Students
Learn to distinguish a result from the evidence that warrants it—and the claims it cannot support.
Now available
Cyber Analytics Under Partial Observability
Measurement, Inference, and Governed Action for IoT and Operational Technology
Security decisions begin where complete visibility ends. This book shows how to make defensible decisions without turning a measurement into an identity, a ranking into a probability, a simulation into proof, or automation into ungoverned action.
Buy on Amazon
Paperback and Kindle editions
The argument
A rigorous method for consequential cyber decisions
Connected devices and operational technology rarely offer complete inventories, clean ground truth, or unlimited time. The book develops a disciplined approach to acting under those constraints while keeping the limits of the evidence visible.
Across two recurring cyber-physical cases, it moves from discovery and device fingerprinting to vulnerability assessment, attack graphs, safe validation, digital twins, post-quantum readiness, federated intelligence, supply-chain integrity, deception, formal verification, and autonomous remediation.
For readers who need to defend their decisions
Built for advanced learning and operational practice
Practitioners
Apply auditable methods when visibility is incomplete, time is constrained, and the cost of overclaiming is real.
Researchers & instructors
Use worked traces, reproducibility contracts, failure modes, and cross-chapter handoffs to examine how conclusions hold up.
Inside the book
Every conclusion has an evidence boundary
The book labels claims as measured, simulated, illustrative, literature-derived, operational observation, or design target. That distinction keeps readers from treating a useful model, trace, or automation plan as stronger evidence than it is.
Cyber Analytics Under Partial Observability
Make the next cyber decision defensible.
Get the book on AmazonFirst edition · 2026 · ISBN 979-8-9973650-0-4