Junjie Wu
AI for Next-Generation Wi-Fi
About Me
I am a postdoctoral researcher at Southwest Jiaotong University. My research focuses on AI for wireless networks, next-generation Wi-Fi optimization, multi-agent reinforcement learning, federated learning, and generative AI for resource allocation.
My work studies how learning-based methods can make future wireless systems more adaptive, efficient, and intent-aware. I am broadly interested in robust AI-driven optimization for networked systems and welcome academic collaboration.
News
One paper on topology-aware dynamic offloading for MEC-enabled MAPC Wi-Fi networks was accepted by IEEE Transactions on Vehicular Technology.
One paper on energy-constrained multi-UAV networks appeared in IEEE Transactions on Wireless Communications.
Mirage was released as an arXiv preprint.
Two papers on AI-driven Wi-Fi resource allocation appeared in IEEE Internet of Things Journal.
Our work on multi-link frame aggregation optimization was published in IEEE Transactions on Wireless Communications.
Selected Publications
Full listTopology-aware Dynamic Offloading for MEC-enabled MAPC Wi-Fi Networks: A Multi-Agent Hierarchical DRL Framework
IEEE Transactions on Vehicular Technology, 2026
[TL;DR]
A topology-aware multi-agent hierarchical DRL framework for dynamic offloading in MEC-enabled MAPC Wi-Fi networks.
# MEC# MAPC Wi-Fi# Hierarchical DRL
Mirage: Transmitting a Video as a Perceptual Illusion for 50,000 X Speedup
arXiv preprint arXiv:2602.07396, 2026
[TL;DR]
A perceptual-illusion based video transmission idea targeting large speedups.
# Video Transmission# Perceptual Systems
Enhancing A2G Robustness in Energy-Constrained Multi-UAV Networks: MADRL for Trajectory Control and Resource Allocation
IEEE Transactions on Wireless Communications, 2026 Citations: 2
[TL;DR]
MADRL is used for robust trajectory control and resource allocation in energy-constrained multi-UAV networks.
# Multi-UAV# MADRL# A2G Networks
Trajectory Design and Beamforming in UAV-Assisted Wireless Networks: A Fine-Tuned M2LLM-Driven DRL-Based Framework
IEEE Transactions on Wireless Communications, 2025 Citations: 5
[TL;DR]
A fine-tuned M2LLM-driven DRL framework supports trajectory design and beamforming in UAV-assisted wireless networks.
# UAV Networks# DRL# Beamforming
DRL Optimization Trajectory Generation via Wireless Network Intent-Guided Diffusion Models for Resource Allocation
IEEE Internet of Things Journal, 2025 Citations: 10
[TL;DR]
Intent-guided diffusion models generate optimization trajectories for DRL-based wireless resource allocation.
# Diffusion Models# Resource Allocation# DRL
Service-Differentiated Joint Distributed Communication and Computing Resource Allocation for Wi-Fi Networks based on Federated Learning and MADRL
IEEE Internet of Things Journal, 2025 Citations: 9
[TL;DR]
Federated learning and MADRL support service-differentiated distributed resource allocation in Wi-Fi networks.
# Wi-Fi Networks# Federated Learning# MADRL
Deep Reinforcement Learning based Multi-link Frame Aggregation Length Optimization in Next Generation Wi-Fi Networks
IEEE Transactions on Wireless Communications, 2024 Citations: 19
[TL;DR]
DRL optimizes multi-link frame aggregation length for next-generation Wi-Fi networks.
# Next-Generation Wi-Fi# Multi-Link# Reinforcement Learning
Optical flow filtering-based micro-expression recognition method
Electronics 9(12), 2056, 2020 Citations: 17
[TL;DR]
An optical-flow filtering method is designed for micro-expression recognition.
# Computer Vision# Micro-Expression Recognition
Research Projects
I have participated in Huawei Innovation Research Program projects on Wi-Fi multi-band and multi-link collaborative transmission, key MAC layer technologies for next-generation Wi-Fi, Wi-Fi sensing standards, and AI-driven Wi-Fi optimization.