Resilient Built Environments
Project Overview
This EPSRC PETRAS secondment project, embedded within the Building Research Establishment (BRE), investigates how multiple independent sensing layers can enhance the resilience of smart homes and office environments against cyber-physical threats. As built environments become increasingly dependent on interconnected IoT devices for monitoring, automation, and security, they also become vulnerable to sophisticated tampering and attack scenarios that can compromise occupant safety and operational continuity. The research compares resilience requirements across domestic and commercial building domains, identifying common vulnerabilities and domain-specific challenges that arise when IoT networks are deployed in different built environment contexts. The project designs controlled experiments to evaluate how independent, complementary IoT sensor networks can detect and mitigate cyber-physical tampering by providing redundant observational capabilities that are difficult for attackers to compromise simultaneously.
Case studies evaluate anomaly detection techniques applied to building sensor data and assess multi-layer sensing architectures that combine diverse sensor modalities. These layered approaches strengthen the overall security posture of connected buildings by ensuring that no single point of compromise can disable the entire monitoring infrastructure, providing defence-in-depth for critical building systems.
Funded by EPSRC through the PETRAS National Centre of Excellence for IoT Systems Cybersecurity and conducted in partnership with GCHQ and BRE, the project informs the creation of future research programmes. These programmes address resilience challenges in connected buildings, contributing to understanding how layered sensing strategies can protect critical building infrastructure from emerging cyber-physical threats in an era of increasing IoT adoption.
Team
Funding
Partners
Repositories
casper-shield-research
CasperShield: Cyber-Physical Behavioural Anomaly Detection in Smart Homes — Research Monorepo
casper-shield-app
CasperShield: Smart Home Security Mobile App (Flutter) + Simulator
anoml-iot-pipeline
AnoML IoT Pipeline offers scripts to capture sensor readings with Grove modules, build datasets, clean and scale data, train anomaly detectors (RNN, CNN, OCSVM) in TensorFlow or scikit-learn, and evaluate models on fog/edge systems. Includes conversion to TFLite/TFMicro for microcontroller deployment.
smarthome-traffic-classifier
Jupyter notebooks and datasets for smart home network traffic classification. Includes binary IoT vs non-IoT detection, multi-class device fingerprinting, and device state recognition. Provides CSV traffic captures for training and evaluation.
bre-smartbuilding-analysis
Network traffic analysis toolkit for the BRE smart building dataset. Provides a shell script that runs Tshark to generate protocol stats, endpoint and conversation summaries, HTTP & DHCP reports, plus a Plotly/Scapy notebook for interactive PCAP visualisations.
smart-home-lab
Provides scripts to stream network and home automation data via Kafka, query Cassandra archives, Jupyter notebooks for PCAP and Home Assistant/OpenHAB analysis, and setup guides for building an IoT testbed with MQTT and data capture.
smarthome-activity-monitoring
Flask-based platform for smart home network activity monitoring. Captures packets via tshark and scapy, transforms flows with CICFlowMeter, and applies ML models to classify device states. Uses APScheduler for periodic capture, SQLAlchemy storage, and templated web dashboard.
smart-home-activity-simulator
Web-based smart home activity simulator using Python, MQTT, and OpenHAB. Configure floors, rooms, and devices; set event times and payloads; then publish them via MQTT. Includes logging, log viewer, Selenium connector, and Bootstrapped frontend assets.