透過室內定位和機器學習方法減少毫微型基地台間干擾之研究;Interference Reduction Assisted by Indoor Localization and Machine Learning Techniques

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Communication Engineering > Electronic Thesis & Dissertation > Item 987654321/74538

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/74538

Title:	透過室內定位和機器學習方法減少毫微型基地台間干擾之研究;Interference Reduction Assisted by Indoor Localization and Machine Learning Techniques
Authors:	邱聖凱;Chiu, Sheng-Kai
Contributors:	通訊工程學系
Keywords:	毫微型基地台;室內定位;機器學習;femtocell;indoor localization;machine learning
Date:	2017-06-29
Issue Date:	2017-10-27 14:01:26 (UTC+8)
Publisher:	國立中央大學
Abstract:	近年來，高品質的行動通訊需求大量增加，未來從4G邁向5G時代，為了要滿足大量且各式各樣的需求，會有許多不同後端網路支援與不同發射功率的基地台共存，異質網路架構不僅可以幫助運營商布建超密集網路，也可以滿足越來越高的頻寬速度及用戶體驗需求，做到更深度的最佳化。其中，毫微型基地台網路(femtocell network)被認為適用於下一代行動通訊的室內傳輸，原因是因為它體型輕巧，能安置在辦公室與住宅，可解決行動通訊覆蓋率不佳及室內傳輸量不足等棘手問題，以提高無線資源重複使用的效率。毫微型基地台在3G末已被提出討論，但遲遲無法發展的原因即是其干擾協調的問題，干擾主要分為來自大型基地台的干擾與毫微型基地台彼此之間互相干擾等兩種情況，這些干擾會影響到使用者聯網品質，造成網路延遲、斷線以及訊號消失等問題。在本篇論文我們將討論關於毫微型基地台彼此之間互相干擾的同層干擾問題，本論文將提出一個系統架構，搭配室內定位、機器學習中的分群演算法和已被提出的干擾協調演算法，透過位置紀錄和學習的方式預測裝置的行動，並提早進行功率控制相關的干擾協調演算法，以減少行動裝置在環境中受到的干擾。;Recently, the demand for high-quality mobile communications has significantly increased. From the 4th generation of mobile phone mobile communication technology standards to the future 5th, in order to meet the large and wide variety of needs, there will be many different back-end networks and different power of base stations coexist. Therefore, heterogeneous network not only help Telecommunications corporations building Ultra-dense network, but could also fulfill the upcoming high-speed Internet access and user experience needs. Femtocell network are considered suitable for next-generation mobile communication because of its small size, which can solve inadequate coverage of mobile communication problem. Femtocell network has been discussed at the end of 3G, but the reason that it could not have been developed is the problem of interference coordination. Interference is divided into two categories, the first is the interference from large base stations and the second is the interference between femtocell base stations, which will both affect network quality of users and cause signal disconnection problems. We are going to discuss the second problem mentioned above. In this thesis, we will present a system including an indoor localization method, clustering methods from machine learning, and two widely used inter-cell interference coordination algorithms. Through this system, we could reduce the interference between femtocell base stations.
Appears in Collections:	[Graduate Institute of Communication Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	273	View/Open

社群 sharing

Loading...