Research

My research focuses on the design of intelligent, reliable, and resource-efficient communication systems for emerging 5G, 6G, Internet of Things, and cyber-physical applications. The central objective is to develop communication and decision-making frameworks that can operate effectively under practical constraints involving latency, reliability, mobility, energy consumption, spectral efficiency, security, and distributed operation. My work integrates wireless communications, constrained optimization, reinforcement learning, federated learning, stochastic modelling, edge intelligence, and system-level simulation. Particular emphasis is placed on methods that remain feasible and dependable under realistic operating conditions rather than achieving high performance only under idealized assumptions.

 

Research Areas

 

AI-Enabled Wireless Networks

 

Future wireless networks must adapt continuously to dynamic traffic, channel conditions, mobility patterns, service demands, and network congestion. My research develops intelligent resource-management frameworks that combine wireless communication principles with reinforcement learning, optimization, and adaptive decision-making. Current topics include spectrum management, mobility management, handover optimization, interference control, energy-aware resource allocation, multi-agent reinforcement learning, constrained reinforcement learning, and intent-aware network orchestration. A major focus of this research is the integration of operational constraints directly into the learning process. Instead of optimizing only the average reward, the proposed frameworks consider reliability, latency, energy, risk, feasibility, and service-level requirements as explicit decision variables.

 

Vehicular Internet of Things

Vehicular communication systems require rapid and reliable decision-making under highly dynamic mobility, traffic, and channel conditions. My research in vehicular IoT investigates intelligent resource allocation, V2X communication, roadside-unit scheduling, spectrum access, service differentiation, congestion management, and mobility-aware coordination. The work considers heterogeneous vehicular services with different reliability, latency, packet-delivery, and computational requirements. Reinforcement learning, stochastic optimization, scheduling, and multiobjective decision-making are used to improve service performance while controlling communication overhead and resource consumption. Current research also examines intent-aware and agentic AI frameworks that allow network controllers to adapt their policies when service priorities or operating conditions change.

 

Healthcare Internet of Things

Healthcare IoT applications require communication and control systems with strict requirements for reliability, latency, safety, sensing accuracy, and interpretability. My research investigates intelligent communication and control frameworks for medical monitoring, automated healthcare systems, telesurgery, physiological sensing, and closed-loop therapeutic applications. This work includes reliability-aware spectrum management, ultra-reliable low-latency communication, wireless body-area networks, medical edge computing, and safety-aware control. A related research direction examines digital-twin and nonlinear predictive-control frameworks for automated insulin delivery. These systems combine continuous glucose monitoring, physiological modelling, prediction, constrained optimization, and safety supervision.

 

6G and Beyond Wireless Systems

Beyond-5G and 6G networks are expected to support highly heterogeneous applications across terrestrial, aerial, satellite, and intelligent propagation environments. My research addresses the communication and resource-management challenges associated with these integrated systems. Current topics include millimetre-wave and terahertz communication, massive MIMO, reconfigurable intelligent surfaces, ultra-reliable communication, distributed intelligence, and sustainable network design. The objective is to develop scalable frameworks that improve coverage, spectral efficiency, reliability, mobility support, and energy efficiency across complex network architectures.

 

Reconfigurable Intelligent Surfaces

Reconfigurable intelligent surfaces provide a mechanism for controlling the wireless propagation environment through programmable reflecting elements. My research investigates RIS-assisted communication for coverage enhancement, interference mitigation, mobility management, energy optimization, and spectrum efficiency. The work considers joint optimization of transmission parameters, surface configurations, user association, mobility, and resource allocation. Artificial intelligence and reinforcement learning are explored as tools for managing RIS-assisted networks under dynamic and partially observable conditions.

 

Non-Terrestrial Networks

Non-terrestrial networks integrate satellites, unmanned aerial vehicles, high-altitude platforms, and terrestrial communication infrastructure. These systems can improve connectivity in remote, underserved, mobile, and disaster-affected environments. My research examines resource allocation, mobility management, handover, service continuity, propagation uncertainty, energy consumption, and intelligent coordination across terrestrial and non-terrestrial network segments. Particular attention is given to integrated network architectures in which terrestrial and aerial or satellite systems cooperate to provide reliable and adaptive connectivity.

 

Edge Intelligence and Federated Learning

Centralized artificial intelligence can be limited by communication overhead, privacy concerns, latency, energy consumption, and heterogeneous data. My research studies distributed and federated learning frameworks for intelligent IoT and wireless systems. Current topics include privacy-preserving learning, communication-efficient federated learning, intrusion detection, adaptive client participation, non-independent data distributions, unreliable devices, energy-aware training, and secure distributed intelligence. The interaction between communication conditions and learning performance is a central research question. This includes analysing how bandwidth, latency, packet loss, device availability, and energy constraints affect distributed model training and inference.

 

Wireless Sensor Networks

Wireless sensor networks remain relevant to environmental monitoring, industrial systems, smart infrastructure, healthcare, and resource-constrained IoT applications. My research in this area includes energy-efficient routing, hop-constrained communication, trust-aware routing, clustering, reliability, and reinforcement-learning-based decision-making. A key objective is to develop routing and resource-management methods that account for limited energy, unreliable links, network lifetime, and feasibility constraints.

 

Research Methodology

My research combines analytical, computational, and learning-based methods. Depending on the problem, the methodology may include mathematical modelling, constrained optimization, Markov decision processes, stochastic processes, reinforcement learning, deep learning, federated learning, nonlinear control, and simulation-based evaluation. The research process typically begins with a clearly defined system model and set of operational assumptions. The proposed method is then compared with representative analytical, heuristic, optimization-based, and learning-based baselines. Particular emphasis is placed on reproducibility, sensitivity analysis, ablation studies, statistical validation, transparent reporting of constraints, and alignment between the mathematical formulation, implementation, and reported conclusions.

 

Current Research Initiatives

Intent-Aware Agentic AI for Vehicular Networks

This research develops autonomous network agents that interpret service intents and adapt wireless access and resource-management policies under changing vehicular conditions. The framework considers reliability, latency, congestion, mobility, and service-priority requirements. It combines constrained reinforcement learning, risk awareness, policy adaptation, and mechanism-level evaluation against established scheduling and learning baselines.

 

Reliability-Constrained Resource Management

This research investigates communication and learning frameworks that explicitly incorporate reliability, feasibility, latency, and energy constraints. Applications include vehicular IoT, healthcare IoT, spectrum access, roadside-unit scheduling, wireless sensor-network routing, and multi-service communication systems.

 

Intelligent Healthcare Communication and Control

This research examines the interaction between communication reliability, sensing uncertainty, prediction, and closed-loop control in connected healthcare environments. Current applications include automated insulin delivery, medical IoT, physiological monitoring, telesurgery, and safety-aware networked control.

 

Integrated Terrestrial and Non-Terrestrial 6G Networks

This initiative studies communication across cellular, UAV, satellite, and RIS-assisted environments. The research focuses on mobility, propagation uncertainty, resource allocation, service continuity, energy efficiency, and intelligent coordination across heterogeneous network segments.

 

Secure and Communication-Efficient Federated Intelligence

This research develops federated-learning frameworks that address privacy, device heterogeneity, unreliable participation, communication overhead, and adversarial threats. Applications include intrusion detection, intelligent IoT, distributed healthcare systems, edge networks, and cyber-physical infrastructure.

 

Research Applications

The broader application domains of my research include intelligent transportation, connected healthcare, industrial IoT, smart cities, wireless automation, satellite and aerial communication, public-safety networks, distributed sensing, and sustainable communication infrastructure. These application domains differ in their operational requirements, but they share a common challenge: communication and computational decisions must be made under uncertainty while satisfying stringent system constraints.

 

Research Collaboration

I welcome research collaboration with universities, research laboratories, industry partners, and interdisciplinary teams working in wireless communications, artificial intelligence, vehicular networking, healthcare IoT, 5G and 6G systems, non-terrestrial networks, federated learning, edge intelligence, and optimization. Potential collaboration formats include joint publications, research proposals, graduate co-supervision, visiting research, comparative studies, special issues, workshops, and shared simulation or experimental platforms.