Research Vision

I build engineering systems that perform reliably in constrained environments. The unifying thread across my work is resilient sensing and intelligent infrastructure — from acoustic underwater channels and low-power sensor networks, to social housing testbeds and national climate-health platforms.

Research Themes

Resilient Communications & the Internet of Underwater Things

The Internet of Underwater Things (IoUT) is one of the most demanding application areas in wireless engineering — harsh acoustic channels, low bandwidth, high latency, and severe energy constraints. My PhD work introduced new clustering and routing protocols (DEKCS) that extended underwater sensor-network lifetime by over 70%, and reinforcement-learning approaches for simultaneous wireless information and power transfer (SWIPT) using AUVs.

Smart Housing & Ambient Intelligence

Working with data from a social landlord in the Scottish Highlands, I have completed IoT systems studies for social housing with direct UK social value:

  • A multi-sensor mould-risk forecasting system that predicts risk 6–48 hours ahead (24-hour AUC-ROC 0.851–0.957, cross-home transfer up to 0.968) — enabling preventative ventilation and heating decisions.
  • A non-intrusive welfare monitoring system for lone elderly residents using per-appliance power monitoring, environmental sensing, radiator valve states, and circuit-level energy signals (F1 0.848 / 0.814; detection rates 87–100%; ~£200–300 per flat) — without cameras, microphones, or wearables.

Climate-Health Engineering & Open Statistical Infrastructure

As technical lead for SOSCHI at the ONS, I lead the platform engineering that turns climate-health methods into reliable institutional infrastructure. The associated climatehealth R package — published on CRAN — operationalises six indicators (temperature-related mortality, wildfire smoke exposure, air pollution, suicides related to extreme heat, malaria, and diarrhoeal disease) endorsed by the United Nations Statistical Commission for the Global Set of Environment and Climate Change Statistics.

Federated Learning & Edge Intelligence

I work on hierarchical federated learning, semantic communication, and edge inference for IoT systems — including a federated anomaly detection framework for the IoUT and semantic telemetry for AUV missions using FathomNet-trained models and visual-anchor navigation.

Network Science for Public Resilience

At the ONS Global Supply Chains Intelligence Programme (Treasury-funded), I applied graph and network engineering methods to map vulnerabilities and critical dependencies in UK supply systems.

SOSCHI — Climate-Health Platform & climatehealth R Package

  • £5m Wellcome Trust-funded programme hosted on the UN Global Platform.
  • Lead the engineering of the SOSCHI platform: services, APIs, deployment, monitoring, maintenance.
  • Lead maintainer of the climatehealth R package on CRAN — implements UN-endorsed climate-health indicators.
  • UN Statistical Commission endorsement (March 2026, 57th session, New York).

Social Housing Smart-Home Engineering

  • Multi-sensor mould-risk forecasting and welfare monitoring for social housing in the Scottish Highlands.
  • Two manuscripts under review at IEEE Internet of Things Journal and IEEE Journal of Biomedical and Health Informatics.
  • Demonstrates affordable, privacy-preserving engineering for healthy housing and independent ageing.

IoUT Cognitive Networking

  • DEKCS clustering protocol (>70% lifetime extension), Q-learning route selection, RL-SWIPT for AUVs.
  • Comprehensive ML-for-IoUT survey under revision at IEEE Communications Surveys and Tutorials.
  • Hierarchical federated anomaly detection under review at IEEE Internet of Things Journal.
  • Semantic telemetry & visual-anchor navigation under review at IEEE Journal of Oceanic Engineering.

EchoLingo Health (Voith AI)

  • Edge AI clinical assistant for low-resource African languages.
  • Real-time transcription, SOAP note automation, ICD/CPT coding, FHIR APIs.
  • Designed for offline operation in low-bandwidth clinical environments.