完整後設資料紀錄
DC 欄位語言
dc.contributor.author陳柏言zh_TW
dc.contributor.author林甫俊zh_TW
dc.contributor.authorChen, Bo-Yanen_US
dc.date.accessioned2018-01-24T07:37:00Z-
dc.date.available2018-01-24T07:37:00Z-
dc.date.issued2016en_US
dc.identifier.urihttp://etd.lib.nctu.edu.tw/cdrfb3/record/nctu/#GT070356531en_US
dc.identifier.urihttp://hdl.handle.net/11536/138869-
dc.description.abstract在現今的3G、4G LTE核心網路下,大眾所使用的網路服務皆是根據通話時間、資料傳輸量與簡訊傳輸量來收取費用,而在未來物聯網環境下,將會出現許多智慧服務,而這些服務裝置將不再使用通話與簡訊服務,並且他們當中很多是非常小的資料量傳輸,卻進行大量的連線、儲存與通知。如果未來物聯網還是以現今的3G、4G LTE的收費方式來收取費用的話,物聯網服務商或電信業者將無法從服務訂閱者收取到公平的收益。為了達成物聯網服務提供商與服務訂閱者之間的雙贏局面,發展出一套適用於物聯網服務的新收費模型與收費架構是非常重要的。 本論文會根據先前研究已經定義出的十種物聯網收費因子,如儲存量、資料傳輸、連線次數、擁塞情況、服務品質、優先度、訂閱、安全、漫遊、群組,及根據上述物聯網收費因子發展出的一套有彈性的物聯網收費模型,更進一步地探討物聯網網路架構下要如何正確地收集每個收費因子與收費模型所需的資料,我們稱此為物聯網收費架構。 雖然在ETSI M2M標準中有明確的指出儲存量、資料傳輸、連線次數可作為物聯網的收費因子,但卻沒有討論要如何去收集上述收費因子所需的資訊與應用,更遑論去探討其他可能的收費因子。而在此研究,我們將不討論安全、擁塞、漫遊、優先度這四個動態收費因子,而是集中探討其餘的六個靜態收費因子的應用。根據M2M標準,每個收費因子都有不同的屬性,而這些屬性決定每個收費因子所需的收費資訊,通常這些資訊可以從網路層或服務層中收集,前者為物聯網平台所使用的基礎網路,後者為物聯網服務平台本身。 物聯網網路層與服務層之間的資料收集將分別討論如下: 1. 網路層收費因子的屬性資訊: 資料傳輸與服務品質兩個收費因子的屬性資訊將在網路層中被收集。根據3GPP標準,當服務訂閱者產生網路資料流時,PGW(PDN-Gateway)可以偵測此資料流並量測每個資料流的使用量與使用的服務品質。在此研究中,將使用封包嗅探器模擬PGW(PDN-Gateway)來收集資料傳輸與服務品質兩個收費因子的屬性資訊。 2. 服務層收費因子的屬性資訊: 儲存量、連線次數、訂閱、分群皆為服務層的收費因子,根據ETSI M2M標準,來自服務訂閱者的HTTP Request送至物聯網平台皆會轉變成RequestIndication primitives。因此服務層中可以藉由分析RequestIndication primitives的內容來收集四個收費因子的屬性資訊。 本論文設計並實作一物聯網收費架構,此設計將彙整網路層與服務層的收費因子資訊,並且根據收費因子資訊來產生收費資料紀錄(CDRs)。而收費資料紀錄最後將被傳送到網路電信業者的帳務伺服器來產生服務訂閱者的帳單。zh_TW
dc.description.abstractIn the traditional networks such as 3G and 4G LTE, the services are charged based on measuring the duration of voice call, the amount of data transfer and the number of SMS messages. However, in IoT/M2M there will be many smart services and many “things” exchange only a small amount of data once in a while, use no voice or SMS services at all, while generate a large amount of storage requests, notification events and network connectivity requests. With this usage pattern of the network, service providers and network operators won’t be able to collect a fair amount of revenue from the subscribers if they still utilize charging models of the past. It is thus important to develop new charging models and related charging architectures for M2M communications in order to meet the unique characteristics of these new communications and ensure a win-win between M2M service providers and subscribers. Ten charging factors for M2M communications have identified in the previous research, including Storage, Data Transfer, Connectivity, Congestion, QoS, Priority, Subscription, Security, Mobility and Grouping. Also, flexible charging models for M2M communications have been developed based on the charging factors identified. In this research, we propose a new charging architecture in M2M communications based on the previously defined charging factors and charging models. Our charging architecture is defined to clarify how to collect all needed data for new charging factors and charging models. Though ETSI identified Storage, Data Transfer and Connectivity as M2M charging factors, it didn’t discuss how to collect data for these charging factors. Neither did they discuss the charging factors beyond the three they defined. In this research, we will not consider Security, Congestion, Mobility and Priority dynamic charging factors but only address how to collect data for the remaining six static charging factors including the three proposed by ETSI. There are different attributes for each charging factor. These attributes for the six charging factors can be collected from either the network layer or the service layer. The latter refers to the M2M service platform while the former refers to the underlying network of the M2M platform. The collection of data from the network layer and the service layer of M2M communications will be discussed separately in the following. 1. Attributes of charging factors in the network layer: There are two charging factors Data Transfer, and QoS whose attributes need to be collected from the network layer. In the 3GPP standard, when a network service data flow is activated by a UE, the PGW (PDN-Gateway) can detect the flow and measure its bearer usage including data transfer amount and QoS. In our research, the PGW will be simulated by a packet sniffer where both data transfer and QoS information can be collected. 2. Attributes of charging factors in the service layer: The charging factors in the services layer include Storage, Connectivity, Subscription and Grouping. According to the ETSI M2M standard, the HTTP requests from an application to the service layer would always be translated to RequestIndication primitives. Thus, the attributes for Storage, Subscription, Connectivity and Grouping can be collected from analyzing the contents of RequestIndication primitives In this thesis, we design an M2M charging architecture that collect and consolidate charging factors information from both network and service layers. The goal of the design is to construct Charging Data Records (CDRs) based on the consolidation of charging factors information. CDRs will then be transferred to the network operator's billing server for the purpose of generating subscriber’s bills.en_US
dc.language.isoen_USen_US
dc.subjectM2M communicationszh_TW
dc.subjectM2M charging factorszh_TW
dc.subjectM2M charging modelszh_TW
dc.subjectM2M charging architecturezh_TW
dc.subjectETSI M2Mzh_TW
dc.subjectOpenMTCzh_TW
dc.subjectM2M communicationsen_US
dc.subjectM2M charging factorsen_US
dc.subjectM2M charging modelsen_US
dc.subjectM2M charging architectureen_US
dc.subjectETSI M2Men_US
dc.subjectOpenMTCen_US
dc.title物聯網收費架構研究zh_TW
dc.titleCharging Architecture in M2M Communicationsen_US
dc.typeThesisen_US
dc.contributor.department網路工程研究所zh_TW
顯示於類別:畢業論文