AI-Powered Energy-Efficient WBANs for Real-Time Health Monitoring

About the Project

Summary of the Proposed Research:

Wireless Body Area Networks (WBANs) are central to next-generation healthcare systems, offering continuous real-time monitoring of patient vital signs through wearable sensors. However, current implementations are limited by high energy consumption, dependence on centralised cloud processing, and static anomaly detection algorithms—issues that hinder their scalability in real-world and resource-constrained environments.

This PhD project proposes a novel, AI-powered, energy-efficient WBAN framework that leverages embedded intelligence and federated learning to enable decentralised, low-latency health anomaly detection. The system will support privacy-preserving, adaptive decision-making directly at the network edge, thereby enhancing responsiveness, efficiency, and clinical relevance.

The aim of the project is to design and implement a smart, low-power WBAN system that enhances anomaly detection accuracy, reduces energy consumption, and improves privacy and adaptability in healthcare applications, particularly for continuous monitoring in remote and clinical environments.

The specific objectives of the project are to:

• Develop and implement lightweight anomaly detection models using Tiny Machine Learning (TinyML) for real-time health monitoring.
• Design an adaptive sampling mechanism and dynamic energy management system to minimise power consumption based on patient condition.
• Integrate federated learning to enable decentralised model updates without sharing sensitive patient data, preserving privacy.
• Build and test embedded AI models for WBAN edge devices using platforms such as Raspberry Pi, ESP32, or Arduino.
• Validate the proposed system using real-world physiological datasets and evaluate its performance across metrics such as accuracy, latency, power efficiency, and reliability.
• Explore real-world deployment potential in applications such as hospital monitoring, elderly care, and remote patient diagnostics.

This research will contribute to the development of intelligent, responsive, and practical healthcare solutions by bridging embedded AI, wireless communication, and biomedical sensing technologies.

How to apply:

For further information, please contact: Dr Masum Billah at masum.billah@staffs.ac.uk

The applications should consist of a cover letter or personal statement of interest, and a CV.

Dr Masum Billah

Course Leader – Embedded Electronic Systems Design Development Engineer Apprenticeship

School of Digital, Technology, Innovation & Business

University of Staffordshire

College Road

Stoke-on-Trent

ST4 2DE

The expected start dates are January and April 2026.

Entry Requirements:

Applicants should have a First- or Upper Second-Class UK Honours degree (or equivalent) in a relevant discipline such as Artificial Intelligence, Electronic Engineering, Electrical and Electronic Engineering, Robotics, or Mechatronics. An MSc with Distinction or Merit is highly desirable.

A strong background in AI, embedded systems, or electronic engineering is required, including proficiency in Python and experience with machine learning frameworks such as TensorFlow, PyTorch, or scikit-learn. Familiarity with TinyML frameworks (e.g., TensorFlow Lite, Edge Impulse) and embedded platforms such as Arduino, ESP32, or Raspberry Pi is advantageous. Prior knowledge of wireless sensor networks and physiological data acquisition (e.g., ECG, heart rate, blood pressure) would further strengthen the application.

The standard minimum IELTS Academic requirement is 6.5 overall, with no less than 6.0 in each component. International applicants may also be required to meet UKVI visa requirements. A valid ATAS certificate (where applicable) must be secured before enrolment.