대한전기학회:학술대회논문집 (Proceedings of the KIEE Conference)
- 대한전기학회 2005년도 제36회 하계학술대회 논문집 C
- /
- Pages.2327-2329
- /
- 2005
시간지연 신경망을 이용한 기지국용 전력증폭기의 보상기법
Power Amplifier Compensation Technique based on Tapped Delayed Neural Networks
- 황보훈 (성균관대학교 전기전자 및 컴퓨터공학과) ;
- 나완수 (성균관대학교 전기전자 및 컴퓨터공학과) ;
- 양영구 (성균관대학교 전기전자 및 컴퓨터공학과) ;
- 박천석 (성균관대학교 전기전자 및 컴퓨터공학과) ;
- 김병성 (성균관대학교 전기전자 및 컴퓨터공학과)
- HwangBo, Hoon (Electrical and Computer Engineering, Sungkyunkwan University) ;
- Nah, Wan-Soo (Electrical and Computer Engineering, Sungkyunkwan University) ;
- Yang, Youn-Goo (Electrical and Computer Engineering, Sungkyunkwan University) ;
- Park, Cheon-Seok (Electrical and Computer Engineering, Sungkyunkwan University) ;
- Kim, Byung-Sung (Electrical and Computer Engineering, Sungkyunkwan University)
- 발행 : 2005.07.18
초록
In this paper, we identify the memory effects of the RF high-power base station amplifiers with Vector Signal Analyzer (VSA). It is found that the model of power- amplifier using Tapped Delayed Neural - Networks with back-propagation algorithm shows very accurate modeling performance. Based on this behavioral modeling, we conducted inverse compensation process which also uses Neural Networks.
키워드