Doctorate thesis defense of Marwa Fattoum Belgacem

Doctorate thesis defense on December 4^th 2023 at 09H00 AM ,in Amphitheater Ibn Khaldoun, SUP'COM 2.

Entitled :Distributed Data Aggregation for Energy Efficiency in Heterogeneous IoT Networks: Water Resource Management

Presented by :Marwa Fattoum Belgacem

Committee

Abstract

Over the past decades, Internet of Things (IoT) has seen a perpetual and great evolution due to the interest of researchers in finding new methodologies of interconnecting diverse devices in many fields. The Wireless Sensor Networks (WSN) are recognized as a very promising competency in the IoT paradigm. Sensor nodes present various challenges in view of energy resources, transmission and processing resources limitation. The energy efficiency is the major issue in WSNs research due to the increase of WSNs applications complexity and the difficult access to the WSN nodes which are frequently deployed in harsh environments. Therefore, several techniques have been developed in order to optimize the energy consumption.

This thesis focuses on proposing new solutions that guarantee the energy efficiency of WSN while preserving the accuracy of the collected data. Firstly, we propose a new Two-Level Clustering method (TLC) based on Position, Data Correlation and nodes Residual Energy. The simulation results proved that the proposed scheme achieves a performance enhancement in terms of energy consumption and network lifetime compared to conventional existing clustering methods. Meanwhile, within the WSN integration with the new IoT technology, the big data issue is provoked, and thus a smart decision making is required. For this, in the second contribution, we introduce Fuzzy Logic advanced intelligent technique which is based on heuristic knowledge and human reasoning. This technique is one of the best problem-solving approach allowing a lower latency and longer lifetime of the network. The proposed clustering process is optimized dynamically. Next, we emphasize the importance of joint clustering and routing to solve energy efficiency problems in WSN by proposing a new joint Clustering and Routing Multi-Objective optimization based on Genetic evolutionary Algorithm (CRMOGA). The optimal solution consists principally in reducing energy consumed for both clustering and routing process. Finally, we propose to take advantage of spatio-temporal correlation of collected data. Then, we propose a Spatio-Temporal Adaptive Sampling approach aware of nodes Residual Energy (STASRE) that adapts the sampling rate dynamically. A reconstruction method based on linear regression is performed to reconstruct missing data. The simulation results prove its efficiency in minimizing energy consumption and maximizing the network lifetime while satisfactory data quality is obtained as witnessed by recovered data at the gateway.

Keywords

IoT, WSN, Energy Efficiency, Clustering, Routing, Data Aggregation, Adaptive Sampling, Spatio-Temporal Correlation, Residual Energy, Fuzzy Logic, Genetic Algorithm, Data Reconstruction.