• Tentative Title: 6G Channel Estimation using Machine Learning & Deep Learning Techniques

(Ongoing M.Sc. Thesis)

Supervisor: Dr. Md. Forkan Uddin, Professor, Department of EEE, BUET

The primary objective of this research work is to investigate the challenges in deploying 6G networks with a focus on reliable channel estimation, and apply Machine Learning and Deep Learning techniques for channel estimation in 6G networks.

• Title: BFMD - A Novel Fingermark Dataset to Evaluate The Performance of Algorithms for Forensic Investigation (Undergraduate Thesis)

Supervisor: Dr. S.M. Mahbubur Rahman, Professor, Department of EEE, BUET

The primary objective of this research work was to build a novel fingermark dataset from different subjects and evaluate the performance of an available fingerprint algorithm on the fingermark dataset.

● Created a unique fingermark dataset comprising samples from different subjects

● Evaluated the accuracy of an available fingerprint algorithm through performance analysis on the dataset

• Title: Privacy-Aware Human Activity Recognition from a Wearable Camera

Supervisor: Dr. Mohammad Ariful Haque, Professor, Department of EEE, BUET

The primary objective of this research work was developing algorithms capable of recognizing activities from videos captured by chest mounted cameras. There were several subjects doing several types of work in their office which were captured using chest mounted cameras. We developed an algorithm based on a deep neural network that can classify these activities with great accuracy.

● Employed a comprehensive approach combining CNN-based feature extraction, MLPs, SVM classification, and YOLOv3 for object detection

● Integrated template matching and blurring techniques for privacy preservation