The School of Computer Science would like to present…
A Structured Survey of Anomaly Types and Detection Models in IoT Frameworks
PhD. Comprehensive Exam by: Atefeh Gilvari
Date: Friday, June 13, 2025
Time: 11:30 AM
Location: Essex Hall, Room 122
In dynamic Internet of Things (IoT) environments, traditional anomaly detection surveys often treat all anomalies as a unified concept, overlooking the distinct characteristics and detection challenges posed by specific anomaly types. This paper presents a structured survey and comparative analysis of anomaly detection methods, organized by key anomaly categories such as drift, novelty, bias, noise, constant-value, and stuck-at anomalies. Each anomaly type is formally defined, with its theoretical foundation, modelling challenges, and applicable parametric, non-parametric, and neural network-based methods systematically reviewed. We analyze how model effectiveness varies across anomaly types and identify state-of-the-art techniques best suited for each. A novel contribution of this work is a taxonomy-driven mapping of models to anomaly types, designed to support targeted method selection and system design in resource-constrained IoT deployments. Our findings emphasize the need for interpretable, adaptive, and type-aware anomaly detection systems and outline open challenges in unified benchmarking, cross-type detectors, and ontology development for anomaly classification.
Large-Scale Sensor Networks, IoT, Anomaly Type, Anomaly Classification, Anomaly Detection
External Reader: Dr. Ning Zhang
Internal Reader: Dr. Boubakeur Boufama
Internal Reader: Dr. Arunita Jaekel
Advisor(s): Dr. Ziad Kobti, Dr. Narayan Kar
