Understanding Traffic Dynamics in Cellular Data Networks and Applications to Resource Management

Overview

This project seeks to undertake a significant modeling exercise on wireless/mobile network data with two goals. One goal is ‘intellectual,’ driven towards understanding the traffic dynamics and discovering any possible structure or relationships. Such understanding will bring new insights that in turn will help to deploy and manage future generation mobile data networks. The second goal is ‘utilitarian.’ Here, we want to use the understanding for doing some form of resource management decisions. We specifically target cellular data networks.

 

We are working on two major research tasks:

 

·      Develop an understanding of the nature of the spatio­temporal traffic dynamics in cellular data networks by developing models from large­scale measurement data collected directly from operational networks, and using such models to effectively forecast load. This modeling study uses a broad range of state­of­the­art tools from the statistical machine learning community.

 

·      Using the forecasted load, design algorithms to exploit various resource management (e.g., energy or capacity provisioning) opportunities in the context of cellular networks. In particular, we are developing algorithms to select base stations to turn­off and choose assignments of demands to base stations, such that the total energy cost is minimized, or deprioritizing non-essential traffic to reduce peak-to-average ratio of the traffic load.  

 

In addition, we are pursuing research in data analytics driven traffic management, use of machine learning in wireless localization and data­driven performance analysis of mobile virtual networks. We are also studying benefits of network functions virtualization in cellular networks and pricing issues.

 

People

• Samir R. Das (PI)
• Luis Ortiz (Co-PI)
• Himanshu Gupta (Co-PI)
• Milind Buddhikot (PI @ Bell Labs)
• Utpal Paul (PhD Student, graduated in 2012)
• Giordano Fusco (PhD Student, graduated in 2013)
• Navid Azimi (PhD Student, graduated in 2014)
• Ayon Chakraborty (PhD student)
• Zafar Qazi (PhD student)

Highlights

Following highlights are based on hourly basestation loads (airtime) on a nationwide CDMA-based 3G network containing roughly 10,000 BSs and 1 million subscribers using 1 week long data.

The above shows the optimal spectral clustering of macro-cells in a subregion based on pairwise cross-correlation, demonstrating large degrees of correlated behavior.


The above shows ‘Granger causal flows’  in a smaller subregion (downtown area) – green arrows for unidirectional and red lines for bidirectional causality between neighboring cells, with sequences of arrows and lines showing causal flows. This demonstrates a significant causal structure.




The above shows diurnal variations of the hourly load in the top two BSs in the seven day period, showing mid-day peaks and lesser activity in the weekends (the last 2 days).


Relevant Publications

• Fatima Zarinni, Ayon Chakraborty, Vyas Sekar, Samir Das, Phillipa Gill, “A First Look at Performance in Mobile Virtual Network Operators,” Proc.  ACM Internet Measurement Conference (IMC), 2014. To appear.
• Zafar A. Qazi, Vyas Sekar, Samir R. Das, “A Framework to Quantify the Benefits of Network Functions Virtualization in Cellular Networks,” arXiv Technical Report No. arXiv:1406.5634, 2014.
• Utpal Paul, Luis Ortiz, Samir.R. Das, Giordano Fusco, Milind M. Buddhikot, “Learning probabilistic models of cellular network traffic with applications to resource management,” Proc. IEEE International Symposium on Dynamic Spectrum Access Networks (DYSPAN), 2014.
• Navid Hamed Azimi, Himanshu Gupta, Utpal Paul, Milind Madhav Buddhikot, Samir R Das, Minimizing Capacity Requirements of Cellular Networks via Delayed Scheduling. Proc. IEEE International Conference on Sensing, Communications and Networking (SECON), 2013.
• Giordano Fusco, Navid Hamed Azimi, Himanshu Gupta, “Capacity Optimization of Femtocell Networks,” Proc. IEEE International Conference on Sensing, Communications and Networking (SECON), 2013.
• Utpal Paul, Milind Buddhikot, Samir R. Das, "Opportunistic Traffic Scheduling in Cellular Data Networks,'' Proc. IEEE Symp. on Dynamic Spectrum Access Networks (DySPAN), Bellevue, WA, October 2012. 
• Utpal Paul, Aanand Prabhu Subramanian, Milind Buddhikot, Samir R. Das, "Understanding Spatial Relationships in Resource Usage in Cellular Data Networks,'' Proc. IEEE NetSciCom 2012 Workshop: 4th International Workshop on Network Science for Communication Networks, Orlando, March 2012.  
• Giordano Fusco, Milind Buddhikot, Himanshu Gupta, and S. Venkatesan, "Finding Green Spots and Turning the Spectrum Novel: Dial Techniques for Green Mobile Wireless Networks," in Proc. IEEE Symposium on Dynamic Spectrum Access Networks (DySPAN), May 2011.
• Utpal Paul, Anand Prabhu Subramanian, Milind M. Buddhikot, Samir R. Das, "Understanding Traffic Dynamics in Cellular Data Networks", Proc. IEEE INFOCOM, Shanghai, China, April 2011.

Sponsor

National Science Foundation


Collaborator

Alcatel-Lucent Bell Labs