標題: 提升無線大型網路容量之傳接收策略研究
Transmitting and Receiving Strategies for Throughput Enhancement in Large-Scale Wireless Networks
作者: 楊子寬
劉俊宏
Zhu-Kuan Yang
Liu,Chun-Hung
電控工程研究所
關鍵字: 隨意網路;隨機幾何;功率控制;干擾消除;機會式排程法;Ad hoc networks;Stochastic geometry;Power control;Interference cancellation;Opportunistic scheduling
公開日期: 2016
摘要: 本論文研究數個在無線隨意網路下之傳接收策略,藉此增加網路效能如傳輸成功率及傳輸速率。我們觀察到通道增益的隨機特性有益於提升平均的訊擾比,因而提出了一些相關的方法來利用通道隨機性。為了增強訊擾比,我們可以透過增強接收訊號強度抑或是抑制干擾的生成。在抑制干擾的方面,我們想出了扣除前K個最強干擾發射端的方法以增加訊擾比。傳輸成功機率及傳輸速率的一般化表示式、非常貼近實際值的上限同時也被推導於本論文中。利用排程法確保傳輸的品質亦是一個增加訊擾比的有效策略。而本論文亦介紹了一個分布性機會式排程法,其利用通道感測資訊進行排程並使用了兩個依據密度之動態閥值,避免傳輸通道增益過弱及干擾影響過強的傳輸情形。本論文所提出的策略研究都大幅地提升傳輸成功率及傳輸速率,其效能分析皆被模擬結果所驗證。
In this thesis, we study some transmitting and receiving strategies for enhancing network performance such as success transmission probability and ergodic link rate in wireless ad hoc networks. We observe the phenomenon that channel randomness can benefit mean signal-to-interference power ratio (SIR) and some methods are proposed to exploit the channel randomness. To increase the SIR, we can either enhance received signal power or suppress the interference. In the view of interference suppression, we come up with an idea to cancel the first K strongest interfering transmitters to improve the SIR. The general expressions of the success probability and ergodic link rate are derived and their tight bounds are also found in this thesis. Using scheduling schemes is also an effective strategy to make sure each transmission is of good quality. In addition, distributed opportunistic scheduling, which is a channel-aware scheduling scheme with two dynamic intensity-based thresholds to refrain the transmissions from the transmitters with weak communication channel gains and/or strong interfering impacts, is proposed in this thesis. The strategies proposed in this thesis significantly improve the success probability and ergodic link rate and their performance is validated by simulation results.
URI: http://etd.lib.nctu.edu.tw/cdrfb3/record/nctu/#GT070360047
http://hdl.handle.net/11536/143196
顯示於類別:畢業論文