PhD graduated
Team : Phare
Departure date : 01/16/2015
Supervision : Guy PUJOLLE
Co-supervision : SECCI Stefano

Resource and crowd management in mobile networks

The dissertation presents different solutions to cater the increasing demand of mobile data traffic in access and backhauling networks, taking into account real usage profiling and human mobility patterns.
Using real mobile phone data, we study human trajectories and behaviors. We show that these trajectories can be estimated at good accuracy levels, using specific mobility metrics, namely the radius of gyration and the centroid. Then, motivated by the fact that crowded spot positions change as a function of time and the occurrence of special events in real networks, we propose two crowded spot estimators based on these user mobility metrics. From an empirical evaluation, we show that they appear as excellent crowd detection solutions for cellular backhauling network management. Crowded spot detection can also be of high utility in the optimization of the access network under the resulting resource contention situations.
Our interests move in the second part of the dissertation into finding good offloading solutions to adequately solve resource contention situations around crowded spots. In particular we study two offloading approaches: trac offloading and content offloading. In the former one, we present two solutions: horizontal offloading over small-cell networks and vertical offloading over Wi-Fi certified Passpoint hotspots. We study benefits and limitations of each solution. We propose a new algorithm based on cooperative game theory for resource allocation and interference management in small-cell networks. The results are very impressive and emphasize the necessity of referring to cooperative game theory concepts in the definition of spectrum contention situations. Furthermore, we show the benefits of using the newly released WiFi Passpoint solution in increasing spectrum capacity gain and decreasing user energy consumption.
Defence : 09/19/2014 - 14h - Site Jussieu 25-26/105
Jury members :
Prof. Deep MEDHI (University of Missouri-Kansas City, Etats-Unis) [Rapporteur]
Prof. Bijan JABBARI (George Mason University, Etats-Unis) [Rapporteur]
Dr. Walid SAAD (Virginia Tech, Etats-Unis)
Dr. Marcelo DIAS DE AMORIN (Université Pierre et Marie Curie, France)
Dr Marceau COUPECHOUX (Télécom ParisTech, France)
Dr. Mahmoud DOUGHAN (Université Libanaise, Liban)
Dr. Stefano SECCI (Université Pierre et Marie Curie, France)
Prof. Guy PUJOLLE (Université Pierre et Marie Curie, France)

2012-2018 Publications

 Mentions légales
Site map |