Discovering Periodic Patterns for Large Scale Mobile Traffic Data: Method and Applications

Understanding the temporal traffic load profile of cellular networks is extremely valuable to many network operation tasks in large mobile networks. Such knowledge is useful for network planning, improving network performance, designing better load balancing schemes, testing handoff algorithms, and proposing new charging mechanisms. This paper proposes a simple yet powerful method to model the temporal traffic profile by a large 3G/LTE cellular network dataset in a metropolitan area, consisting of 9 thousand base stations and 3.5 million subscribers. Specifically, using the spectrum-based analysis, we extract three major frequency components, which captures the weekly, daily, and hourly temporal patterns in the traffic load across base stations. By clustering the traffic utilizing the features extracted from spectrum-domain components, we find that urban scale cellular traffic can be classified into five groups, which maps to five types of geographic locations. Besides the comprehensive analysis, we apply this model to two applications: predicting the future traffic load, and designing a load based pricing scheme, where we demonstrate the usefulness of our model and analysis results.

[1]  Halim Yanikomeroglu,et al.  HetHetNets: Heterogeneous Traffic Distribution in Heterogeneous Wireless Cellular Networks , 2015, IEEE Journal on Selected Areas in Communications.

[2]  Chris H. Q. Ding,et al.  K-means clustering via principal component analysis , 2004, ICML.

[3]  Donald W. Bouldin,et al.  A Cluster Separation Measure , 1979, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  Ying Zhang,et al.  Understanding the characteristics of cellular data traffic , 2012, CellNet '12.

[5]  Carlo Ratti,et al.  Visualizing signatures of human activity in cities across the globe , 2015, ArXiv.

[6]  Ying Zhang.  User mobility from the view of cellular data networks , 2014, IEEE INFOCOM 2014 - IEEE Conference on Computer Communications.

[7]  Shobha Venkataraman,et al.  Characterizing data usage patterns in a large cellular network , 2012, CellNet '12.

[8]  Muhammad Ali Imran,et al.  How much energy is needed to run a wireless network? , 2011, IEEE Wireless Communications.

[9]  Qiang Xu,et al.  Identifying diverse usage behaviors of smartphone apps , 2011, IMC '11.

[10]  Dan Wang,et al.  Time dependent pricing in wireless data networks: Flat-rate vs. usage-based schemes , 2014, IEEE INFOCOM 2014 - IEEE Conference on Computer Communications.

[11]  Zhifeng Zhao,et al.  The predictability of cellular networks traffic , 2012, 2012 International Symposium on Communications and Information Technologies (ISCIT).

[12]  Di Yuan,et al.  Analysis of Cell Load Coupling for LTE Network Planning and Optimization , 2012, IEEE Transactions on Wireless Communications.

[13]  Guoqiang Peter Zhang,et al.  Time series forecasting using a hybrid ARIMA and neural network model , 2003, Neurocomputing.

[14]  David Lazer,et al.  Inferring friendship network structure by using mobile phone data , 2009, Proceedings of the National Academy of Sciences.

[15]  K. K. Ramakrishnan,et al.  Over the top video: the gorilla in cellular networks , 2011, IMC '11.

[16]  Athina Markopoulou,et al.  Quantifying the potential of ride-sharing using call description records , 2013, HotMobile '13.

[17]  Sangtae Ha,et al.  TUBE: time-dependent pricing for mobile data , 2012, SIGCOMM '12.

[18]  Honggang Zhang,et al.  Spatial modeling of the traffic density in cellular networks , 2014, IEEE Wireless Communications.

[19]  Yu-Chee Tseng,et al.  Time-Dependent Smart Data Pricing Based on Machine Learning , 2017, Canadian Conference on AI.

[20]  Dan Wang,et al.  The effectiveness of time dependent pricing in controlling usage incentives in wireless data network , 2013, SIGCOMM.

[21]  Jonathan Rodriguez,et al.  Frequency Allocation for HetNet CoMP: Energy Efficiency Analysis , 2013, ISWCS.

[22]  Lusheng Ji,et al.  Characterizing and modeling internet traffic dynamics of cellular devices , 2011, SIGMETRICS '11.

[23]  Sergei Vassilvitskii,et al.  k-means++: the advantages of careful seeding , 2007, SODA '07.

[24]  Sangtae Ha,et al.  Incentivizing time-shifting of data: a survey of time-dependent pricing for internet access , 2012, IEEE Communications Magazine.

[25]  Ramón Cáceres,et al.  A Tale of One City: Using Cellular Network Data for Urban Planning , 2011, IEEE Pervasive Computing.

[26]  Nitesh V. Chawla,et al.  Inferring user demographics and social strategies in mobile social networks , 2014, KDD.

[27]  Sangtae Ha,et al.  Time-Dependent Broadband Pricing: Feasibility and Benefits , 2011, 2011 31st International Conference on Distributed Computing Systems.

[28]  Jyri Hämäläinen,et al.  Topology and irregularity in cellular networks , 2015, 2015 IEEE Wireless Communications and Networking Conference (WCNC).

[29]  Josip Lorincz,et al.  A measurement study of short-time cell outages in mobile cellular networks , 2016, Comput. Commun..

[30]  Jyri Hamalainen,et al.  Looking at Cellular Networks Through Canonical Domains and Conformal Mapping , 2015 .

[31]  Minas Gjoka,et al.  On the Decomposition of Cell Phone Activity Patterns and their Connection with Urban Ecology , 2015, MobiHoc.

[32]  J. A. Hartigan,et al.  A k-means clustering algorithm , 1979 .

[33]  Magnus Olsson,et al.  Energy performance evaluation revisited: Methodology, models and results , 2016, IEEE International Conference on Wireless and Mobile Computing, Networking and Communications.

[34]  Albert-László Barabási,et al.  Understanding individual human mobility patterns , 2008, Nature.

[35]  Qin Wang,et al.  Mobile bandwidth sharing: Pricing analysis under buffet plans , 2016, 2016 18th Asia-Pacific Network Operations and Management Symposium (APNOMS).

[36]  T. Graepel,et al.  Private traits and attributes are predictable from digital records of human behavior , 2013, Proceedings of the National Academy of Sciences.