Un codage réseau contraint pour les réseaux de capteurs sans fil

Abstract : Today, it is likely that Machine-to-machine (M2M) architectures are the future of computer networks. These communication systems promise new strains of pervasive services and a better awareness of the world around us. However, they suffer several limitations caused by the very nature of their components: a low data-rate, unreliable transmissions, limited storage and processing capabilities, finite lifespan and unpredictable delays often caused by a high packet loss rate. These are the same issues that are still preventing, since the late eighties, the awaited large-scale deployment of wireless sensor networks in the world of telecommunications. Recent developments in the Information Theory are finally bringing a wind of change on the common ways of considering data communications, via a new concept known as the Network Coding Theory. This new data delivery approach is transforming the way we operate, manage and even understand existing and future communications networks. In this context, we propose in this thesis to study the impact, benefits and feasibility of network coding in wireless sensor networks. Our goal is to evaluate the cost/benefit value of its practical concepts for different M2M network paradigms in general, and for sensor networks in particular. Primarily, we assess the potential impact of opportunistic and linear network coding on sensor networks strictly in terms of performance (i.e., data rate and latency). Then we address the problem of reliable communications in coded sensor networks, since such systems are characterized by the absence of end-to-end flow control mechanisms. In the last chapter, we explore the feasibility of network coding in a real testbed. We investigate how an innovative coding technique specifically designed for low-power/low-rate networks can use opportunistic coding to enhance the performance and the reliability of the network. More broadly, this thesis presents one answer to the question: “Which network coding for what sensor network?” It does not claim to be the sole answer to this issue, but rather a series of theoretical and technological blocks that denote both the feasibility of network coding in constrained environments and also the need to adapt it to their particular criteria
Document type :
Theses
Complete list of metadatas

Cited literature [80 references]  Display  Hide  Download

https://pastel.archives-ouvertes.fr/pastel-00794618
Contributor : Abes Star <>
Submitted on : Tuesday, February 26, 2013 - 11:22:49 AM
Last modification on : Wednesday, July 4, 2018 - 4:38:00 PM
Long-term archiving on : Monday, May 27, 2013 - 5:55:12 AM

File

TH2012PEST1103_complete.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : pastel-00794618, version 1

Citation

Ismail Salhi. Un codage réseau contraint pour les réseaux de capteurs sans fil. Autre [cs.OH]. Université Paris-Est, 2012. Français. ⟨NNT : 2012PEST1103⟩. ⟨pastel-00794618⟩

Share

Metrics

Record views

1624

Files downloads

3033