Defence of Doctoral Thesis in Electronics with Mazhar Hussain

Thu 03 Apr 2025 10.00–14.00
Sundsvall
Room C306, C-building
Lägg till i din kalender

Welcome to a public defense of doctoral thesis in Electronics with Mazhar Hussain who will present and defend his thesis "Multi-Sensor Data Fusion for Improved Estimation and Prediction of Physical Quantities".

In recent years, the integration of multi-sensor data has become essential across various fields, from smart cities and healthcare to robotics and environmental monitoring. This research explores data fusion techniques - ranging from traditional analytics to machine learning and deep learning, to enhance decision-making and predictive modeling. This thesis focus on three case studies include analyzing molten glass viscosity, predicting city bus fuel consumption, and classifying hazardous gases.

Welcome to listen when Mazhar Hussain presents his work on multi-sensor data fusion and its impact on real-world applications.

 

Date: April 3rd, 2025

Time: 10:00 CET

Place: Campus Sundsvall, Room C306, C-building, Youtube and Zoom.

Doctoral thesis: Multi-Sensor Data Fusion for Improved Estimation and Prediction of Physical Quantities

Respondent: Mazhar Hussain

Supervisor and Chair: Professor Jan Lundgren, Mittuniversitetet

Co-supervisor: Professor Mattias O'Nils, Mittuniversitetet 

Opponent: Professor Niclas Björsell, Högskolan i Gävle

Examining committee: 

Professor Annalisa Liccardo, Universita Degli Studi di Napoli Federico II, Italien

Professor Fredrik Gustafsson, Linköpings Universitet

Professor Tomas Nordström, Umeå Universitet

Backup: Associate Prof. Göran Thungström, Mittuniversitetet

Join the seminar on Zoom

Abstract

In recent years, there has been a significant increase in multi-sensor data across various fields, spanning from environmental monitoring and industrial automation to smart agriculture, surveillance systems, healthcare analytics, robotics, remote sensing, smart cities, and beyond. The fundamental drive behind leveraging multimodal data is the amalgamation of complementary information extracted from various sensors, facilitating more comprehensive insights and informed decision-making compared to reliance on a single modality.

The analysis of multi-sensor data presents substantial challenges due to its vastness and the presence of structured, semi-structured, and unstructured data, spanning different modalities with distinct sources, types, and distributions. Data fusion, the integration of information from diverse modalities, becomes crucial in addressing inference problems arising from multi-sensor data. Both analytics-based and learning-based data fusion approaches are widely used, with learning-based approaches, leveraging machine learning and deep learning methods, showing notable effectiveness.

However, the question of "where" and "how" to fuse different modalities remains an open challenge. To explore this, the study focused on three applications as case studies to employ data fusion approaches for estimating and predicting physical quantities. These applications include analysing the correlation between the change in the geometrical dimension of a free-falling molten glass gob and its viscosity using Pearson correlation coefficient as analytical method, predicting fuel consumption of city buses through machine learning methods, and classifying and measuring hazardous gases i.e. hydrogen sulfide (H2S) and methyl mercaptan (CH3SH) using deep learning methods.

Results from these case studies indicate that the choice between traditional, machine learning, or deep learning-based data fusion depends on the specific application, as well as the size and quality of the data. Despite this, advancements in computing power and deep learning technology havemade data more accessible and have enhanced its complementarity. Therefore, a comprehensive review to compare a range of deep learning-based data fusion strategies is conducted. The review provides an examination of various feature extraction methods, as well as an outline and identification of the research fields that stand to derive the greatest benefits from these evolving approaches.

 

Read the thesis in Diva

The page was updated 3/11/2025