« ITSGWN» is an Associate Team between the INRIA projectteam TREC of INRIA ParisRocquencourt, and the EECS department of UC Berkeley in the USA, funded from 2011 to 2014. This Associate Team participates in the Inria@SiliconValley initiative.
28th October 2012  Contact: F. Baccelli
News
V. Anantharam and F. Baccelli obtained new results on error exponents for multiuser channels using the novel techniques they have developed based on point process theory. Progress was made on characterizing the error exponent region for a large class of multiple access channels. Three talks have been given on this work lately (see below).
Mohsen Bayati (Graduate School of Business Stanford), Marc Lelarge and Andrea Montanari (EE and Statistics Stanford) obtained new results about the universality of a certain phase transition arising in polytope geometry and compressed sensing. A preliminary version of the work was presented at ISIT 2012; here is the extended version.
J. Salez and V. Anantharam worked on a information theoretic conjecture with wide ranging implications for the design of rateless codes. Significant progress was made on this conjecture, but it remains unresolved. It is hoped to continue work on this in 2013. They also worked on the structure of Nash equilibria in large routing games. The aim was to characterize the equilibria as fixed points of an equation coming from the studying the interaction of a single agent with the rest of the environment. This work has been successfully completed and a paper presenting this work is being prepared.
The Associate Team
 Presentation of the collaboration This collaboration is centered on the interplay between stochastic geometry and network information theory, whith a particular emphasis on wireless networks.
 Members

INRIA Project Team: The INRIA Project Team at the origin of this proposal is TREC
(ParisRocquencourt and ENS).
The coordinator on the INRIA side is Francois Baccelli, 23 avenue d’Italie, CS 81321 75214 Paris Cedex 13. Tel: (33) 139 63 54 29, Mobile: (33) 678 06 85 44 email: Francois.Baccelli@ens.fr. The INRIA team is composed of the coordinator and Bartek Blaszczyszyn . The team has strong collaborations on the topics of the proposal with Paul Muhlethaler of Hipercom (ParisRocquencourt).
 International Partner: The international partner is the EECS Department of UC Berkeley, USA. The coordinators on the US side are Venkat Anatharam, A. Sahai, and David Tse, Electrical Eng. and Computer Science Dept. University of California, 271 Cory Hall, Berkeley, CA 947201774. All are professors at UC Berkeley in the EECS department and members of the Wireless Foundations Center. The collaboration extends to Prof. Abbas El Gamal with whom there is an ongoing joint collaboration: Prof. Abbas el Gamal, Information Systems Laboratory, Department of Electrical Engineering, Stanford University, Packard 262, Stanford, CA 94305.

INRIA Project Team: The INRIA Project Team at the origin of this proposal is TREC
(ParisRocquencourt and ENS).
 Research Program The research is centered on the interplay between stochastic geometry and network information theory, whith a particular emphasis on wireless networks. In terms of research, three main lines of thought are:
 1. Error exponents and stochastic geometry (VA, FB, BB, AS). We continue the approach started in [AB08] on the analysis of the error exponent (reliability function) for classical channels in Information Theory. Our aim is to combine techniques of point process theory, stochastic geometry, large deviations and perturbation analysis to obtain general results on error exponents associated with stationary and ergodic noise for classical channels of information theory as well as network information theory.
 2. Stochastic geometry and network Information Theory (VA, FB, BB, AEG, PM, DT). We investigate the interplay between Network Information Theory [EK10] and stochastic geometry [BB09] to analyze large wireless networks. The aim of this research is to derive innovative distributed communication strategies alleviating the interference limitations and allowing one to transfer information more efficiently through large wireless networks. The first step of this general program was completed lately in [BET10] in collaboration with A. El Gamal [Stanford, USA] and D. Tse, where a global estimation of the improvement brought by simultaneous decoding in such networks was performed using stochastic geometry.
 3. Cognitive radio and stochastic geometry (FB, BB, PM, AS). We propose new probabilistic models based on stochastic geometry to analyze cognitive radio in large networks with randomly located users using e.g. carrier sensing multiple access, or in large cellular networks. The aim is to derive analytical results on the impact of the interaction between primary and secondary users, on their medium access probability, coverage probability and throughput. These results give insight on the statistical guarantees which can be offered to primary users and more generally on the possibilities offered by cognitive radio to improve the effectiveness of spectrum utilization [NB10,NB11].
 Publications
 [AB08] V. Anantharam and F. Baccelli, A Palm Theory Approach to Error Exponents, Proc. IEEE ISIT’08, Toronto, July 2008.
 [BB09] F. Baccelli and B. Blaszczyszyn, Spatial Modeling of Wireless Communications – A Stochastic Geometry Approach, Foundations and Trends in Networking, NOW Publishers.
 [AB10] V. Anantharam and F. Baccelli, InformationTheoretic Capacity and Error Exponents of Stationary Point Processes under Random Additive Displacements, Lecture at the SCS Programme of the Isaac Newton Institute for Mathematical Sciences, Stochastic Network Workshop, March 2010, arxiv1012.4924.
 [EK10] A. El Gamal and Y.H. Kim, Lecture Notes on Network Information Theory, January, 2010.
 [NB10] T.V. Nguyen and F. Baccelli, A Probabilistic Model of Carrier Sensing Based Cognitive Radio, IEEE DYSPAN’10, Singapore, 2010.
 [BET11] F. Baccelli A. El Gamal and D. Tse, Interference Networks with PointtoPoint Codes, IEEE Tr. IT, 2011, arxiv1102.2868.
 [NB11] T.V. Nguyen and F. Baccelli, A Stochastic Geometry Model for Cognitive Radio Networks, the Computer Journal, 2011 comjnl10.1093.
 [AB12] V. Anantharam and F. Baccelli, Capacity and Error Exponents of Matched Dimensions MAC Channels in the Poltyrev Regime, In preparation, 2012.
 Visits, Postdocs and Internships
 The postdoc of J. Salez (former PhD student of TREC) at UC Berkeley under the joint supervision of V. Anantharam and D. Aldous is now completed. Justin Salez is now Maître de Conférences in the Mathematics Department of Université Paris 7.
 F. Baccelli visited U.C. Berkeley in Spring 2012, 15 April, 15 May, 2012.
 V. Anantharam visited INRIA in Summer 2012, 03 June  14 June, 2012.
 A. El Gamal visited Paris in March 2012 and gave a lecture.
 V. Anantharam visited INRIA/ENS from April to July 2011 and gave several lectures.
 K. Woyach, PhD student of A. Sahai, spent a week at INRIA in April 2011 and gave a lecture.
 A. Sahai spent a week at INRIA in September 2010 and gave a survey lecture on Cognitive Radio at the LINCS.
 F. Baccelli held a 6 month Miller Institute Professor position at UC Berkeley in 2009.
 Workshops, Talks on joint research, Events
 Venkat Anantharam gave an invited lecture on [AB10][AB12] "Capacity and Error Exponents in the Poltyrev Regime." at HP Labs, Palo Alto on 24 October 2012.
 Venkat Anantharam gave an invited lecture on [AB12] "Capacity and Error Exponents for Multipleaccess Channels in the Poltyrev Regime" at the Department of Electrical Engineering, Indian Institute of Technology, Powai, Mumbai, India on 09 August 2012.
 A TREC inria@siliconvalley workshop entitled "Random Spatial Models in Communication Science" was held at INRIA Paris on June 13 2012.
 F. Baccelli gave a lecture on the joint research with V. Anantharam at the Information Theory event of GdR ISIS, 20120406.
 F. Baccelli gave a lecture on the joint work with V. Anantharam at Laboratoire de Probabilité in Paris, February 2012.
 Venkat Anantharam gave a lecture at the Colloquium of Departement d'Informatique of ENS in June 2011.
 Anant Sahai gave a lecture on Cognitive radio at the Inaugural Workshop of the programme held at UC Berkeley in June 2011.
 David Tse gave a lecture on [BET11] at IEEE ISIT in St Petersburg, July 2011.
 Plans for 2013
 F. Baccelli will visit UC Berkeley in March.
 V. Anantharam will visit Paris in June.
 Joint research on all 3 topics.