Open to research collaboration

Md. Sazedur Rahman

PhD researcher at Missouri S&T building secure and trustworthy federated learning systems for underground mines and other safety-critical environments.

Md. Sazedur Rahman
Current role Graduate Research Assistant
W2C Lab, Missouri S&T
About me

Making distributed intelligence safer when data and devices cannot be fully trusted.

My research focuses on securing federated learning against malicious and unreliable clients in spatially distributed underground mining environments.

I develop trust-aware aggregation, poisoning-attack detection, and robust learning methods that accommodate non-IID data, intermittent connectivity, low illumination, dust, occlusion, and other conditions found in real mines. I also work on computer vision, anomaly detection, and trustworthy AI for cyber-physical systems.

Federated Learning Cybersecurity Trustworthy AI Computer Vision Autonomous Mining Robust Aggregation
Research directions

Secure distributed intelligence for challenging environments.

My work connects algorithmic robustness with the operational realities of autonomous underground mining and connected cyber-physical systems.

01

Secure Federated Learning

Trust-weighted aggregation and detection methods for label-flip, sign-flip, additive-noise, intermittent, and unreliable-client behaviors.

Related projects
02

AI for Underground Mining

Safety-sign recognition, situational awareness, and robust perception using MineSigns and other real and synthetic mining datasets.

Related projects
03

Trustworthy Cyber-Physical Systems

Anomaly detection, malicious-sensor identification, privacy, and resilient collaboration across vehicles, robots, and IoT devices.

Related projects

Projects

01

TrustFed-LFD

A federated learning defense that limits the influence of client updates associated with label flipping.

02

MineSigns Dataset

Video collected in an underground mine and annotated across 13 safety-sign classes for perception research.

Publication record

Publications.

Selected peer-reviewed work in federated learning security, intelligent mining, and connected systems.

2026
ACM DEBS 2026

Label-Flip Attack Detection via Trust-Weighted Aggregation in Federated Learning for Underground Mine Security

Md. Sazedur Rahman, Sanjay Madria, Samuel Frimpong

2026
COMSNETS 2026

Future Mining: Learning for Safety and Security

Md. Sazedur Rahman, M. R. Jewel, Sanjay Madria

2025
IEEE MASS 2025

Detecting Untargeted Attacks and Mitigating Unreliable Updates in Federated Learning for Underground Mining Operations

Md. Sazedur Rahman, Mohamed Elmahallawy, Sanjay Madria, Samuel Frimpong

2024
International Conference on Computing and Communication Networks · pp. 535–547 · 2024

Detection of Coal Miner with a Comprehensive Dataset Using Transfer Learning Techniques

Md. Sazedur Rahman, K. Hoque, M. B. Hossain, D. Das, T. Wu

2024
IEEE 21st International Conference on Mobile Ad-Hoc and Smart Systems (MASS) · 2024

CAV-AD: A Robust Framework for Detection of Anomalous Data and Malicious Sensors in CAV Networks

Md. Sazedur Rahman, Mohamed Elmahallawy, Sanjay Madria, Samuel Frimpong

2024
5th International Conference on Data Analytics & Management (ICDAM) · 2024

Enhancing Multi-Class Skin Lesion Classification: An Ensemble-Based Transfer Learning Approach

SMMH Md Manirujjaman, Mahmudul Hoque, Md. Sazedur Rahman, Md. Zahim Hassan, et al.

2024
2nd International Conference on Big Data, IoT and Machine Learning · 2024

Computing Skyline Query on Incomplete Data

Md. Sazedur Rahman, KMA Hasan

2023
International Conference on Innovations in Bio-Inspired Computing and Applications · 2023

Vehicle-BD: A Benchmark Dataset of Bangladeshi Local Vehicles

Md. Sazedur Rahman, Md. Zahim Hassan, S. Ibtisum

2022
4th International Conference on Sustainable Technologies for Industry 4.0 · 2022

Bangladeshi Local Vehicle Recognition with a Comprehensive Dataset Using Transfer Learning Techniques

Md. Sazedur Rahman, M. Z. Hassan, S. N. Hossain, N. Masrur, J. Rabbi

Latest updates

News and milestones.

2026

Label-flip defense paper accepted at the 20th ACM DEBS.

2026

Future Mining paper accepted at COMSNETS 2026.

October 2025

Presented federated-learning security research at IEEE MASS in Chicago.

Academic CV

PhD student in Computer Science, Missouri S&T · Graduate Research Assistant, W2C Lab

Download CV (PDF)
Let’s connect

Interested in secure federated learning or intelligent mining systems?

I welcome research conversations, collaborations, and academic opportunities.

mrvfw@umsystem.edu