Characterizing and Improving WiFi Latency in Large-Scale Operational Networks

WiFi latency is a key factor impacting the user experience of modern mobile applications, but it has not been well studied at large scale. In this paper, we design and deploy WiFiSeer, a framework to measure and characterize WiFi latency at large scale. WiFiSeer comprises a systematic methodology for modeling the complex relationships between WiFi latency and a diverse set of WiFi performance metrics, device characteristics, and environmental factors. WiFiSeer was deployed on Tsinghua campus to conduct a WiFi latency measurement study of unprecedented scale with more than 47,000 unique user devices. We observe that WiFi latency follows a long tail distribution and the 90th (99th) percentile is around 20 ms (250 ms). Furthermore, our measurement results quantitatively confirm some anecdotal perceptions about impacting factors and disapprove others. We deploy three practical solutions for improving WiFi latency in Tsinghua, and the results show significantly improved WiFi latencies. In particular, over 1,000 devices use our AP selection service based on a predictive WiFi latency model for 2.5 months, and 72% of their latencies are reduced by over half after they re-associate to the suggested APs.

[1]  Haibo He,et al.  Learning from Imbalanced Data , 2009, IEEE Transactions on Knowledge and Data Engineering.

[2]  Mike Y. Chen,et al.  Improved access point selection , 2006, MobiSys '06.

[3]  Praphul Chandra,et al.  Wi-Fi Telephony: Challenges and Solutions for Voice over WLANs (Communications Engineering Series) , 2006 .

[4]  Srinivasan Seshan,et al.  Developing a predictive model of quality of experience for internet video , 2013, SIGCOMM.

[5]  Kajal T. Claypool,et al.  Latency can kill: precision and deadline in online games , 2010, MMSys '10.

[6]  Vyas Sekar,et al.  Shedding light on the structure of internet video quality problems in the wild , 2013, CoNEXT.

[7]  Feng Qian,et al.  Periodic transfers in mobile applications: network-wide origin, impact, and optimization , 2012, WWW.

[8]  Chris Chafe,et al.  Effect of Time Delay on Ensemble Accuracy , 2004 .

[9]  Ning Ding,et al.  Characterizing and modeling the impact of wireless signal strength on smartphone battery drain , 2013, SIGMETRICS '13.

[10]  Yunhao Liu,et al.  Voice over the dins: Improving wireless channel utilization with collision tolerance , 2013, 2013 21st IEEE International Conference on Network Protocols (ICNP).

[11]  Lusheng Ji,et al.  Understanding the impact of network dynamics on mobile video user engagement , 2014, SIGMETRICS '14.

[12]  Stefan Savage,et al.  Jigsaw: solving the puzzle of enterprise 802.11 analysis , 2006, SIGCOMM.

[13]  Srinivasan Seshan,et al.  Modeling web quality-of-experience on cellular networks , 2014, MobiCom.

[14]  Costin Raiciu,et al.  Towards Wifi Mobility without Fast Handover , 2015, NSDI.

[15]  Ratul Mahajan,et al.  Analyzing the MAC-level behavior of wireless networks in the wild , 2006, SIGCOMM 2006.

[16]  Rebecca E. Grinter,et al.  Measuring and Mitigating Web Performance Bottlenecks in Broadband Access Networks , 2014 .

[17]  Stefan Savage,et al.  The Shaman Automatic 802.11 Wireless Diagnosis System , 2010 .

[18]  Suman Banerjee,et al.  Observing home wireless experience through WiFi APs , 2013, MobiCom.

[19]  Nick Feamster,et al.  Web performance bottlenecks in broadband access networks , 2013, SIGMETRICS '13.

[20]  Nick Feamster,et al.  Community contribution award -- Measuring and mitigating web performance bottlenecks in broadband access networks , 2013, Internet Measurement Conference.

[21]  Edmund Wong,et al.  Large-scale Measurements of Wireless Network Behavior , 2015, SIGCOMM.

[22]  Peter Steenkiste,et al.  Fixing 802.11 access point selection , 2002, CCRV.

[23]  Dan Pei,et al.  How bad are the rogues' impact on enterprise 802.11 network performance? , 2015, 2015 IEEE Conference on Computer Communications (INFOCOM).

[24]  Stefan Savage,et al.  Jigsaw: solving the puzzle of enterprise 802.11 analysis , 2006, SIGCOMM.

[25]  Ratul Mahajan,et al.  Analyzing the MAC-level behavior of wireless networks in the wild , 2006, SIGCOMM.

[26]  David Wetherall,et al.  How Speedy is SPDY? , 2014, NSDI.

[27]  Paramvir Bahl,et al.  MultiNet: connecting to multiple IEEE 802.11 networks using a single wireless card , 2004, IEEE INFOCOM 2004.

[28]  Vishnu Navda,et al.  DiversiFi: robust multi-link interactive streaming , 2015, CoNEXT.

[29]  Marco Conti,et al.  Design and performance evaluation of an asymptotically optimal backoff algorithm for IEEE 802.11 Wireless LANs , 2000, Proceedings of the 33rd Annual Hawaii International Conference on System Sciences.

[30]  Dimitrios Koutsonikolas,et al.  Realizing the full potential of PSM using proxying , 2012, 2012 Proceedings IEEE INFOCOM.

[31]  Gang Zhou,et al.  All or none? The dilemma of handling WiFi broadcast traffic in smartphone suspend mode , 2015, 2015 IEEE Conference on Computer Communications (INFOCOM).

[32]  Konstantina Papagiannaki,et al.  PIE in the Sky: Online Passive Interference Estimation for Enterprise WLANs , 2011, NSDI.

[33]  Jie Wu,et al.  Designing a Practical Access Point Association Protocol , 2010, 2010 Proceedings IEEE INFOCOM.

[34]  James Parker,et al.  on Knowledge and Data Engineering, , 1990 .

[35]  Samir Ranjan Das,et al.  Passive Measurement of Interference in WiFi Networks with Application in Misbehavior Detection , 2013, IEEE Transactions on Mobile Computing.