Espace membre

Cet espace est dédié aux étudiants, aux enseignants et au personnel administratif de l'école

Valider

Mot de passe oublié?

Actualités de l'école

07/12/2018

Doctorate thesis defense of Sarra REBHI




Doctorate thesis defense on December 07th 2018 at 09H00 ,in Sup’Com Amphitheater Ibn Khaldoun.


Entitled :Development of New Cognitive Techniques for Radio over Fiber Systems

Presented by :

Sarra REBHI


Committee


President

Fethi CHOUBANI

Professor at Sup'Com, Tunisia

 

 

 

Examiners

Houria REZIG

Professor, ENIT, Tunisia

 

Habib FATHALLAH

Associate Professor, FSB, Tunisia

 

Leila NAJJAR

Associate Professor, SUP’COM, Tunisia

 

 

 

Thesis Director :

Mourad MENIF

Professor, SUP’COM, Tunisia

 

Abstract


This thesis falls within 5G small-cell transport systems. The new fronthauling solutions for metro/aggregation segment and dense small-cells coverage are considered as ones of the most significant challenges for the upcoming 5G cellular systems. The 5G transport technologies require high flexible optical transport system aware of radio service specifications and able to accommodate high traffic and provide high throughput in strict requirement networks.


In this context, the aim of our thesis work is to develop a cognitive RoF system composed of a flexible platform able to integrate several adaptation techniques. This platform is managed by a high-efficiency cognitive management system responsible for orchestrating and performing complex decisions in a strict latency requirement environment.


The first part of our thesis consists in developing a flexible and dynamic RoF architecture. We propose a cognitive architecture able to meet the unpredictable resource requirements in a dense small cell coverage. The RoF architecture provides high reconfiguration capabilities through the integration of high flexible nodes and reconfigurable equipments. The high rate requirement and ever increasing control data limit the use of classical management systems. The introduced cognitive management system is based on Software Defined Networking paradigm that segregates the data plane from the control plane and offers high programmability capabilities. The cognitive management system is suited to hybrid network and the hierarchical topology. We expose the developed management operation and we detail the cognitive processing. Then, we set the needed monitoring tasks and the required SDN controllers.


The next step in our thesis consists in implementing a performing algorithm aligned with cited requirements and multi-criteria RoF system. We implement a multi-objective optimization algorithm able to perform a complex decision in a multi-tasks optimization network.


Finally, we propose some cognitive adaptation techniques and we integrate our decision algorithms in the cognitive SDN based management system providing optimal solutions for network adaptation and enhancing the RoF system performance.


Keywords :


Radio over Fiber, cognitive techniques, Software Defined Networking, Millimeter wave band, optical pre-equalizer, adaptive modulation, multi-objective optimization.