透過機器學習降低NB-IoT網路的通道干擾之研究;The Study of reduction of channel interference in NB-IoT networks through Machine Learning

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Communication Engineering > Electronic Thesis & Dissertation > Item 987654321/77388

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/77388

Title:	透過機器學習降低NB-IoT網路的通道干擾之研究;The Study of reduction of channel interference in NB-IoT networks through Machine Learning
Authors:	鍾元禎;YUAN-ZHEN, ZHONG
Contributors:	通訊工程學系
Keywords:	LTE;NB-IoT;機器學習;功率分配;LTE;NB-IoT;Machine Learning;Power Allocation
Date:	2018-07-18
Issue Date:	2018-08-31 14:36:25 (UTC+8)
Publisher:	國立中央大學
Abstract:	隨著4G邁入5G的時代，然而因為5G得到來，將為物聯網帶來極大的衝擊，能夠有更多的支援無線設備的連入和支援龐大的數據流量，並且以低於現今成本的聯網服務。而在5G正式上路之前，低功耗廣域網路(Low Power Wide Area Network,LPWAN)提供產業持續前進的動力，其中較受注目的包括使用授權頻段的NB-IoT(NarrowBand-IoT) 。近來，NB-IoT (FDD)的應用可能會受到在LTE (TDD)的架構下影響，當NB-IOT小基地台利用LTE的頻段進行上行資料傳送，在使用LTE的頻段的時候，會受到附近的LTE大基地台(Marco cell)的下行功率溢漏干擾。為此，本篇論文即是研究在NB-IoT (FDD)在LTE (TDD)的環境下，如何有效且精準的配置功率。最後模擬的結果可以看出本論文提出的KM-Q演算法在任何溢漏環境下都能夠配置出較佳的功率參數，並建立溢漏迴歸模型作為防止溢漏干擾之最終目標。;The 5G is coming soon after the 4G, the 5G will bring a great impact on the Internet of Things.It will be able to support the connection of wireless devices and support huge data traffic. It will provide IoT services at lower cost than today. Before the 5G is opened officially,The LPWAN (Low Power Wide Area Network,LPWAN) provides the power which the industry goes forward continually. It is worth noting that the use of licensed band NB-IoT (NarrowBand-IoT). Recently, the NB-IoT (FDD) application possibly can be affected by the LTE (TDD) construction. When the NB-IoT small base station uses the LTE frequency band for the uplink transmission.It can be interference by downlink power to leak from nearby the LTE eNB (Marco cell). For this reason, this paper proposed the NB-IoT (FDD) how to effectively and accurately allocate the power under the LTE (TDD) environment. Finally,it simulates the result indication the KM-Q algorithm which this paper proposed in any leakage environment all to be able to show the better power parameter, and establishes the leakage regression model achievement to prevent leakage the interference for the ultimate goal.
Appears in Collections:	[Graduate Institute of Communication Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	161	View/Open

社群 sharing

Loading...