Optimizing Massive Random Access: Leveraging Correlation Models and Sparse Recovery Algorithms
Thesis event information
Date and time of the thesis defence
Place of the thesis defence
L5, Linnanmaa
Topic of the dissertation
Optimizing Massive Random Access: Leveraging Correlation Models and Sparse Recovery Algorithms
Doctoral candidate
Master of Science Hamza Djelouat
Faculty and unit
University of Oulu Graduate School, Faculty of Information Technology and Electrical Engineering, CWC - Radio Technologies (CWC-RT)
Subject of study
Communications engineering
Opponent
Docent Mikko Vehkaperä, Nokia
Custos
Professor Markku Juntti, University of Oulu
Optimizing Massive Random Access: Leveraging Correlation Models and Sparse Recovery Algorithms
Massive machine-type communications (mMTC) serve as the backbone for future internet of things (IoT) applications. Characterized by sporadic traffic from a massive number of connected devices, mMTC presents unique challenges for traditional channel access protocols in terms of efficient channel access and resource allocation. Therefore, the success of the IoT applications in the next 5G/6G era is bounded by the optimal design of novel radio resource management protocols that support energy efficient and scalable, and unlimited wireless connectivity.
This thesis aims to optimize random access in mMTC by employing sophisticated correlation models inherited in both the wireless propagation channel and IoT traffic patterns. The aim is to go beyond the often-used simplistic models in the literature by employing more realistic and practical channel models as well as capturing a more nuanced realization of real-world device interactions within the large-scale IoT network.
The contribution of this thesis unfolds in two main directions: First, we propose, develop and optimize various solutions, formulated within the frameworks of both deterministic and Bayesian sparse recovery to account and model the different level of available prior information, both in terms of channel statistics and traffic patterns. Second, the thesis goes beyond its algorithmic contributions to provide an in-depth theoretical analysis for related performance benchmarks.
By capitalizing on the practical correlation structures found in mMTC, this research achieves a significant milestone, delivering solutions that not only match but potentially exceed the performance benchmarks of the state-of-the-art solutions, all while minimizing communication overhead and reducing power consumption.
This thesis aims to optimize random access in mMTC by employing sophisticated correlation models inherited in both the wireless propagation channel and IoT traffic patterns. The aim is to go beyond the often-used simplistic models in the literature by employing more realistic and practical channel models as well as capturing a more nuanced realization of real-world device interactions within the large-scale IoT network.
The contribution of this thesis unfolds in two main directions: First, we propose, develop and optimize various solutions, formulated within the frameworks of both deterministic and Bayesian sparse recovery to account and model the different level of available prior information, both in terms of channel statistics and traffic patterns. Second, the thesis goes beyond its algorithmic contributions to provide an in-depth theoretical analysis for related performance benchmarks.
By capitalizing on the practical correlation structures found in mMTC, this research achieves a significant milestone, delivering solutions that not only match but potentially exceed the performance benchmarks of the state-of-the-art solutions, all while minimizing communication overhead and reducing power consumption.
Last updated: 15.4.2024