Research

Learning-based optimization of intelligent wireless networks

My research applies deep reinforcement learning to the design and optimization of next-generation wireless networks. At the Wireless Intelligent Networks (WIN) Lab, I build ns-3 simulation environments to train and evaluate RL agents — such as Deep Q-Networks (DQN) — for Wi-Fi 7 multi-link operation, and I study how learning-based methods can optimize emerging 6G technologies.

More broadly, I'm interested in the intersection of reinforcement learning, networking, and applied machine learning: how adaptive, data-driven agents can manage the growing complexity of modern communication systems while remaining reproducible and reliable.

Research Interests

Reinforcement Learning for Networks

Designing and training DQN agents for intelligent WiFi and cellular network optimization.

Wi-Fi 7 & Multi-Link Operation

Application-aware, learning-based uplink traffic allocation across MLO links.

6G Technologies

Optimizing Fluid Antenna, Pinching Antenna, RSMA, and NOMA in cellular systems.

Network Simulation

Building reproducible ns-3 experimentation environments for RL research.

Applied Machine Learning

Predictive modeling and production ML pipelines on large real-world datasets.

Optimization

Multi-objective ("quadrilemma") optimization for resource-constrained systems.

Publications

I publish under the name Duy P. Ngo. See my Google Scholar profile.

  • 2026

    Application-Aware Learning-Based Uplink Traffic Allocation for Wi-Fi 7 MLO

    Brian Trinh, Samarth Kulkarni, Duy P. Ngo, Krishna Murthy Kattiyan Ramamoorthy

    IEEE 32nd International Symposium on Local and Metropolitan Area Networks (LANMAN), 2026