基於代價賽局在裝置對裝置間通訊下之資源分配與使用者劃分;Pricing Game-Based Resource Allocation and User Grouping for Device-to-Device Communications

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Communication Engineering > Electronic Thesis & Dissertation > Item 987654321/65555

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/65555

Title:	基於代價賽局在裝置對裝置間通訊下之資源分配與使用者劃分;Pricing Game-Based Resource Allocation and User Grouping for Device-to-Device Communications
Authors:	蘇峻霆;Su,Jyun-ting
Contributors:	通訊工程學系
Keywords:	裝置對裝置間通訊;賽局理論;代價方程式
Date:	2014-08-28
Issue Date:	2014-10-15 17:03:43 (UTC+8)
Publisher:	國立中央大學
Abstract:	為解決通訊系統中上行鏈結(Uplink)的頻帶覆用問題，我們提出裝置對裝置間通訊(Device-to-Device Communications)來改善頻寬擁擠的問題，D2D通訊允許使用者間直接做送收且不經由基地台聯繫，在建構出裝置對裝置間通訊的網路架構之下，我們將面臨到新的干擾問題，如：D2D裝置對基地台的干擾、不同群組間干擾及傳統使用者對D2D通訊的干擾，為了解決新產生的干擾問題，我們引用經濟學中經常使用的賽局理論(Game Theory)將干擾問題模組化，在賽局理論中藉由尋找納許平衡點(Nash Equilibrium)來決定D2D使用者的傳輸功率。　　將決策賽局中的使用者區分為D2D使用者及傳統模式使用者(Cellular Users)，藉由制定賽局中的獲益方程式(Payoff Function)使整個系統可以達到最大整體效益，然而，尚未經過設計獲益方程式中可能不存在或是僅存在非最佳的Nash Equilibrium，因此，將代價方程式(Pricing function)引入至獲益方程式來改善原先的方法。在系統之中代價方程式的選擇上可以依循不同需求，本論文中將以保護傳統模式使用者為出發點與控制D2D users群組間的干擾來制定代價方程式。根據數學分析，提出的獲益方程式是存在至少一個的納許平衡點，因此，我們可以從兩種不同角度出發去劃分共用頻帶的使用者。　　首先，藉由功率控制讓不同模式使用者共享頻帶且彼此的干擾量均可被接受，在這種方式中我們可以找出多組使用者配對，接著比較各組使用者群組間的平衡點來決定哪組為最佳效率。其二，亦可在固定功率的情況之下，依據不同使用者的距離位置來做使用者劃分並選擇存在平衡點的使用者群組，由模擬顯示，在兩種選擇方法之下皆僅存在於唯一的Nash Equilibrium。 ;In order to overcome the problem of frequency reuse in communication systems in uplink, we propose the Device-to-Device communication (D2D Communications) to improve the problem of crowded resource.D2D communications allows user in close proximity to communicate directly without through the base station (BS). After construct the D2D communications network, we face new interference problems, e.g. the interference from D2D user to BS, interference between difference D2D pairs and the interference from cellular user to D2D pair’s receiver. In order to clear up the new interference problems, quoting the game theory, which usual use in economics, to model new interference problems. In game theory, we can find Nash equilibrium (NE) to decide each D2D users transmit power. The users in strategic game distinguish D2D users and cellular users. By establish payoff function in strategic game make overall systems can achieve maximum total payoff. However, undesigned payoff function wills not existent or reach inefficient NE. So the pricing function is necessary to be introduced into the payoff function to ameliorate original methods. A choice of pricing function we can follow difference demands in our D2D network. In this thesis, beginning with protect cellular users and control interference between D2D groups to set up pricing function. According to mathematical analysis, the proposed payoff function exist at least one NE. Therefore, we can start from two distinct ways to grouping users in same resource. First, by means of power control make distinct mode users share same resource and the interference between each user are acceptable. In this way, we can find a lot of user group and compare the Nash equilibrium in each user group to decide which group is the most efficiency. Second, in situation of fixed maximum power constraint. According to the distance of between difference users to grouping user, choosing user group which has NE. By simulation results, both methods have uniqueness of NE.
Appears in Collections:	[Graduate Institute of Communication Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	632	View/Open

社群 sharing

Loading...