Mobile Networks and Applications

, Volume 21, Issue 3, pp 402–413 | Cite as

Understanding and Predicting Data Hotspots in Cellular Networks

  • Ana Nika
  • Asad Ismail
  • Ben Y. Zhao
  • Sabrina Gaito
  • Gian Paolo Rossi
  • Haitao Zheng


The unprecedented growth in mobile data usage is posing significant challenges to cellular operators. One key challenge is how to provide quality of service to subscribers when their residing cell is experiencing a significant amount of traffic, i.e. becoming a traffic hotspot. In this paper, we perform an empirical study on data hotspots in today’s cellular networks using a 9-week cellular dataset with 734K+ users and 5327 cell sites. Our analysis examines in details static and dynamic characteristics, predictability, and causes of data hotspots, and their correlation with call hotspots. We show that using standard machine learning methods, future hotspots can be accurately predicted from past observations. We believe the understanding of these key issues will lead to more efficient and responsive resource management and thus better QoS provision in cellular networks. To the best of our knowledge, our work is the first to empirically characterize traffic hotspots in today’s cellular networks.


Data hotspots Cellular networks Machine learning 


  1. 1.
  2. 2.
    Nan E, Chu X, Guo W, Zhang J (2013) User data traffic analysis for 3G cellular networks. In: Proc. of CHINACOM, pp 468– 472Google Scholar
  3. 3.
    Paul U, Subramanian A, Buddhikot M, Das S (2012) Understanding spatial relationships in resource usage in cellular data networks. In: Proc. of INFOCOM workshops, pp 244–249Google Scholar
  4. 4.
    Laner M, Svoboda P, Schwarz S, Rupp M (2012) Users in cells: a data traffic analysis. In: Proc. of WCNC, pp 3063–3068Google Scholar
  5. 5.
    Paul U, Subramanian A, Buddhikot M, Das S (2011) Understanding traffic dynamics in cellular data networks. In: Proc. of INFOCOM, pp 882–890Google Scholar
  6. 6.
    Shafiq M Z, Ji L, Liu A Z, Pang J, Wang J (2012) Characterizing geospatial dynamics of application usage in a 3G cellular data network. In: Proc. of INFOCOM, pp 1341–1349Google Scholar
  7. 7.
    Shafiq M Z, Ji L, Liu A X, Wang J (2011) Characterizing and modeling Internet traffic dynamics of cellular devices. In: Proc. of SIGMETRICS, pp 305–316Google Scholar
  8. 8.
    Zhou X, Zhao Z, Li R, Zhou Y, Zhang H (2012) The predictability of cellular networks traffic. In: Proc. of ISCIT, pp 973–978Google Scholar
  9. 9.
  10. 10.
    Sinnott R W (1984) Virtues of the haversine. Sky Telescop 68(2):159MathSciNetGoogle Scholar
  11. 11.
    Defays D (1977) An efficient algorithm for a complete link method. Comput J (Br Comput Soc) 20(4):364–366MathSciNetMATHGoogle Scholar
  12. 12.
    Isaacman S, Becker R, Cáceres R, Kobourov S, Martonosi M, Rowland J, Varshavsky A (2011) Identifying important places in people’s lives from cellular network data. In: Proc. of 9th international conference on pervasive computing, pp 133– 151Google Scholar
  13. 13.
    Hall M, Frank E, Holmes G, Pfahringer B, Reutemann R, Witten I H (2009) The WEKA data mining software: an update. SIGKDD Explor Newsl 11(1):10–18CrossRefGoogle Scholar
  14. 14.
    Cortes C, Vapnik V (1995) Support-vector networks. Mach Learn 20(3):273–297MATHGoogle Scholar
  15. 15.
    Breiman L (2001) Random forests. Mach Learn 45(1):5– 32MathSciNetCrossRefMATHGoogle Scholar
  16. 16.
    Guyon I, Elisseeff A (2003) An introduction to variable and feature selection. J Mach Learn Res 3:1157–1182MATHGoogle Scholar
  17. 17.
    Mohri M, Rostamizadeh A, Talwalkar A (2012) Foundations of machine learning. The MIT PressGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2015

Authors and Affiliations

  1. 1.Department of Computer ScienceUniversity of CaliforniaSanta BarbaraUSA
  2. 2.Università degli Studi di MilanoMilanItaly

Personalised recommendations