Spatio-temporal models of cellular traffic are crucial for a number of wireless network applications, such as protocol design and tuning for real networks, malware detection, network planning and contingency analysis and spectrum market design.

Overview


We have built an integrated multi-network modeling environment for spectrum management. It uses a first-principles based approach that involves the integration of a diverse set of public and commercial datasets and computational models into a common architecture for data exchange. The system consists of following modules:

Mobility Model: An "activity-based mobility" model uses realistic human mobility patterns caused by daily activities during a normative day. The mobility patterns are constructed using TRANSIMS, a synthetic urban traffic generator for a synthetic population. The regions are divided into voronoi cells, which represent the coverage area by a cell tower. Mobility model tracks the sequence of cells traversed by each person.

Wireless Device Ownership: A classification and regression tree appoach is applied to National Health Interview Survey (NHIS) data from CDC to build a model that determines the likelihood of owning a wireless device based on demographic labels.

Modeling Calls and Usage Patterns: This module uses a discrete event simulation approach to model calling patterns, using well known distributions for call duration and arrival rates. The caller is randomly chosen from the synthetic population but the callee can be chosen (1) from the social network of the caller, (2) age based homophily criteria and (3) randomly.

Application Areas: This modeling environment enable us to study a wide range of applications areas such as spread of information in mobile social networks, viral marketing, spread of malware, interdependent infrastructures and contingency planning, network and protocol design in cognitive networks, spectrum markets, etc. Examples of these areas are shown below.

 

Applications


  • Hotspots can arise because of traffic jams, failure of cell towers due to power outages, and malware and denial of service (DoS) attacks on the network. Hotspots can spread to neighboring regions and can cause non-localized impact. We use percolation techniques to analyze the impact of cascading hotspots.

  • The tool EpiNet provides a simulation framework to study the spread of malware in wireless networks. We compare the dynamics of Bluetooth worm epidemics over realistic wireless networks and networks generated using random waypoint mobility models and show that realistic wireless networks exhibit very different structural properties. Importantly, these differences have significant qualitative effect on spatio-temporal dynamics of worm propagation.

  • This system can be used to study human initiated cascading failures in societal infrastructures. We study an evacuation scenario caused by a chemical plume in a densely populated urban region. Interdependencies between transportation and communication infrastructures are examined; and methods developed to identify critical base stations and regions, and quantify the impact on the overall call traffic.

  • Our modeling environment has been used to study the impact of geographic complementarity on the demand bids of wireless service providers, purchase value of licenses, utilization of capacity and the level of demand unmet. We also evaluate the performance of market clearing mechanisms on the spectrum markets for regions with different levels of complementarity, and the potential impact of sequential ordering in running the auctions among the regions.

  • A computer simulation based study of a large-scale, nuclear event in a densely populated urban region is carried out . The study focuses on interaction between human behavior and physical infrastructures in the aftermath of the crisis. Emergency boadcasts immediately after the event, advise people to shelter in place. The results show that this relatively mild intervention can have a large beneficial impact.

  • Offloading cellular traffic through opportunistic communications is a promising solution to handle mobile data traffic. We investigate the target-set selection problem for information delivery in the emerging Mobile Social Networks (MoSoNets). We propose to exploit opportunistic communications to facilitate the information dissemination and thus reduce the amount of cellular traffic. In particular, we study how to select the target set with only k-users, such that we can minimize the cellular data traffic.

Publications

 

Journals Publications

  1. R. Beckman, K. Channakeshava, F. Huang, J. Kim, A. Marathe, M. Marathe, S. Saha, G. Pei and A. Vullikanti, 2013. Integrated Multi-Network Modeling Environment for Spectrum Management. IEEE Journal on Selected Areas in Communication (JSAC), special issue on Network Science, vol 31, issue 6, June 2013, pages 1158-1168.
  2. Pei G, Parthasarathy S, Srinivasan A, Kumar VS Anil (2013) Approximation algorithms for throughput maximization in wireless networks with delay constraints. IEEE/ACM Transactions on Networking, 21(6): 1988-2000.
  3. C. Barrett, K. Channakeshava, F. Huang, J. Kim, A. Kumar, A. Marathe, M. Marathe, and G. Pei, 2012. Human Initiated Cascading Failures in Societal Infrastructures. PLoS ONE, vol. 7, no. 10, October.
  4. Han B, Hui P, Kumar VS Anil, Marathe M, Shao J, Srinivasan A (2012) Mobile Data Offloading through Opportunistic Communications and Social Participation. IEEE Transactions on Mobile Computing, 11(5): 821-834.
  5. C. Barrett, K. Channakeshava, S. Eubank, V. S. Anil Kumar, and M. Marathe. From biological and social network metaphors to coupled bio-social wireless networks. International Journal of Autonomous Adaptive Communications, 4(2):122–144, 2011.

Conference Publications

  1. Channakeshava K, Bisset K, Marathe M, Kumar VS Anil (2014) Reasoning about mobile malware using high performance computing based population scale models. In Proceedings of the 2014 Winter Simulation Conference, 3048-3059. Savannah, GA, December 7-10, 2014. http://dl.acm.org/citation.cfm?id=2694232
  2. Abdelhamid S, Alam M, Alo R, Arifuzzaman S, Beckman P, Bhattacharjee T, Bhuiyan H, Bisset K, Eubank S, Esterline A, Fox E, Fox G, Hasan S, Hayatnagarkar H, Khan M, Kuhlman C, Marathe M, Meghanathan N, Mortveit H, Qiu J, Ravi S, Shams Z, Sirisaengtaksin O, Swarup S, Vullikanti A, Wu T (2014) CINET 2.0: A CyberInfrastructure for Network Science. In The 10th IEEE International Conference on eScience, 2014., 324-331. Sao Paulo, Brazil, October 20-24, 2014. http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=6972280&tag=1 10.1109/eScience.2014.21
  3. S. Chandan, S. Saha, C. Barrett, S. Eubank, A. Marathe, M. Marathe, S. Swarup, and A. Vullikanti, 2013. Modeling the Interaction between Emergency Communications and Behavior in the Aftermath of a Disaster. International Conference on Social Computing, Behavioral Cultural Modeling, & Prediction (SBP) , Washington DC, April 2-5, pages 476-485.
  4. J. Kim, V.S.A. Kumar, A. Marathe, G. Pei, S. Saha (2012). Analysis of Policy Instruments for Enhanced Competition in Spectrum Auction. Proceedings of IEEE Dynamic Spectrum Access Networks (DySPAN). Bellevue, Washington, October 16-19.
  5. J. Kim, V.S.A. Kumar, A. Marathe, G. Pei, S. Saha (2012). Clearing Secondary Spectrum Market with Spatio-temporal Partitioning. Proceedings of IEEE Dynamic Spectrum Access Networks (DySPAN). Bellevue, Washington, October 16-19.
  6. J. Kim, V.S. Anil Kumar, A. Marathe, G. Pei, S. Saha and B. Sunapanasubbiah. Modeling Cellular Network Traffic with Mobile Call Graph Constraints, Proceedings of the 2011 Winter Simulation Conference, 2011.
  7. G. Pei, V.S. Anil Kumar, S. Parthasarathy and A. Srinivasan. Approximation algorithms for throughput maximization in wireless networks with delay constraints, IEEE Conference on Computer Communications (INFOCOM), 2011.
  8. K. Bisset, J. Chen, C. Kuhlman, V. S. Anil Kumar, and M. Marathe. Interaction-based HPC modeling of social, biological and economic contagion over large networks. In S. Jain, R. R. Creasey, J. Himmelspach, K. P. White, and M. Fu, editors, Proceedings of the Winter Simulation Conference, Phoenix, Arizona, USA, Dec 11-14 2011.
  9. K. Channakeshava, K. Bisset, M. Marathe, V. S. Anil Kumar, and S. Yardi. High performance scalable and expressive modeling environment to study mobile malware in large dynamic networks. In Proceedings of 25th IEEE International Parallel & Distributed Processing Symposium, Anchorage, Alaska, May 16–20 2011.
  10. J. Kim, V.S. Anil Kumar, A. Marathe, G. Pei, S. Saha and B. Sunapanasubbiah. Impact of Geographic Complementarity in Dynamic Spectrum Access, Proceedings of IEEE International Dynamic Spectrum Access Networks (DySPAN), 2011.
  11. R. Beckman, K. Channakeshava, F. Huang, V.S.A. Kumar, A. Marathe, M. Marathe, G. Pei (2010). Synthesis and Analysis of Spatio-Temporal Spectrum Demand Patterns: A First Principles Approach. IEEE Dynamic Spectrum Access Networks (DySpan), April, Singapore.
  12. R. Beckman, K. Channakeshava, F. Huang, V.S.A. Kumar, A. Marathe, M. Marathe, G. Pei (2010). Implications of Dynamic Spectrum Access on the Efficiency of Primary Wireless Market. IEEE Dynamic Spectrum Access Networks (DySpan), April, Singapore.
  13. C. Barrett, R. Beckman, K. Channakeshava, F. Huang, V. S. Anil Kumar, A. Marathe, M. Marathe, and G. Pei. Cascading failures in multiple infrastructures: From transportation to communication network. In Proceedings of the Conference on Interacting Critical Infrastructures for the 21st Century, pages 1–8, Beijing, China, Sep 20–22 2010.
  14. B. Han, P. Hui, V. S. Anil Kumar, M. Marathe, G. Pei, and A. Srinivasan. Cellular traffic offloading through opportunistic communications: A case study. In Proceedings of the 5th ACM Workshop on Challenged Networks (CHANTS), Chicago, Illinois, Sep 2010.
  15. Karthik Channakeshava and Deepti Chafekar and Keith Bisset and Anil Vullikanti and Madhav Marathe. EpiNet: A Simulation Framework to Study the Spread of Malware in Wireless Networks, SIMUTools09, ICST Press, Rome Italy, March 2009.