Published Papers
NanoMST: A Hardware-Aware Multiscale Transformer Network for TinyML-Based Real-Time Inertial Motion Tracking
O. Tariq, D. Han
IEEE Internet of Things (IoT) Journal, 2025
This paper introduces an ultra-lightweight multiscale transformer architecture designed specifically for TinyML applications in inertial motion tracking.
DeepILS: Towards Accurate Domain Invariant AIoT-enabled Inertial Localization System
O. Tariq, B. Dastagir, M. Bilal and D. Han
IEEE Internet of Things Journal, 2025
This paper presents a domain-invariant inertial localization system that maintains accuracy across diverse environments without requiring environment-specific training.
Meta-Swin: Lightweight Image Super-Resolution Swin Transformer for Metaverse Applications
O. Tariq, Dastagir M.B.A., M. Bilal, and D. Han
2nd International Conference on Intelligent Metaverse Technologies & Applications (iMETA), 2024
This paper introduces a lightweight Swin Transformer for image super-resolution designed specifically for metaverse applications with constrained computational resources.
EmoHEAL: A Fusion-Based Framework for Emotion Recognition Using Wearable Sensors
O. Tariq, Y. Oh, and D. Han
2024 IEEE SENSORS, Kobe, Japan, 2024
This paper presents a multimodal fusion framework for emotion recognition using data from wearable sensors, with applications in health monitoring and affective computing.
FedNav: A Federated Learning Approach for Secure AIoT-enabled Inertial Odometry
O. Tariq, M. Bilal, and D. Han
Proc. IEEE Annual Congress on AIoT, Melbourne, Australia, 2024
This paper introduces a federated learning framework for inertial odometry that preserves user privacy while enabling collaborative model training for improved navigation accuracy.
Compact WalshHadamard Transform-Driven S-Box Design for ASIC Implementations
O. Tariq, Dastagir M.B.A., and D. Han
Electronics, vol. 13, no. 16, 2024
This paper presents a novel approach to S-Box design using WalshHadamard transforms, optimized for area-efficient ASIC implementations with strong cryptographic properties.
DeepIOD: Towards A Context-Aware IndoorOutdoor Detection Framework Using Smartphone Sensors
O. Tariq, Dastagir M.B.A., and D. Han
Sensors, vol. 24, no. 16, 2024
This paper presents a context-aware framework for indoor-outdoor detection using smartphone sensors, enabling adaptive behavior in location-based applications.
TabCLR: Contrastive Learning Representation of Tabular Data Classification for Indoor-Outdoor Detection
M.B.A. Dastagir, O. Tariq, and D. Han
IEEE Access, 2024
This paper introduces a contrastive learning approach for tabular data to improve indoor-outdoor detection accuracy using smartphone sensor data.
2D Particle Filter Accelerator for Mobile Robot Indoor Localization and Pose Estimation
O. Tariq, and D. Han
IEEE Access, vol. 12, 2024
This paper presents a real-time FPGA-based accelerator for particle filter SLAM, enabling efficient indoor localization and pose estimation for mobile robots.
A Smart Card-Based Approach for Privacy-Preserving Authentication of Non-Fungible Tokens Using Non-Interactive Zero Knowledge Proof
M.B.A. Dastagir, O. Tariq, and D. Han
Proc. SmartWorld, UIC, ScalCom, DigitalTwin, PriComp, Meta, Haikou, China, 2022
This paper proposes a privacy-preserving authentication scheme for NFTs using smart cards and zero-knowledge proofs to enable secure verification without revealing sensitive information.
HILO: High-Level and Low-Level Co-Design, Evaluation, and Acceleration of Feature Extraction for Visual-SLAM Using PYNQ Z1 Board
M.B.A. Dastagir, O. Tariq, and D. Han
Proc. IPIN-WiP, Beijing, China, 2022
This paper presents a co-design approach for accelerating visual SLAM feature extraction on FPGA platforms, optimizing for both high-level algorithm performance and low-level hardware efficiency.
Papers Under Review
Quantum-Inspired Reinforcement Learning for Secure and Sustainable AIoT-Driven Supply Chain Systems
Dastagir, M. B. A., O. Tariq, Al-Kuwari, S., & Farouk, A.
IEEE Internet of Things Journal, 2025 (Under Review)
This paper proposes a quantum-inspired reinforcement learning approach to enhance security and sustainability in AIoT-based supply chain systems.
Quantum Latent Generative Modeling Learning for Intelligent Transportation System
Dastagir, M. B. A., O. Tariq, Al-Kuwari, S., & Farouk, A.
IEEE Intelligent Transportation Systems Transactions, 2025 (Under Review)
This paper presents a quantum-enhanced generative modeling approach for intelligent transportation systems to improve traffic pattern prediction and management.
ConvXformer: Differentially Private Hybrid ConvNeXt-Transformer for Inertial Navigation
O. Tariq, M. Bilal, MU Hassan, D. Han, J. Crowcroft
IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2025 (Under Review)
This paper presents a differentially private hybrid architecture combining ConvNeXt and Transformer models for privacy-preserving inertial navigation.
TRIP: Towards Realtime Deep Inertial Odometry for Pose-Invariant Pedestrian Navigation
O. Tariq, Y. Oh, and D. Han
IEEE Transactions on Consumer Electronics, 2025 (Under Review)
This paper introduces a real-time deep learning framework for inertial odometry that maintains accuracy regardless of device orientation, improving pedestrian navigation in challenging environments.