Titulo Estágio
Enhanced Wifi Offloading in Mobile Networks
Áreas de especialidade
Comunicações, Serviços e Infraestruturas
Sistemas Inteligentes
Local do Estágio
DEI-FCTUC
Enquadramento
With the steady growth of smart-phones sales, the demand for services that generate mobile data traffic has grown tremendously, with average generation of mobile data traffic growing 81% in 2012, and the global data traffic expected to increase 8-fold until 2017. This recent boost up of mobile data consumption is struggling the 3G/4G cellular networks, which are not always prepared to receive such demand.
From an operator perspective, WiFi offloading is a promising solution to the recent boost up of mobile data consumption that is making excessive demands on 3G/4G cellular networks in metropolitan areas. The idea consists in shifting the traffic off of typical cellular networks to WiFi networks. From a user perspective this also provides benefits by enabling cheaper connectivity via WiFi networks that are in many cases already available (e.g. [url=https://corp.fon.com/en]Fon[/url]).
Objetivo
The main goal of this project is to advance the state-of-the art of WiFi offloading techniques to address the growing issue of increased usage of cellular networks that may choke the available bandwidth to end-users.
Recent works take into account user mobility for optimizing data offloading. For example, by determining the best locations for access points (AP) by creating a time dependent graph to represent the interaction between people's mobility and traffic demands, and locations suitable to receive an AP (Mucceli & Viana, SECON 2014).
We look at the AP placement/selection problem from a different perspective than above: considering that hotspots are already deployed and available (e.g. [url=https://corp.fon.com/en]Fon[/url] APs), devise a mechanism for selection of APs for data offloading that takes into account user mobility/routines. The network card will select the AP offering the best signal, but this is not necessarily the best one in terms of available capacity, connection duration, etc. For that, prediction will be an important aspect to consider.
Plano de Trabalhos - Semestre 1
1) State-of-the art study on:
- current AP selection/association techniques
- mobility prediction mechanisms
2) Implement and evaluate a simple scenario where users are moving around and selecting APs for communication/data offloading with current association/AP selection mechanisms – evaluate relevant metrics (throughput, communication duration, etc);
3) Definition and characterization of which information is expected to be available from APs (both live and historical info);
4) Study of metrics that take relevant information from the previous point to facilitate the selection of the best AP;
5) Intermediate report.
Plano de Trabalhos - Semestre 2
1) Explore the parameter space of the previous metric and define a mechanism that performs AP selection based on it;
2) Implement and evaluate the developed mechanism for selection of APs and compare with the initial (plain) results;
3) Implement prediction mechanisms that can be used to improve the efficacy of AP selection, taking into account factors such as user mobility patterns, hotspot (historical) response, etc;
4) Develop/employ prediction mechanisms to assess the expected behavior of metrics such as the connection duration and available capacity of APs;
5) Use these metrics to select APs that are suitable to support more data offloading;
6) Implement and evaluate the developed mechanism for selection of APs in a realistic propagation simulator ([url=http://vehicle2x.net/]GEMV2[/url]);
7) Writing of scientific article and dissertation.
Condições
Possibility of an internship at [url=http://www.inria.fr/en/]INRIA[/url], Paris, France, with financial support of ~500€/month.
Internship in collaboration with a researcher from INRIA (Paris, France).
Observações
(Mucceli & Viana, SECON 2014) From routine to network deployment for data offloading in metropolitan areas, IEEE SECON, Singapore, June 2014. Available [url=https://hal.inria.fr/hal-00931786v3/document]here[/url].
The implementation and evaluation of the developed mechanisms shall be performed using the [url=http://vehicle2x.net/]GEMV2[/url] simulator.
Orientador
João Vilela (DEI), Aline Viana (INRIA, France)
jpvilela@dei.uc.pt 📩