Jisung Pyo, Jaehyuk Choi
Published in IEEE Wireless Communications Letters (IF=5.5 / Q1), 2025
This paper propose a multi-link-based CSI sampling and integration framework combined with a CSI data loss recovery preprocessing method to enhance network resource efficiency, mitigate data loss, and improve the robustness of Wi-Fi sensing models.
Lawrence Lubwama, Jungik Jang, Jisung Pyo, Joon Yoo, Jaehyuk Choi
Published in MDPI Sensors (IF=3.4 / Q2), 2025
Wi-Filter leverages Wi-Fi CSI and lightweight models for dynamic frame filtering at the edge, enabling resource-efficient and scalable video analytics.
Jungik Jang, Jisung Pyo, Young-Il Yoon, Jaehyuk Choi
Published in IEEE Access (IF=3.4 / Q2), 2024
This paper proposes Meta-Transformer, a meta-learning-based AMC framework using few-shot learning and transformer-based encoders to achieve high performance and scalability with unseen modulations.
Jungik Jang, Jisung Pyo, Young-Il Yoon, Sang Yong Seo, Eun Jae Lee, Gyeong Hun Jung, Jaehyuk Choi
Published in MILCOM 2023-2023 IEEE Military Communications Conference (MILCOM) , 2023
This paper presents a scalable automatic modulation classification (AMC) scheme based on meta-learning and few-shot learning, enabling flexible recognition of unseen modulations with minimal data.
표지성, 최재혁
Published in 2023 한국통신학회 추계종합학술발표회 , 2023
본 연구는 Wi-Fi 센싱 모델의 도메인 종속성을 해결하기 위해 메타 러닝과 트랜스포머를 결합한 방식을 제안하며, 새로운 환경이나 클래스에서도 높은 분류 성능과 일반화 능력을 보장한다.