I'm a PhD researcher in Electrical Engineering at the University of Rhode Island (CYPHER Lab), working at the intersection of electrical engineering, computer networking, and cybersecurity. I program security logic directly into the network switches that carry power-grid and ICS/SCADA traffic — obfuscating devices against reconnaissance, deploying data-plane decoys, and catching threats at wire speed.
Before research, I spent eight years running IT/OT and SCADA systems at Ghana Water Company, a national water utility — so I build systems meant to survive production, not just a paper. My work is jointly funded by the Office of Naval Research (ONR) and the National Science Foundation (NSF).
- Programmable Network Defense — P4/SDN security architectures for real-time detection and autonomous mitigation at the data plane (BMv2, Intel Tofino ASICs)
- ICS/SCADA Obfuscation & Deception — anti-reconnaissance and data-plane decoys (e.g. DNP3 outstations) that mislead attackers while gathering intelligence
- ML-Driven Threat Detection — GRU/RNN models for traffic anomaly detection and lightweight classifiers for resource-constrained ICS/OT edge devices
- Risk Quantification & Governance — FAIR quantitative risk modeling, FDNA cascading-failure analysis for power grids, and multi-jurisdictional AI governance
- Agent-P4-SDN-Based DNS Threat Defense — Closed-loop autonomous defense pipeline that detects and mitigates DNS-based attacks using P4 programmable switches and SDN orchestration.
- dnp3_decoy — Programmable data-plane decoy that impersonates DNP3 outstations (RTUs) on a Tofino switch, presenting attackers with realistic virtual devices, MAC addresses, and OS fingerprints.
- adaptive_routing — P4-based adaptive load balancer that distributes traffic across equal-cost paths using real-time link telemetry, improving on static ECMP — applicable to resilient grid communications.
- Quantifying Systemic Risk in Critical Power Infrastructure (FDNA) — Functional Dependency Network Analysis modeling single-node failures into grid-wide cascades to prioritize resilience investment.
- MemProof — Admission-control protocol defending against external corpus poisoning in RAG systems by unauthorized sources.
Programmable Networks & Data Plane
Languages
ML / AI
Security Operations & GRC
Platforms & Tools
- P. Akekudaga, M. A. Bonsu. Navigating Multi-Jurisdictional Privacy Compliance in AI: An Empirical Analysis of Regulatory Gaps. IEEE UEMCON 2025.
- P. Akekudaga, O. Keskin. Quantifying Systemic Risk in Critical Power Infrastructure Using FDNA: From Single-Node Failure to Grid-Wide Cascades. Submitted, SmartNets 2026.
- P. Akekudaga, O. Keskin. A Framework for Financial Markets Impact Assessment of Data Breaches Using Interpretable ML and Event Study Methods. ASIA '25.
- M. A. Bonsu, P. Akekudaga. Resilient IoT Security: Early Flood Attack Detection Using a GRU Deep Learning Model. WJARR, 2025.
Full record: ORCID 0009-0003-2263-0142
- 🏆 Winner, SIRAcon '25 Research Competition — Society of Information Risk Analysts
- 🥇 Cyber 9/12 Strategy Competition — Semi-finalist & "Most Creative Proposal to the White House"
- 🎓 Member: IEEE, FAIR Institute, SIRA, NSBE, ISACA
