[Tccc] PhD proposal in mobile networking, in Sophia Antipolis, France

Lucile Sassatelli sassatelli
Thu Nov 10 05:24:27 EST 2011


*PhD thesis proposal*

*Title*: *Optimal routing policies and network coding for mobile social 
networks *

*Advisors*: Lucile SASSATELLI (Assistant professor, contact: 
sassatelli at i3s.unice.fr, http://www.i3s.unice.fr/~sassatelli/)

*Funding*: This PhD thesis is proposed in the framework of the ANR JCJC 
project TRouP WilMA, and will be done at the I3S laboratory, 
CNRS/University Nice Sophia Antipolis, France.

The salary will amount to 1450 euros per month after tax, for 3 years.

*Application*: Send CV, motivation and recommendation letters, as well 
as referees and master marks. Application deadline: December 1st, 2011.

*Description*:

The traditional infrastructure-centric networking paradigm has already 
proven inadequate to stand the growing demand for contents and updates 
generated by widespread mobile applications [N09,V09]. A promising 
alternative to overloading the telecommunication operators' systems lies 
in the exploitation of user interactions through the design of smart 
decisions to convey information. Such approach therefore considers 
Mobile Ad Hoc Networks (MANETs or Delay Tolerant Networks - DTNs), which 
are meant to allow communication between mobile users without any 
infrastructure.

In DTNs, owing to high-speed mobility and possible disruptions, a big 
challenge is that of routing. Conventional Internet routing protocols 
(e.g., RIP and OSPF) as well as routing schemes for MANETs, such as DSR, 
AODV, assume that a complete instantaneous path exists between a source 
and a destination. Thus, if no end-to-end paths exist most of the time, 
these protocols fail to deliver data. Several proposals try to address 
this routing problem, and have all been optimized for mobility models in 
which nodes scatter uniformly over the whole space along time, such as 
in the independent identically distributed (i.i.d.) model, random walk 
and random waypoint (RWP). Such models fail to capture the real nature 
of MANETs, driven by space and social acquaintances. Recently, a 
clustered mobility model with heterogeneous nodes with anisotropic 
mobility patterns [GGL08], as typically encountered in realistic 
mobility traces, has been introduced. For such general model, we want to 
propose solutions to the opportunistic routing problem, assuming that 
nodes are not aware of the network topology.

The goal of the PhD thesis is to devise optimal routing policies and 
identify delay/energy trade-offs when the mobile nodes are oblivious of 
their own and other nodes' motion, for nodes with heterogeneous physical 
layer and clustered mobility, with the help of network coding. Network 
coding is a new network paradigm that has emerged in the last ten years. 
It consists in allowing intermediate nodes to modify information packets 
before forwarding them, and has proven to bring great advantages for 
several metrics in various network scenarii.

First, we will consider that a node has a single home-point, that there 
is a single unicast connection in the network. The optimal forwarding 
policies will be investigated, building on [ABP10], both theoretically 
and from simulations. Specifically, we will assess the performance of 
such policies with synthetic and real-world contact traces. How to fit 
such routing policies to real-world mobility will be addressed with the 
help of state-of-the-art clustering algorithms.

One step ahead, the PhD student will tackle the more intricate case of 
multiple home points, i.e., when a node belongs to several communities. 
Furthermore, the gain using intra-session network coding will be 
studied, and the model will be extended to the general case with network 
coding, multiple session, general buffer sizes and managements and 
contact bandwidths.

The third step will address the design of smart inter-session coding 
schemes for multi-session DTNs. Inter-session network coding is an 
active field of research, and effective routing schemes taking advantage 
of it are still to be found. Additionally, delay/throughput trade-offs 
under such model and routing scheme will be derived.

*Pre-requisites:*

Background in wireless communications, networking and statistics.

*References:*
[N09] www.nytimes.com/2009/09/03/technology/companies/03att.html
[V09] venturebeat.com/2009/05/11/iphone-users-eating-up-atts-network
[AGL09] G. Alfano, M. Garetto and E. Leonardi, "Capacity scaling of 
wireless networks with inhomogeneous node density: Upper bounds," IEEE 
Journal on Selected Areas in Communications, 27(7):1147-1157, 2009.
[ABP10] E. Altman, T. Basar, and F. De Pellegrini, "Optimal Monotone 
Forwarding Policies in Delay Tolerant Mobile Ad-Hoc Networks", Elsevier 
Performance Evaluation, issue 67 (2010) pp. 299-317.

-- 
Lucile SASSATELLI
Maitre de conf?rence - Assistant professor
E-mail: sassatelli at i3s.unice.fr	, Tel: +33(0) 4 92 94 27 72
Webpage: http://www.i3s.unice.fr/~sassatelli/

Laboratoire I3S - UMR6070 - UNSA CNRS	
06903 Sophia Antipolis Cedex - France

Universit? de Nice
IUT Nice C?te d'Azur - D?partement R?seaux et T?l?coms

-------------- next part --------------
A non-text attachment was scrubbed...
Name: PhDproposal.pdf
Type: application/pdf
Size: 63219 bytes
Desc: not available
Url : https://lists.cs.columbia.edu/pipermail/tccc/attachments/20111110/44ddb4bd/PhDproposal-0001.pdf



More information about the TCCC mailing list