Real networks. Real environments. Real Intelligence.
Future wireless networks will need more than pattern matching. They will need AI systems that can model changing network conditions, anticipate the effects of control actions, and reason across traffic, topology, energy, and mobility.
Researching AI-native wireless networks through representation learning, world models, foundation models, reinforcement learning, and sustainable AI.
The goal is to move beyond text-centric AI toward systems that can support planning, automation, and sustainability in the dynamic, partially observable, continuous-state environments that define wireless networks.
Research Areas
Wireless World Models and Foundation Models
Developing shared latent models for wireless networks and smart farms that support prediction, simulation, planning, and autonomous control.
Graph Intelligence for Networked Physical Systems
Using graph neural networks to model topology, interference, sensing, and resource dependencies across interconnected wireless systems.
Predictive Orchestration and Adaptive Forecasting
Connecting forecasting directly to operational decisions for proactive orchestration of radio, compute, energy, irrigation, and sensing resources.
Sustainable AI and Energy-Intelligent Infrastructure
Designing AI methods that reduce energy waste while improving resilience through anomaly detection, thermal optimisation, and efficient edge intelligence.
AI-Native Wireless Agriculture, SDR, and Open RAN
Building intelligent agricultural systems with wireless sensing, drone data, SDR, and Open RAN for affordable, sustainable, and AI-native rural connectivity.
Autonomous Robotics for Field and Site Operations
Developing autonomous systems for inspection, monitoring, navigation, and physical task automation in mobile sites and agricultural environments.
Affiliations
Research collaborations and institutional ties across wireless systems, machine intelligence, and networked infrastructure.
