 21, 2011, issued Jun 3, 2014.
- HyungJune Lee and Rajesh Kumar Sinha, "Method and System for Reliable Service Period Allocation in 60GHz MAC," U.S. Patent US8446825 B2, filed June 29, 2011, issued May 21, 2013.
Copyright Notice: Since most of these papers are published, the copyright has been transferred to the respective publishers. Therefore, the papers cannot be duplicated for commercial purposes. The following is ACM's copyright notice; other publishers have similar ones.
Copyright ©200x by the Association for Computing Machinery, Inc. Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that new copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted.
Copyright ©200x by the Association for Computing Machinery, Inc. Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that new copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted.