近年來網路與應用程式多元發展，使用者可藉由終端設備，於不同場合選擇相映應的存取裝置來上傳或下載資訊。但不同的應用程式，如結合數據之網際網路資料存取、語音及影像的三合一整合服務(Triple Play)，其對傳輸網路品質之需求，會隨著媒體資訊之不同而有不同服務品質(Quality of Service, QOS)的要求。固定式網路(Fixed Network)是基於實體電路所提供的資料傳遞(Modem, ADSL/xDSL, FTTB/FTTH)，現今無線產業開放所形成的多元選擇(CSD, GPRS, HSDPA 或未來4G/LTE)，或者是行動固網整合應用服務FMC(Fixed Mobile Convergence)所衍生之新型網路型態，當Triple Play再加上Mobility所造就的Quadruple play，在功能手機(Feature Phone)與智慧手機(Smart Phone)普及的同時，是否能對不同之Media提供相對應的QOS，或提供更穩定的網路傳輸品質，實是值得探討。本論文將透過IP網路量測，將不同類型的偵測封包，如TCP，UDP和ICMP，以模擬應用程式所使用的通訊協定來估測delay, jitter; packet loss 與throughput的特徵量測。我們依據QOS規格要求，量化各型基本網路特性之數據，可以了解不同應用於特定網路運作性能。對於FMC網路，如整合固定式與3G網路的Femtocell，依據此法更可以取得明確之性能數據(performance data)，讓一般使用者對不同應用所需之QOS要求，從多元網路選擇最佳效能的系統參考，對於網路維運商或者新設網路系統所承載的應用服務，能進行事前長期量測並進行網路優化(Network optimization)，或藉此除錯(Troubleshooting)，例如網路負載增加對既有系統運用之影響，可運用此法以使用者角度來進行的端點對端點(end to end)的量測方式，以適時擴充系統。面對未來LTE將全面IP化的來臨，對於相關應用與系統訊息傳遞的基本品質要求將有極大的助益。

Multimedia applications have to be delivered through heterogeneous networks. Users can choose different access terminal devices to upload or download information by their request. However, for the “Triple Play” which combines Internet data access, voice and video services together, the Quality of Service (QoS) requirement would be different for different media communication applications. For fixed network, information is provided based on circuit entity (Modem, ADSL/xDSL, FTTB/FTTH), while in wireless network, several access technologies to acquire information are available, such as CSD, GPRS, HSDPA and 4G/LTE. The wireless network enables mobile data access and upgrades the Triple Play to Quadruple Play, which allows users to access and enjoy rich information media content at anytime and anywhere. With faster data access rate and higher mobility wireless network access, together with the variety of new developed Smart Phones, how to provide appropriate QoS or network transmission resource quality for different types of information media becomes important. We proposed to measures network metrics through the IP network for different types of data packets to simulate the applications to yield the network transmission metrics, such as delay, jitter, packet loss and throughput estimated in TCP, UDP, and ICMP protocols. With different QoS for different application requirements, the measured network metrics can be classified into different specified QoS categories. Under this setup, we can understand the network performance for different applications in FMC network (integration of fixed network and 3G Femtocell). With the help network metrics measurements, we can predict QoS performance for an application that is carried out through specified network metrics. The users can select the most suitable data access channel from the multi-network system with reference to QoS requirements for different applications. In addition, for network operators or newly established network system to serve for specified applications, the proposed network metric measurement method can help to maintain constant network condition monitoring, optimizing and troubleshooting to improve the performance of existing network facilities and timely expand the network service capability from user terminals through end-to-end measurement. In the future, for the longer term evolution (LTE) and all-IP networks, the proposed methodology may help much in exploiting the relation between application requirement and system signaling messages.

[1] TWNIC,2011/07, “2011年臺灣無線網路使用調查報告,”
From http://www.twnic.net.tw/download/200307/1107d.pdf
[2] Ericsson, 2011年11月, “市場數據報告-掌握網路型社會脈動,”
From http://www.ericsson.com/res/site_TW/docs/article_2012_vol01_02.pdf pp.3-8
[3] DIFFSERV—The scalable end -to- end quality of service model, from
http://www.cisco.com/en/US/technologies/tk543/tk766/technologies_white_paper09186a00800a3e2f.pdf
[4] Quality of Service (QoS) concept and architecture (Release 9),
TS 23.107,3GPP,pp.15-17,2011-12.
[5] Ivica Kostanic, Nenad Mijatovic and Stephen D. Vest, “Measurement Based
QoS Comparison of Cellular Communication Networks,”IEEE,pp.3, 2009.
[6] 林 秉 暉 ,” UMTS網路下使用MIH服務以預先完成IMS之QoS研究,”
通訊工程學系碩士論文,國立中央大學,2008.
[7] Mobile radio interface Layer 3 specification;Core network protocols; Stage
3(Release 10) TS 24.008 V10.7.0 , 3GPP, pp.544-552, 2012-06.
[8] IP Multimedia Subsystem (IMS);Stage 2(Release 11) TS 23.228,3GPP,
pp.23-25 ;43-46 ,2012-06.
[9] SDP: Session Description Protocol RFC 4566, pp.1-30,2006.
[10] Alberto Diez Albaladejo., Fabricio Carvalho de Gouveia.,Marius Iulian Corici
Thomas Magedanz., “The PCC Rule in the 3GPP IMS Policy and Charging
Control Architecture ,” IEEE GLOBECOM,2008.
[11] Cisco IOS IP SLAs Configuration Guide, Release 12.4,from
http://www.cisco.com/en/US/docs/ios/12_4/ip_sla/configuration/guide/hsla_c.html
[12] Nadia Ghiata and Marius Marcu,”Measurement methods for QoS in VoIP
review,”Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT), pp2-3,2011.
[13] Shue, C.A.,Gupta, M.,Myers, S.A.,” IPSec: Performance Analysis and
Enhancement ,” IEEE International Conference ,pp1-3,2007.
[14] Services and service capabilities(Release9) TS 22.105,3GPP,pp.8-16, 2010-09 .
[15] Policy and charging control architecture (Release 8) TS 23.203 V8.14.0 ,3GPP,
pp.1-32, 2012-06.