Information Technology Reference
In-Depth Information
6 Conclusion and Future Work
The explosive growth of smart phones and rapid deployment of 3G networks
have brought great convenience to people's lives. But there are still a lot of
issues to be concerned. In this paper, we present TraSt, a framework that can
record every detail information for each packet and help testers to do tests and
evaluations on 3G networks and smart phone applications. We also do some
experiments to explore four aspect of Chinese three major operators charging
system and evaluate four kinds of popular smart phone applications. Our study
offers some insight. For the operators' charging system, we could see that in
general, the operators' charging results are quite accurate. The reasons why
they are different from some trac statistic applications' results are: first, some
protocol overheads such as ARP are not charged by operators but counted by
these applications; second, packet loss may lead some packets to be charged
by operators but not received by the smart phone or vice versa; third, some
services such as MMS (Multimedia Messaging Service), will produce trac and
be counted by these applications, but operators will charge it separately. For
the applications of same kind, we could see their trac consumption are always
quite different. The reason is that the degree of optimization and user experience
of different application are not the same.
At present, TraSt use file to store packets, when trac is very large, it can
not meet the real-time requirements. So in the future, we will use more ecient
way such as interprocess communication to get packets.
References
1. Analysis Report on China's Smart Phone Market in March 2012,
2. Android NDK,
http://developer.android.com/sdk/ndk/overview.html
3. Jnetpcap,
http://jnetpcap.com/
4. Ministry of Industry and Information Technology. 3G into the Large-scale Devel-
opment Stage,
http://www.miit.gov.cn
5. Tcpdump,
http://tcpdump.org/
6. RFC 3489: STUN — simple traversal of user datagram protocol (UDP) through
network address translators, NATs (2003)
7. 3GPP TS32.240: Telecommunication management; Charging management; Charg-
ing architecture and principles (2006)
8. G, A.: Global 3G Deployments UMTS HSPA HSPA+ (2010)
9. Huang, J., Xu, Q., Tiwana, B., Mao, Z.M., Zhang, M., Bahl, P.: Anato-
mizing application performance differences on smartphones. In: MobiSys 2010,
pp. 165-178 (2010)
10. Koutsopoulou, M., Kaloxylos, A., Alonistioti, A., Merakos, L.F., Kawamura, K.:
Charging, accounting and billing management schemes in mobile telecommunica-
tion networks and the internet, pp. 50-58 (2004)
11. Chan, M.C., Ramjee, R.: TCP/IP Performance over 3G Wireless Links with Rate
and Delay Variation. In: MobiCom 2002, pp. 71-82 (2002)
12. Mahajan, R., Zhang, M., Poole, L., Pai, V.: Uncovering Performance Differences
among Backbone ISPs with Netdiff. In: NSDI 2008, pp. 205-218 (2008)
