Challenges
Data Integrity attacks occur due to physical and internal cyber exploits
Various attack types and strategies are possible, and the utility does not know which one will happen.
Different attacks are treated in silos, causing too many defense solutions
Can we build one framework that works for various attacks and multiple domains?
Attack Scale
Fraction of compromised IoT Device Endpoints varied from very low to large
Attack Strength
Average Margin of Perturbation per IoT device's data stream
Data Integrity Attack Types
Additive, Deductive, Camouflage, Alternating Switching
Attack Strategies
Data Order Aware, Incremental/Ramp, KL Distance minimization, Scaling
Intellectual Merits
A novel theory of Pythagorean Mean based Invariant which characterizes benign cluster-level behavior
The Pythagorean Invariants contains properties that deviate under various attack features and
Invariant deviation properties have the ability to indicate attack type, strategy, and severity, (Context)
Context informs a subsequent device level anomaly detector.
Developed two device level anomaly detectors:
(1) context embedded KL divergence;
(2) bio inspired information theory built on Renyi Entropy.
Our novel bio inspired information theoretic frameworks can detect low margin data falsification attacks.
Unifying data driven smart energy and transportation security frameworks based on anomaly detection.
Publications/Products
1. S. Bhattacharjee, P. Madhavarapu, S. Silvestri, S. K. Das, “Attack Context Embedded Data Driven Trust Diagnostics in Smart Metering Infrastructure” ACM Transactions on Priv. and Sec., 2021. Paper Link, Code Link
2. S. Bhattacharjee, P. Madhavarapu, S. K. Das, “A Diversity Index Scoring Framework for Identifying Smart Meters Launching Stealthy Data Falsification Attacks”, ACM Asia’ CCS, 2021 Paper Link, Code Link
3. M. Islam, J. Talusan, S. Bhattacharjee, F. Tiasus, S. Vazirizade, A. Dubey, K. Yasumoto, S. K. Das “Anomaly based Incident Detection in Large Scale Transportation Systems”, IEEE/ACM Intl. Conf. on Cyber Physical Sys. (IEEE ICCPS), 2022. Paper Link , Code Link (without revealing private data)
4. P. Roy, S. Bhattacharjee, H. Al-Sheakh, S. K. Das, “Resilience against Bad Mouthing Attacks in Mobile Crowdsensing Systems via Cyber Deception” IEEE World of Wireless MObile and Multimedia Networks (IEEE WoWMoM), 2021 Paper Link
5. P. Roy, S. Bhattacharjee, S.K. Das, “Real Time Stream Mining based Attack Detection in Distribution Level PMUs for Smart Grids,” IEEE Global Communications Conference (IEEE Globecom), 2020 Paper Link
6. H. Al-Sheakh, S. Bhattacharjee, “Towards a Unified Trust Framework for Detecting IoT Device Attacks in Smart Homes”, IEEE Conference on Mobile Ad-Hoc Sensor and Smart Systems (IEEE MASS), Dec. 2020 Paper Link
7. S. Bhattacharjee, M. J. Islam, S. Abedzadeh, "Robust Anomaly based Attack Detection in Smart Grids under Data Poisoning Attacks" ACM Asia 'CCS Workshop on Cyber Physical System Security, June 2022. Paper Link
8. Y. Ishimaki, S. Bhattacharjee, H. Yamana and S. K. Das, "Towards Privacy-preserving Anomaly-based Attack Detection against Data Falsification in Smart Grid," IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (IEEE SmartGridComm), 2020 Paper Link
9. P. Madhavarapu, S. Bhattacharjee, S. K. Das, “A Generative Model for Evasion Attacks in Smart Grid”, IEEE INFOCOM BigSecurity Workshop, 2022. Paper Link, Code Link
10. P. Madhavarapu, P. Roy, S. Bhattacharjee, S. K. Das, “Active Learning Augmented Folded Gaussian Model for Anomaly Detection in Smart Transportation”, IEEE ICC IoTSN Symposium, 2022., Code Link
11. P. Roy, S.Bhattacharjee, S. Abedzadeh, S.K. Das, ``Noise Resilient Learning for Attack Detection in Smart Grid PMU Infrastructure", IEEE Trans. on Dependable and Secure Computing (Special Issue on Reliability and Robustness in AI based Cybersecurity Solutions), Vol. 21(2), pp. 618-635, March. 2024. Paper Link
12. M.J. Islam, J.P. Talusan, S.Bhattacharjee, F. Tiasus, A. Dubey, K. Yasumoto, S.K. Das, "Scalable Pythagorean Mean based Incident Detection in Smart Transportation Systems", ACM Transactions on Cyber Physical Systems, June 2023. Paper Link
13. S. Bhattacharjee, S.K. Das, "Building a Unified Data Falsification Threat Landscape for Internet of Things/Cyber Physical Systems", IEEE Computer, Vol.56(3), March 2023 Paper Link
14. A. Paul, S. Bhattacharjee, S.K. Das, "Securing Smart Water Metering Infrastructures: A Scalable Approach to Detecting False Data Injection", IEEE SmartComp Workshop on Big Data and IoT Security, June, 2023.
15. S. Bhattacharjee, S.K. Das, "Information Integrity in Participatory Crowd-Sensing via Robust Trust Models", Advances in Mobile Crowdsensing: From Theory to Practice Springer Cham, July 2023. Paper link
16. S. Bhattacharjee, S.K. Das, "Unifying Threats against Information Integrity in Participatory Crowd Sensing", IEEE Pervasive Computing, 2023.
17. J. Okonofua, R. Meyer, S. Bhattacharjee, "Information Diversity based Detection for ON-OFF Low Strength DDoS Attacks in Smart Home IoT" International Conference on Distributed Computing and Networking (ICDCN '24), pp. 292–297, 2024. https://doi.org/10.1145/3631461.3631954
18. S. Abedzadeh and S. Bhattacharjee, "On the Role of Re-Descending M-Estimators in Resilient Anomaly Detection for Smart Living CPS," 2024 IEEE International Conference on Smart Computing (SMARTCOMP), Japan, 2024, pp. 198-205. Paper Link
19. A. Oluyomi, S. Abedzadeh, S. Bhattacharjee, S.K. Das, "Unsafe Event Detection in Smart Water Metering Infrastructure via Noise Resilient Learning", IEEE/ACM International Conference on Cyber Physical Systems, pp. 259-270, 2024. Paper Link
20. Md. Jaminur Islam, J. Talusan, S. Bhattacharjee, F. Tiausas, A. Dubey, K. Yasumoto, S.K. Das. "Scalable Pythagorean Mean-based Incident Detection in Smart Transportation Systems". ACM Trans. Cyber-Phys. Syst. 8, 2, Article 20 (April 2024), 25 pages Paper Link
21. M. Maliha, S. Bhattacharjee "A Unified Time Series Analytics based Intrusion Detection Framework for CAN BUS Attacks" ACM Conference on Data and Application Security and Privacy (CODASPY '24), 19–30, 2024. Paper Link
People
Sajal K. Das
PI, Missori Univ. S & T
Shameek Bhattacharjee
PI, Western Michigan Univ
(WMU)
Venkata Praveen Kumar Madhavarapu,
Former PhD Student (MST)
Prithwiraj Roy,
Former PhD. Student (MST)
Mohammad Jaminur Islam,
Former PhD. Student, WMU
Sahar Abedzadeh, PhD. Student, WMU
Joseph Okonofua, Undergraduate Student WMU
Alvaro Rivas, Undergraduate Student WMU
Maisha Maliha, PhD Student WMU
Ayanfeoluwa Oluyomi, PhD Student MST
Broader Scientific Impact
Generic threat landscape for data integrity attacks in smart living CPS
Formal Analysis: closed form expressions to predict security performance given any dataset.
Proved invariant based anomaly detection applies to Smart Energy Metering, Phasor Measurement Units, and Smart Transportation
Proven compatibility with customer privacy preserving frameworks (e.g., Fully Homomorphic Encryption)
Closed form approximation of failure/ evasion points of anomaly detector