Mobile Network Performance from User Devices: A Longitudinal, Multidimensional Analysis

In the cellular environment, operators, researchers and end users have poor visibility into network performance for devices. Improving visibility is challenging because this performance depends factors that include carrier, access technology, signal strength, geographic location and time. Addressing this requires longitudinal, continuous and large-scale measurements from a diverse set of mobile devices and networks. This paper takes a first look at cellular network performance from this perspective, using 17 months of data collected from devices located throughout the world. We show that (i) there is significant variance in key performance metrics both within and across carriers; (ii) this variance is at best only partially explained by regional and time-of-day patterns; (iii) the stability of network performance varies substantially among carriers. Further, we use the dataset to diagnose the causes behind observed performance problems and identify additional measurements that will improve our ability to reason about mobile network behavior.

[1]  Fabio Ricciato,et al.  Large-scale RTT measurements from an operational UMTS/GPRS network , 2005, First International Conference on Wireless Internet (WICON'05).

[2]  Eng Keong Lua,et al.  Internet Routing Policies and Round-Trip-Times , 2005, PAM.

[3]  Youngseok Lee Measured TCP Performance in CDMA 1x EV-DO Network? , 2006 .

[4]  Mark Claypool,et al.  Characterization by measurement of a CDMA 1x EVDO network , 2006, WICON '06.

[5]  Wing Cheong Lau,et al.  An Empirical Study on 3G Network Capacity and Performance , 2007, IEEE INFOCOM 2007 - 26th IEEE International Conference on Computer Communications.

[6]  Azer Bestavros,et al.  TCP over CDMA2000 Networks: A Cross-Layer Measurement Study , 2007, PAM.

[7]  Jarmo Prokkola,et al.  HSDPA Performance in Live Networks , 2007, 2007 IEEE International Conference on Communications.

[8]  Xin Liu,et al.  Experiences in a 3G network: interplay between the wireless channel and applications , 2008, MobiCom '08.

[9]  Fabio Ricciato,et al.  Network-Wide Measurements of TCP RTT in 3G , 2009, TMA.

[10]  Samir Ranjan Das,et al.  Performance comparison of 3G and metro-scale WiFi for vehicular network access , 2010, IMC '10.

[11]  Markus Fiedler,et al.  Influence of the Packet Size on the One-Way Delay in 3G Networks , 2010, PAM.

[12]  Paramvir Bahl,et al.  Anatomizing application performance differences on smartphones , 2010, MobiSys '10.

[13]  Ramachandran Ramjee,et al.  Bartendr: a practical approach to energy-aware cellular data scheduling , 2010, MobiCom.

[14]  Markus Rupp,et al.  Dissecting 3G Uplink Delay by Measuring in an Operational HSPA Network , 2011, PAM.

[15]  Nina Taft,et al.  Passive and Active Measurement , 2012, Lecture Notes in Computer Science.

[16]  Markus Rupp,et al.  A comparison between one-way delays in operating HSPA and LTE networks , 2012, 2012 10th International Symposium on Modeling and Optimization in Mobile, Ad Hoc and Wireless Networks (WiOpt).

[17]  Ahmed Elmokashfi,et al.  Characterizing Delays in Norwegian 3G Networks , 2012, PAM.

[18]  Paul Barford,et al.  Cell vs. WiFi: on the performance of metro area mobile connections , 2012, Internet Measurement Conference.

[19]  Feng Qian,et al.  An in-depth study of LTE: effect of network protocol and application behavior on performance , 2013, SIGCOMM.

[20]  Ramesh Govindan,et al.  Diagnosing Path Inflation of Mobile Client Traffic , 2014, PAM.