Skip to Main content Skip to Navigation

Accès massif pour l'IoT dans la 5G et au delà

Abstract : The integration of the Internet of Things in next generations of wireless communication systems has raised very interesting technical challenges, particularly at the MAC layer, where existing resource allocation techniques become inefficient with the increasing number of devices. Indeed, the drastic evolution of the number of connected devices requires a system that is able to support sporadic traffic and at the same time guarantees reliable performance and stringent quality of service. In this context, conventional orthogonal multiple access techniques are no longer able to cope with this increasing and explosive traffic demand due to the extremely limited transmission bandwidth. Thus, non-orthogonal multiple access (NOMA) is considered as a promising approach to meet the massive access demand and bursty traffic by allowing multiple users to access the same time-frequency resource. In this thesis, we consider NOMA multiplexing in the power domain for uplink and downlink transmissions. To guarantee the required level of quality of service in the network, the clustering algorithms that we have proposed take into account the global capacity of the network as well as the fairness between the users, and at the same time allow to establish a massive connectivity. Indeed, we propose resource allocation and power control techniques to improve system performance in terms of throughput and number of connected users. To overcome the problem of collisionsdue to inter-cellular interference, we use reinforcement learning techniques to optimize the resource allocation. The proposed techniques show the interest of NOMA compared to the classic orthogonal access and point out future directions of research to beinvestigated in order to make possible the integration of this technique, which is rather in its infancy, in the future. communication systems.
Complete list of metadata
Contributor : ABES STAR :  Contact
Submitted on : Thursday, September 22, 2022 - 9:18:15 AM
Last modification on : Friday, September 23, 2022 - 4:34:16 AM


Version validated by the jury (STAR)


  • HAL Id : tel-03783230, version 1



Mohamed Ali Adjif. Accès massif pour l'IoT dans la 5G et au delà. Traitement du signal et de l'image [eess.SP]. Université de Limoges, 2022. Français. ⟨NNT : 2022LIMO0063⟩. ⟨tel-03783230⟩



Record views


Files downloads