Research Activity
My research focuses are deep learning, wireless networking, reinforcement learning, natural laguage processing, large language models, computer vision, image processing and artificial intelligence.
Publications
"Federated Deep Reinforcement Learning for Joint Coverage and Connectivity in Cooperative UAV Networks for Emergency Communications."
Shafkat Khan Siam, Muhammad Yeasir Arafat, Krishnendu Guha, Zilong Liu, Enric Pardo, Hyundong Shin, Md. Noor-A-Rahim. IEEE IoT, 2026.
In disaster-affected areas where conventional communication infrastructure is severely damaged, unmanned aerial vehicles (UAVs) can serve as critical aerial platforms for providing emergency coverage to ground points of interest. This paper proposes FedJCC, a federated deep reinforcement learning framework for joint coverage and connectivity optimization in cooperative UAV networks for emergency communications. The problem is formulated as a constrained Markov decision process, and a hierarchical UAV architecture is adopted, in which cluster heads maintain communication relaying while cluster members provide ground coverage. FedJCC is built on a dueling double deep Q-network (D3QN) with prioritized experience replay, where each UAV is trained locally and model parameters are periodically aggregated through federated averaging. A virtual force-based collision avoidance mechanism derived from the improved adaptive artificial potential field method ensures safe navigation, and a dynamic deployment adjustment mechanism reduces the number of active UAVs while preserving coverage performance. Simulation results show that FedJCC achieves a coverage ratio of 98.8%, substantially outperforming state-of-the-art combinatorial optimization baselines, with improved packet delivery ratio and average end-to-end delay across different scenarios. The coverage-connectivity tradeoff analysis confirms that FedJCC operates closest to the ideal region in which both objectives are jointly maximized.
"AI-Driven Multi-Tier Aerial Communication Networks: A Review of Routing, Computing, Handover, Resource Management, and Optimization Techniques."
Shafkat Khan Siam, Muhammad Yeasir Arafat, Muhammad Morshed Alam, Xu Zhu, Ping Yang, Zilong Liu, Krishnendu Guha, Md. Noor-A-Rahim. Springer Nature, Artificial Intelligence Review, 2026.
Multi-tier aerial communication networks (MACNs), integrating satellites, high-altitude platforms, and unmanned aerial vehicles, are emerging as a cornerstone of next-generation global connectivity. Their promise of resilient and ubiquitous coverage, however, is hindered by highly dynamic topologies, severe energy and computational constraints, environment-sensitive channels, diverse quality-of-service requirements, and limited real-world validation. Artificial intelligence (AI) has increasingly been explored as a flexible framework to address these challenges, enabling adaptive routing, distributed computing and task offloading, handover management, intelligent resource allocation, and large-scale network optimization. This survey provides a comprehensive and structured review of methods for MACNs, with particular emphasis on AI-driven solutions and their relationship to classical and hybrid alternatives. We critically evaluate representative approaches in terms of scalability, efficiency, data demands, and practical deployability, and identify emerging trends such as graph neural networks with reinforcement learning for dynamic routing, predictive learning for mobility management, and federated learning for distributed computation. Persistent challenges remain in lightweight edge intelligence, real-world testbeds, reproducible benchmarking, and simulation-to-deployment transfer. To address these issues, we offer a research roadmap emphasizing compressible and interpretable models, standardized benchmarks, realistic validation, and hybrid designs that balance adaptability with computational and energy overhead. Finally, we identify open challenges and future research directions, offering insights into the design of AI-driven MACNs that are efficient, scalable, and adaptive to evolving network and service demands.
"Rethinking Gradient Weight’s Influence over Saliency Map Estimation."
Masud An Nur Fahim, Nazmus Saqib, Shafkat Khan Siam, Ho Yub Jung. MDPI, Sensors, 22 (17), 6516, 2022.
Class activation map (CAM) helps to formulate saliency maps that aid in interpreting the deep neural network’s prediction. Gradient-based methods are generally faster than other branches of vision interpretability and independent of human guidance. The performance of CAM-like studies depends on the governing model’s layer response and the influences of the gradients. Typical gradient-oriented CAM studies rely on weighted aggregation for saliency map estimation by projecting the gradient maps into single-weight values, which may lead to an over-generalized saliency map. To address this issue, we use a global guidance map to rectify the weighted aggregation operation during saliency estimation, where resultant interpretations are comparatively cleaner and instance-specific. We obtain the global guidance map by performing elementwise multiplication between the feature maps and their corresponding gradient maps. To validate our study, we compare the proposed study with nine different saliency visualizers. In addition, we use seven commonly used evaluation metrics for quantitative comparison. The proposed scheme achieves significant improvement over the test images from the ImageNet, MS-COCO 14, and PASCAL VOC 2012 datasets.
"Denoising Single Images by Feature Ensemble Revisited."
Masud An Nur Fahim, Nazmus Saqib, Shafkat Khan Siam, Ho Yub Jung. MDPI, Sensors, 22 (18), 7080, 2022.
Image denoising is still a challenging issue in many computer vision subdomains. Recent studies have shown that significant improvements are possible in a supervised setting. However, a few challenges, such as spatial fidelity and cartoon-like smoothing, remain unresolved or decisively overlooked. Our study proposes a simple yet efficient architecture for the denoising problem that addresses the aforementioned issues. The proposed architecture revisits the concept of modular concatenation instead of long and deeper cascaded connections, to recover a cleaner approximation of the given image. We find that different modules can capture versatile representations, and a concatenated representation creates a richer subspace for low-level image restoration. The proposed architecture’s number of parameters remains smaller than in most of the previous networks and still achieves significant improvements over the current state-of-the-art networks.
Projects
Federated deep reinforcement learning for cooperative UAV networks:
This is the first chapter of my PhD research at University College Cork. I developed FedJCC, a federated deep reinforcement learning framework in which a hierarchical UAV swarm jointly optimizes ground coverage and relay connectivity for post-disaster emergency communications. Each UAV trains a dueling double deep Q-network locally with prioritized experience replay, and the swarm collaborates by sharing only model parameters through federated averaging. The framework also includes virtual force-based collision avoidance and a dynamic deployment mechanism that reduces the active swarm size while preserving coverage. This work has been published in IEEE IoT.
AI-driven multi-tier aerial communication networks:
This is my ongoing PhD research at University College Cork on applying AI to multi-tier aerial networks that integrate satellites, high-altitude platforms, and UAVs. As part of this work, I led a comprehensive survey of AI-driven routing, edge computing, handover, resource management, and optimization techniques for these networks, published in Artificial Intelligence Review (Springer Nature).
Image denoising:
This project focused on using self-supervised learning and ensemble methods
for image denoising. I developed an aggregated multiscale self-supervised denoising model,
which leverages multiple features from different scales to enhance the image quality. I have
done my Master’s thesis on this topic. The thesis can be download from here.
Also the code and main idea for my thesis in details provided here.
I also co-authored a paper on this topic, where image denoising have been done with a supervised
training method. We have developed a custom network where shallow features are extracted and ensambled
to remove noise from images. The paper is titled “Denoising Single Images by Feature Ensemble
Revisited.”
Explainable AI:
This project aimed to understand what a deep learning model sees when it classifies a specific class, an area of research known as explainable AI. I worked on developing a method to estimate the gradient weights of the saliency maps, which are used to visualize the regions of interest in an image. I also co-authored a paper on this topic, titled “Rethinking Gradient Weight’s Influence over Saliency Map Estimation.”
FRVT 1:1 face recognition:
This project involved developing advanced face recognition algorithms using deep learning and computer vision techniques. I was part of the team that implemented and evaluated different models and achieved state-of-the-art results on the FRVT 1:1 benchmark.