Myungho Kim
Research Subjects
Myungho Kim has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Ji-Yeon Park
Hanwha System
2 shared publications - Joung-Sik Kim
Hanwha System
2 shared publications - Na-Rae Yi
Hanwha System
2 shared publications - Sung Oh
Kyung Hee University
2 shared publications - Sung-Ho Bae
Kyung Hee University
2 shared publications - Hyungjun Ju1 shared publication
Latest Publications
- 16 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. PP (15209210) , 1-12
AS-RAM: Automated Search with Rich Attention Modules for Network Traffic Prediction
By effectively forecasting network traffic patterns, future cellular networks have the potential to manage themselves and incorporate intelligent, streamlined operations autonomously. Existing methods have relied on handcrafted architectures to predict network traffic, which still have two main limitations. First, human-designed models may have sub-optimality in modeling. Second, existing handcrafted designs did not take spatio-temporal long-term dependencies into account explicitly even though they play crucial roles in network traffic prediction. To address these issues, we propose a method using NAS (Neural Architecture Search) to find more optimized architectures than handcrafted ones for network traffic prediction. Specifically, we introduce a new search space that integrates attention modules enabling better extraction and reflection of the unique characteristics of network traffic data. These attention modules are seamlessly incorporated into the architecture from an optimization perspective. To verify the effectiveness of the proposed method, we validate the performance of the proposed framework using the Italia Telecom dataset. Experimental results show that our framework significantly improves the performance compared to the baseline in terms of root mean squared error between predicted value and ground-truth value. Specifically, our method improves the performance by 9.5%, 9.8%, and 10.2% on SMS, CALL, and INTERNET traffic datasets, respectively.
- 16 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. PP (15209210) , 1-12
AS-RAM: Automated Search with Rich Attention Modules for Network Traffic Prediction
By effectively forecasting network traffic patterns, future cellular networks have the potential to manage themselves and incorporate intelligent, streamlined operations autonomously. Existing methods have relied on handcrafted architectures to predict network traffic, which still have two main limitations. First, human-designed models may have sub-optimality in modeling. Second, existing handcrafted designs did not take spatio-temporal long-term dependencies into account explicitly even though they play crucial roles in network traffic prediction. To address these issues, we propose a method using NAS (Neural Architecture Search) to find more optimized architectures than handcrafted ones for network traffic prediction. Specifically, we introduce a new search space that integrates attention modules enabling better extraction and reflection of the unique characteristics of network traffic data. These attention modules are seamlessly incorporated into the architecture from an optimization perspective. To verify the effectiveness of the proposed method, we validate the performance of the proposed framework using the Italia Telecom dataset. Experimental results show that our framework significantly improves the performance compared to the baseline in terms of root mean squared error between predicted value and ground-truth value. Specifically, our method improves the performance by 9.5%, 9.8%, and 10.2% on SMS, CALL, and INTERNET traffic datasets, respectively.
- Open Access
- 30 September 2024
- Published by The Korean Institute of Broadcast and Media Engineers in Journal of Broadcast Engineering
- Vol. 29 (5) , 654-661
A Correlation-based Tabular data Augmentation method for Network QoS Prediction
- 30 September 2024
- Published by The Korean Institute of Broadcast and Media Engineers in Journal of Broadcast Engineering
- Vol. 29 (5) , 654-661
Open AccessA Correlation-based Tabular data Augmentation method for Network QoS Prediction
- 31 October 2023
- Published by Korean Institute of Information Technology in The Journal of Korean Institute of Information Technology
- Vol. 21 (10) , 117-126
Study for Machine Learning-based Demodulation Scheme Robust to Electromagnetic Interference
전자기 스펙트럼 작전 환경에서는 통신 시스템의 성능 확보를 위해 레이더 간섭을 보상할 수 있는 기술이 필요하다. 본 논문에서는 레이더 펄스 신호가 존재할 때 통신 시스템의 수신기에서 간섭 신호의 영향을 학습하여 오류 발생을 감소시킬 수 있는 기계 학습 기반의 복조 구조를 제안한다. 제안하는 복조기는 심층 신경망을 이용하여 구성되며, 은닉층의 개수와 노드 개수에 따른 심볼 오류율 성능을 분석하여 저복잡도의 복조 신경망 구성 방안을 제시한다. 컴퓨터 시뮬레이션 결과를 통해 간섭 대 신호비가 0.5보다 큰 상황에서 신호 대잡음비 25 dB 기준으로 기존의 최대 우도 복조 방식은 약 10%의 심볼 오류율을 나타내었으나, 제안하는 기계학습 기반 복조는 0.01%의 심볼 오류율을 나타내는 것을 확인하였다.
- 31 October 2023
- Published by Korean Institute of Information Technology in The Journal of Korean Institute of Information Technology
- Vol. 21 (10) , 117-126
Study for Machine Learning-based Demodulation Scheme Robust to Electromagnetic Interference
전자기 스펙트럼 작전 환경에서는 통신 시스템의 성능 확보를 위해 레이더 간섭을 보상할 수 있는 기술이 필요하다. 본 논문에서는 레이더 펄스 신호가 존재할 때 통신 시스템의 수신기에서 간섭 신호의 영향을 학습하여 오류 발생을 감소시킬 수 있는 기계 학습 기반의 복조 구조를 제안한다. 제안하는 복조기는 심층 신경망을 이용하여 구성되며, 은닉층의 개수와 노드 개수에 따른 심볼 오류율 성능을 분석하여 저복잡도의 복조 신경망 구성 방안을 제시한다. 컴퓨터 시뮬레이션 결과를 통해 간섭 대 신호비가 0.5보다 큰 상황에서 신호 대잡음비 25 dB 기준으로 기존의 최대 우도 복조 방식은 약 10%의 심볼 오류율을 나타내었으나, 제안하는 기계학습 기반 복조는 0.01%의 심볼 오류율을 나타내는 것을 확인하였다.