Current Result Document :
| 한글제목(Korean Title) |
이기종 사물 인터넷 환경에서의 효율적인 네트워크 슬라이싱 리소스 관리를 위한 협업 학습 |
| 영문제목(English Title) |
A Collaborative Learning for Efficient Network Slicing Resource Management in Heterogeneous IoT Environment |
| 저자(Author) |
담프로힘
맛사
김석훈
Prohim Tam
Sa Math
Seokhoon Kim
|
| 원문수록처(Citation) |
VOL 22 NO. 01 PP. 0121 ~ 0122 (2021. 04) |
한글내용
(Korean Abstract) |
|
영문내용
(English Abstract) |
With the widespread Internet of Things (IoT) utilization in numerous scenarios and application services, the management and orchestration entities require upgrading the conventional architecture and emerge with intelligent models and ultra-reliable mechanisms. Mission-critical IoT applications are significant to consider in the heterogeneous network environment. With false priority and high failure rates, serious losses in terms of human lives in emergent situations, great business assets, and privacy leakage will occur. The proposed Collaborative Learning for Efficient Network Slicing (CL-ENS) scheme tackled the problems by converging
Support Vector Regression (SVR) learning in decentralized controllers, Deep Neural Network (DNN) in centralized global control, and SDN-enabled resource management for each critical slice. To demonstrate the theoretical approach, Mininet emulator was conducted to evaluate models and capture the performance metrics.
|
| 키워드(Keyword) |
|
| 파일첨부 |
PDF 다운로드
|
