-
SAD-LoRA: Spectral Alignment for Low-Rank Knowledge Distillation
2026
O. Tariq, S. M. Raza, J. Son · ICML 2026 ColorAI Workshop, Seoul, South Korea
Aligns the LoRA adapter subspace with the teacher's data-weighted spectral subspace during knowledge distillation, improving rank efficiency with zero inference overhead.
Poster
Poster
-
Uncertainty Aware and Decoder Aligned Learning for Video Summarization
2026
O. Tariq, M. Raza, S. Jeongbae · 2026 Int. Joint Conf. on Neural Networks (IJCNN), Maastricht, Netherlands
Uncertainty-aware framework with decoder-aligned learning for robust video summarization.
Accepted
-
NOS-Gate: Queue-Aware Streaming IDS for Consumer Gateways under Timing-Controlled Evasion
2026
M. Bilal, O. Tariq, H. Ahmed · IEEE Transactions on Consumer Electronics
Queue-aware streaming intrusion detection system robust to timing-controlled evasion attacks on consumer gateways.
-
ConvXformer: Differentially Private Hybrid ConvNeXt-Transformer for Inertial Navigation
2026
O. Tariq, M. Bilal, M. U. Hassan, D. Han, J. Crowcroft · IEEE Trans. Systems, Man, and Cybernetics: Systems, pp. 1–14
Differentially private hybrid ConvNeXt-Transformer architecture providing formal ε-DP guarantees for distributed multimodal sensor fusion without sacrificing localization accuracy.
Early Access
-
ADP-QFed: Privacy-Preserving Quantized Federated Learning for Intelligent Edge Sensing in IoT Systems
2025
O. Tariq, M. Bilal, D. Han · IEEE Internet of Things Journal
Combines adaptive differential privacy with quantized federated learning for privacy-preserving collaborative edge intelligence in IoT systems.
IF 8.9
-
Quantum-Inspired Reinforcement Learning for Secure and Sustainable AIoT-Driven Supply Chain Systems
2025
M. B. A. Dastagir, O. Tariq, S. Al-Kuwari, A. Farouk · IEEE Internet of Things Journal
Quantum-inspired reinforcement learning for secure and sustainable AIoT-driven supply chain optimization.
IF 8.9
-
NanoMST: A Hardware-Aware Multiscale Transformer Network for TinyML-Based Real-Time Inertial Motion Tracking
2025
O. Tariq, D. Han · IEEE Internet of Things Journal
Ultra-lightweight multiscale transformer (298K params, 8-bit quantization) for TinyML inertial motion tracking — 4.7× compute reduction over LSTM baselines with real-time edge performance.
IF 8.9
-
DeepILS: Towards Accurate Domain Invariant AIoT-Enabled Inertial Localization System
2025
O. Tariq, B. Dastagir, M. Bilal, D. Han · IEEE Internet of Things Journal
Domain-invariant AIoT-enabled inertial localization achieving sub-meter accuracy across diverse environments without environment-specific retraining.
IF 8.9
-
Meta-Swin: Lightweight Image Super-Resolution Swin Transformer for Metaverse Applications
2024
O. Tariq, M. B. A. Dastagir, M. Bilal, D. Han · 2nd Int. Conf. on Intelligent Metaverse Technologies & Applications (iMETA), Dubai, UAE
Lightweight Swin Transformer for image super-resolution designed for resource-constrained metaverse applications.
-
EmoHEAL: A Fusion-Based Framework for Emotion Recognition Using Wearable Sensors
2024
O. Tariq, Y. Oh, D. Han · 2024 IEEE SENSORS, Kobe, Japan
Multimodal fusion framework for emotion recognition using wearable sensor data with health monitoring applications.
-
FedNav: A Federated Learning Approach for Secure AIoT-Enabled Inertial Odometry
2024
O. Tariq, M. Bilal, D. Han · IEEE AIoT Congress, Melbourne, Australia
Federated learning framework for privacy-preserving collaborative inertial odometry training across distributed IoT devices.
-
2D Particle Filter Accelerator for Mobile Robot Indoor Localization and Pose Estimation
2024
O. Tariq, D. Han · IEEE Access, vol. 12, pp. 18473–18487
Real-time FPGA-based particle filter SLAM accelerator for mobile robot indoor localization — 60% latency reduction over software baseline at 30 FPS.
IF 3.6
-
TabCLR: Contrastive Learning Representation of Tabular Data Classification for Indoor-Outdoor Detection
2024
M. B. A. Dastagir, O. Tariq, D. Han · IEEE Access
Contrastive learning approach for tabular sensor data to improve indoor-outdoor detection classification accuracy.
IF 3.6
-
Compact Walsh–Hadamard Transform-Driven S-Box Design for ASIC Implementations
2024
O. Tariq, M. B. A. Dastagir, D. Han · Electronics, vol. 13, no. 16
Area-efficient S-Box design for ASIC using Walsh-Hadamard transforms with strong cryptographic properties.
IF 2.6
-
DeepIOD: Context-Aware Indoor–Outdoor Detection Framework Using Smartphone Sensors
2024
O. Tariq, M. B. A. Dastagir, D. Han · Sensors, vol. 24, no. 16
Context-aware indoor-outdoor detection using smartphone sensors for adaptive location-based application behavior.
IF 3.5
-
Smart Card-Based Privacy-Preserving NFT Authentication via Non-Interactive Zero-Knowledge Proof
2022
M. B. A. Dastagir, O. Tariq, D. Han · Proc. SmartWorld/UIC/ScalCom, Haikou, China
Privacy-preserving NFT authentication using smart cards and zero-knowledge proofs.
-
HILO: High-/Low-Level Co-Design for Visual-SLAM Feature Extraction on FPGA
2022
M. B. A. Dastagir, O. Tariq, D. Han · Proc. IPIN-WiP, Beijing, China
Co-design approach for accelerating visual SLAM feature extraction on FPGA, optimizing algorithm performance and hardware efficiency jointly.