Proactive Metering of Mobile Internet User Experience

Abstract

Having 67% worldwide share, mobile internet is very important for Internet Service Providers (ISPs). Since mobile Internet access is a collective service, Key Performance Indicators (KPIs) measuring quality of data traffic on select network segments/servers may not correctly indicate true user experience. For this reason, mobile ISPs are investing in sophisticated high-end commercial speed analysis systems which typically collect and analyze network traffic data from key network segments/servers. Unfortunately, their utility is quite limited as long as the proactive network intervention is considered. In this work, we develop a MapReduce based network speed analysis system which measures end-to-end network speed to quantify true user experience across multiple geographic regions and service categories. Also functioning as an online decision support system, it enables network administrators with timely ISP network intervention right before potential arrival of mass number of user complaints. The system has been tested with a leading mobile ISP in Turkey. The results confirm its effectiveness.

Keywords

• mobile internet
• internet service providers
• network speed analysis
• big data
• mapreduce

References

1. [1] Cisco Visual Networking Index: Global Mobile Data Traffic Forecast Update, http://www.cisco.com/c/en/us/solutions/collateral/service-provider/visual-networking-index-vni/white_paper_c11-520862.pdf
2. [2] Analysis and modeling of mobile traffic using real traces, Personal, Indoor, and Mobile Radio Communications (PIMRC), 2017 IEEE 28th Annual International Symposium on, Montreal, QC, Canada, ISBN: 978-1-5386-3531-5.
3. [3] Traffic Measurement and Analysis of a Broadband Wireless Internet Access, Vehicular Technology Conference, 2009. VTC Spring 2009. IEEE 69th, Barcelona-Spain, ISBN: 978-1-4244-2517-4.
4. [4] Challenges of future multimedia QoE monitoring for internet service providers, Robitza, W., Ahmad, A., Kara, P.A. et al. Multimed Tools Appl (2017) 76: 22243, https://doi.org/10.1007/s11042-017-4870-z.
5. [5] End-to-end internet speed analysis of mobile networks with mapReduce, Networks, Computers and Communications (ISNCC), 2016 International Symposium on, Yasmine Hammamet, Tunisia, ISBN: 978-1-5090-0284-9.
6. [6] QoS mechanism in content delivery network, Modern Problems of Radio Engineering. Telecommunications and Computer Science (TCSET), 2016 13th International Conference on, Lviv, Ukraine, ISBN: 978-6-1760-7807-4.
7. [7] Comparison of TCP variants in Long Term Evolution (LTE), Electrical, Electronics and Information Engineering (ICEEIE), 2017 5th International Conference on, Malang, Indonesia, ISBN: 978-1-5386-0355-0.
8. [8] An adaptive rate-based congestion control with weighted fairness for large round trip time wireless access networks, Electrical Engineering (ICEE), 2016 24th Iranian Conference on, Shiraz, Iran, ISBN: 978-1-4673-8789-7.

9. [9] Detecting and Localizing End-to-End Performance Degradation for Cellular Data Services Based on TCP Loss Ratio and Round Trip Time, IEEE/ACM Transactions on Networking, Volume: 25, Issue: 6, p: 3709 –3722, Dec. 2017, ISSN: 1558-2566.
10. [10] Data rate fluctuations from user perspective in 4G mobile networks, Software, Telecommunications and Computer Networks (SoftCOM), 2014 22nd International Conference on, ISBN: 978-9-5329-0052-1.
11. [11] Speedtest-Like Measurements in 3G/4G Networks: The MONROE Experience, Teletraffic Congress (ITC 29), 2017 29th International, Genoa, Italy, ISBN: 978-0-9883045-3-6. [12] http://www.speedtest.net/
12. [13] Managing services in the telecom cloud: An example for CDN, Transparent Optical Networks (ICTON), 2016 18th International Conference on, Trento, Italy, ISBN: 978-1-5090-1467-5.
13. [14] Analyzing the potential of mobile opportunistic networks for big data applications, IEEE Network, Volume: 29, Issue: 5, September-October 2015, ISSN: 0890-8044.
14. [15] Jeffrey Dean and Sanjay Ghemawat. MapReduce: Simplified data processing on large clusters. In Proceedings of the 6th Symposium on Operating System Design and Implementation (OSDI 2004), pages 137-150, San Francisco, California, 2004.
15. [16] Dean, J., & Ghemawat, S. (2010). MapReduce: a flexible data processing tool.Communications of the ACM, 53(1), 72-77. [17] Hadoop, http://hadoop.apache.org/.
16. [18] Zikopoulos, P., & Eaton, C. (2011). Understanding big data: Analytics for enterprise class hadoop and streaming data. McGraw-Hill Osborne Media.
17. [19] Characterizing Flow, Application, and User Behavior in Mobile Networks: A Framework for Mobile Big Data, IEEE Wireless Communications, Volume: 25, Issue: 1, February 2018, ISSN: 1536-1284.
18. [20] Traffic flow management in a real mobile phone network using linear optimization, IEEE Latin America Transactions, Volume: 16, Issue: 2, Feb. 2018, ISSN: 1548-0992. [21] Dumpcap, https://www.wireshark.org/docs/man-pages/dumpcap.html
19. [22] Blum, R. (2003). Network Performance Open Source Toolkit: Using Netperf, tcptrace, NISTnet, and SSFNet. John Wiley & Sons. [23] Tcptrace, http://www.tcptrace.org/
20. [24] Measurement and Classification of Smart Systems Data Traffic Over 5G Mobile Networks, Dighriri M., Lee G.M., Baker T. (2018), In: Dastbaz M., Arabnia H., Akhgar B. (eds) Technology for Smart Futures. Springer, Cham, ISBN: 978-3-319-60137-3.
21. [25] Online Internet Traffic Measurement and Monitoring Using Spark Streaming, GLOBECOM 2017 - 2017 IEEE Global Communications Conference, Singapore, Singapore, ISBN: 978-1-5090-5019-2.
22. [26] New Model and Open Tools for Real Testing of QoE in Mobile Broadband Services and the Transport Protocol Impact: The Operator's Approach, IEEE Latin America Transactions, Volume: 13, Issue: 2, Feb. 2015, ISSN: 1548-0992.