Kanghee Lee
Research Subjects
Kanghee Lee has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
Ara Khil
Soongsil University
1 shared publication- Bong Joo Kang
Division of Advanced Materials Korea Research Institute of Chemical Technology (KRICT) Daejeon Republic of Korea
1 shared publication - Bumki Min
Korea Advanced Institute of Science and Technology
1 shared publication - Fabian Rotermund
Department of Physics, Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Republic of Korea
1 shared publication - Hyun-Chul Kim
Department of Artificial Intelligence, Kyungpook National University, Daegu, Republic of Korea
1 shared publication - Jaesung Lim
Ajou University
1 shared publication
Latest Publications
- Open Access
- 10 August 2020
- Published by Association for Computing Machinery (ACM)
- p. 1-2
Traffic classification using deep learning
As Deep Learning (DL) algorithms have rapidly become a methodology of choice in various domains, they have recently entered also the field of the Internet traffic classification, successfully demonstrating impressive results. Most of the research work up to this point has focused on improving the accuracy of classification systems, yet there has been little attempt to provide (i) systematic comparison of the various DL algorithms used and (ii) analysis on where the higher accuracy come from, particularly when comparing with the traditional machine learning algorithms like C4.5. To fill this gap, we conduct experiments with four DL algorithms proposed for traffic classification, including CNN, LSTM, Stacked Auto-Encoder (SAE), and Hierarchical Attention Networks (HAN). Further, we propose to leverage and visualize hierarchical attention layers to highlight which parts of the traffic packet traces were most informative for accurate classification, which provides hints about why (and how) DL algorithms achieve the state-of-the-art level high accuracy. We view this paper as the first step towards answering the aforementioned "why" question, which is critical in understanding the real benefit and contribution of deep learning to the field of the Internet traffic classification, and advancing its state-of-the-art.
- 10 August 2020
- Published by Association for Computing Machinery (ACM)
- p. 1-2
Open AccessTraffic classification using deep learning
As Deep Learning (DL) algorithms have rapidly become a methodology of choice in various domains, they have recently entered also the field of the Internet traffic classification, successfully demonstrating impressive results. Most of the research work up to this point has focused on improving the accuracy of classification systems, yet there has been little attempt to provide (i) systematic comparison of the various DL algorithms used and (ii) analysis on where the higher accuracy come from, particularly when comparing with the traditional machine learning algorithms like C4.5. To fill this gap, we conduct experiments with four DL algorithms proposed for traffic classification, including CNN, LSTM, Stacked Auto-Encoder (SAE), and Hierarchical Attention Networks (HAN). Further, we propose to leverage and visualize hierarchical attention layers to highlight which parts of the traffic packet traces were most informative for accurate classification, which provides hints about why (and how) DL algorithms achieve the state-of-the-art level high accuracy. We view this paper as the first step towards answering the aforementioned "why" question, which is critical in understanding the real benefit and contribution of deep learning to the field of the Internet traffic classification, and advancing its state-of-the-art.
- 1 January 2017
- Published by Optica Publishing Group
- p. NW3A.2
Electrically Controlled Second-Order Nonlinear Generation of Terahertz Waves
We report electric-field-induced second-harmonic generation of terahertz waves in photodoped gallium arsenide with direct-current bias fields. A fundamental wave of 0.6 THz was selectively incident and its second-harmonic wave of 1.2 THz was clearly observed.
- 1 January 2017
- Published by Optica Publishing Group
- p. NW3A.2
Electrically Controlled Second-Order Nonlinear Generation of Terahertz Waves
We report electric-field-induced second-harmonic generation of terahertz waves in photodoped gallium arsenide with direct-current bias fields. A fundamental wave of 0.6 THz was selectively incident and its second-harmonic wave of 1.2 THz was clearly observed.
- 1 May 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
MMSE-Based Distributed Beamforming in AF Relay System under Network Power Constraint
An amplify-and-forward (AF) wireless relay strategy consisting of a source with multiple M antennas, distributed N relays with only a single antenna, and a destination with multiple L antennas is investigated in this paper. This paper presents an optimal relay amplifying matrix at the relays, transmit beamforming vector at the source, and receive beamforming vector at the destination under the network power constraint of the source and relays using the minimum mean square error (MMSE) criterion. Then, this paper proves that, for the considered amplify-and-forward (AF) wireless relaying system, minimizing the mean square error (MSE) is equivalent to maximizing the signal-to-noise ratio (SNR). Additionally, this paper presents the relationship between the achievable rate and the MMSE. This paper also investigates the relay amplifying matrix, SNR, and MMSE cost function (CF) behaviors numerically and analytically. Finally, this paper numerically and analytically shows that the MMSE CF values are less than 1 at all times, regardless of N.
- 1 May 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-5
MMSE-Based Distributed Beamforming in AF Relay System under Network Power Constraint
An amplify-and-forward (AF) wireless relay strategy consisting of a source with multiple M antennas, distributed N relays with only a single antenna, and a destination with multiple L antennas is investigated in this paper. This paper presents an optimal relay amplifying matrix at the relays, transmit beamforming vector at the source, and receive beamforming vector at the destination under the network power constraint of the source and relays using the minimum mean square error (MMSE) criterion. Then, this paper proves that, for the considered amplify-and-forward (AF) wireless relaying system, minimizing the mean square error (MSE) is equivalent to maximizing the signal-to-noise ratio (SNR). Additionally, this paper presents the relationship between the achievable rate and the MMSE. This paper also investigates the relay amplifying matrix, SNR, and MMSE cost function (CF) behaviors numerically and analytically. Finally, this paper numerically and analytically shows that the MMSE CF values are less than 1 at all times, regardless of N.
- 1 January 2013
- Published by Springer Nature
- p. 995-1002
Optimization of a Cart Capacity Using the Three-Dimensional Single Bin Packing Problem for a Serving Robot
- 1 January 2013
- Published by Springer Nature
- p. 995-1002
Optimization of a Cart Capacity Using the Three-Dimensional Single Bin Packing Problem for a Serving Robot