Skip to Main content Skip to Navigation
Theses

Exploitation ciblée du canal dans l’IoT : amélioration de la localisation et de l’allocation du spectre dans les réseaux LPWAN

Abstract : The main objective of this work is to improve various LPWAN functions, by exploiting the knowledge of the propagation properties of the wireless channel. First, localization techniques (without GPS) are studied and then enhanced by proposing an original parametric TDoA technique. In this context, preprocessing techniques for observables specific to the TDoA technique are proposed. Using a specific simulator implemented in the framework of the thesis, the results show that the proposed approaches are more efficient than the classical TDoA technique. On the other hand, the use of CSI for localization is advocated, the thesis studies its spatial and temporal variability. Through a measurement campaign, the reliability of the use of ESP in substitution of RSSI is affirmed by its increased range. At this stage, the LoRaWAN PDR is modeled according to the ESP. Then, the work is oriented towards the improvement of the PDR by proposing spectrum allocation algorithms that exploit the frequency dependency of the PDR. Thus, two decentralized policies are proposed to learn an appropriate frequency allocation scheme based on the ESP as channel quality information. Experimentally, these proposed algorithms outperform the conventional UCB policy with less packet loss at the end of the process.
Complete list of metadata

https://tel.archives-ouvertes.fr/tel-03708143
Contributor : ABES STAR :  Contact
Submitted on : Wednesday, July 6, 2022 - 3:34:12 PM
Last modification on : Wednesday, July 6, 2022 - 3:34:12 PM

File

NAGY_ABDELGHANY_Ahmed.pdf
Version validated by the jury (STAR)

Identifiers

  • HAL Id : tel-03708143, version 1

Citation

Ahmed Abdelghany. Exploitation ciblée du canal dans l’IoT : amélioration de la localisation et de l’allocation du spectre dans les réseaux LPWAN. Networking and Internet Architecture [cs.NI]. Université Rennes 1, 2021. English. ⟨NNT : 2021REN1S122⟩. ⟨tel-03708143⟩

Share

Metrics

Record views

70

Files downloads

10