Jean Bolot

I study mobile and home networks, their users, and the data they generate

I am the director of the Technicolor Research Lab, which we just opened in downtown Palo Alto, steps from the Stanford University campus. The goal of the lab is to create and develop the next generation of personalized TV, movie and human experiences. Until recently, I led the Research Group of Sprint, located in the San Francisco Bay Area at the northern tip of Silicon Valley, working with a wonderful group of colleagues and academic collaborators. At Sprint, our goal was to understand the performance and the economics of the mobile Internet, and the behavior of its users. In general, my research focuses on areas at the interface of data mining, mobile and home networking, and economics.

Before Sprint, I was a founding team member of Ensim, a Silicon Valley startup in the area of virtualization and data center automation software. Before Ensim, I was a visiting professor in Computer Science at UC Berkeley, visiting professors Randy Katz and Steve McCanne, and a researcher at INRIA in France.

[ What's new] [ Research - Papers and Talks] [ Collaborators] [ Contact information]

What's New

• New paper: Modeling the Economic Value of the Location Data of Mobile Users (with F. Baccelli), to appear in IEEE Infocom 2011, Shanghai, China, March 2011.
• New paper: The Problem of Sensing Unused Spectrum (with D. Willkomm, S. Machiraju, A. Wolisz), to appear in IEEE Networking 2011, Valencia, Spain, May 2011.
• Paper: Bayesian Inference for Localization in Cellular Networks (with F. Baccelli, H. Zang), to appear in IEEE Infocom 2010, San Diego, CA, March 2010. We present a general technique based on Bayesian inference to locate mobile devices in cellular networks. We consider in particular the case when only one base station is visible from by the device, meaning that triangulation cannot be done, which happens most of the time in practice (yes, it does).
• Paper: The Role of PASTA in Network Measurement (with F. Baccelli, S. Machiraju, D. Veitch), IEEE/ACM Transactions on Networking, vol 17, no 4, August 2009. We consider the optimal way to measure networks such as the Internet using active probing. We argue that the celebrated PASTA (Poisson Arrivals See Time Averages) property used to justify the sending of probe packets at Poisson times is of limited use. We give concrete examples of when Poisson probes should or should not be used, explain why, and offer guidelines on suitable alternative sending processes.
• Paper: Economic Incentives to Increase Security in the Internet: The Case for Insurance (with M. Lelarge), IEEE Infocom 2009, April 2009, Rio de Janeiro. We consider the problem of designing incentives to entities in the Internet so they invest in security at a socially efficient level, and find that insurance is a powerful such incentive.
• Paper: Primary User Behavior in Cellular Networks and Implications for Dynamic Spectrum Access, IEEE Communications Magazine, vol 47, no 3, March 2008. We measure and analyze the temporal and spatial characteristics of spectrum usage in a large scale CDMA cellular network. We then discuss what this means for cognitive radio usage in that network.
• Paper: Increasing Wireless Security through Network Diversity (with T. Ye, D. Veitch), Computer Communication Review , January 2009.
• Best paper award: Primary Users in Cellular Networks: A Large-scale Measurement Study, IEEE Dyspan (Dynamic Spectrum and Cognitive Radio) 2008. We measure and analyze the temporal and spatial characteristics of spectrum usage in a large scale CDMA cellular network. We then discuss what this means for cognitive radio usage in that network.
• Paper: A Local Mean-Field Analysis of Security Investments in Networks, NetEcon (Network Economics) 2008 - ACM Sigcomm workshop. In our Sigmetrics 2008 paper, we developed a very general model to analyze the economics of epidemic processes, by overlaying an economic framework on a general epidemic propagation model. In this paper, we use a local mean-field approach to solve the model, and derive insight in a specific problem, namely the economic aspects of security investments in networks subject to virus and botnet attacks.
• Paper: Mobile Calls Graphs: Beyond Power-Law and Lognormal Distributions, ACM KDD (Data Mining) 2008 . We study the structure of the social networks created by cell phone users. We find that the networks do not follow the usual power law or lognormal distributions. Instead, we find they follow a Double Pareto LogNormal (DPLN) distribution. We introduce the notion of "social wealth" and describe a generative process for the DPLN distribution.
• Paper: Network Externalities and the Deployment of Security Features and Protocols in the Internet, ACM Sigmetrics 2008.
• Paper: Mining Call and Mobility Data to Improve Paging Efficiency in Cellular Networks, ACM Mobicom 2007. We study the mobility patterns of cell phone users and develop mobility profiles to derive a family of very effective profile-based paging techniques.

Research - Papers and Talks

Mining large scale cell phone data: social networks, location and mobility, spectrum usage

Network measurement: theory and practice

Economics of networked systems

Economics of security

Wireless security

Resource allocation in the Internet

Application-level resource allocation: adaptive voice and video over IP

Current research collaborators