Méthodes algorithmiques pour l'optimisation mono-objectif et multi-objectif : application aux réseaux de radiodiffusion

Abstract : The purpose of my thesis is the dynamic construction of radio communication networks subject to multiple constraints and the optimal (if possible) use of the set of available frequencies at time t. This is an NP-Hard problem with important economical issues. I have designed and used original hybrid meta-heuristics for solving this kind of problems and providing the best possible QoS. Frequencies are rare and expansive therefore we can ask if a single frequency would not be enough, for a while, in order to cover the needs of a given geographical area. This would allow the use of the remaining frequencies for other applications. This is the principle of Single Frequency Networks (SFN) which necessitate the simultaneous optimization of transmission delays, of the allocated frequency and of their design. I have designed an original and e_cient software which performs these operations. We give experimental results for real benchmarks provided by TDF.
Document type :
Theses
File URL :
http://docnum.univ-lorraine.fr/prive/DDOC_T_2015_0135_BEDOUI.pdf
Complete list of metadatas

https://hal.univ-lorraine.fr/tel-01751940
Contributor : Thèses Ul <>
Submitted on : Thursday, March 29, 2018 - 1:29:12 PM
Last modification on : Tuesday, April 24, 2018 - 1:30:18 PM

Identifiers

  • HAL Id : tel-01751940, version 1

Collections

Citation

Akram Bedoui. Méthodes algorithmiques pour l'optimisation mono-objectif et multi-objectif : application aux réseaux de radiodiffusion. Autre [cs.OH]. Université de Lorraine, 2015. Français. ⟨NNT : 2015LORR0135⟩. ⟨tel-01751940⟩

Share

Metrics

Record views

19