Comparative analysis of state-of-the-art cybersecurity frameworks
| Study | Methodology | Objective | Limitations addressed by our work |
|---|---|---|---|
| Nguyen et al. (2021) | Survey on FL for IoT | Reviews FL challenges in IoT | Theoretical survey; does not propose or validate an integrated solution |
| Guo et al. (2024) | FTL | Intrusion detection for IoT | Lacks mechanism to ensure node trust or address zero-day threats |
| Anitha and Murugan (2024) | FL + Blockchain | Secure data sharing | Does not leverage GenAI or FTL for non-IID data handling |
| Kumar et al. (2024) | FCD (FL + crypto) | Unified defense against utility and privacy attacks | Focuses on cryptographic defenses; lacks GenAI for synthetic data or FTL adaptation |
| Anitha and Murugan (2024) | FL + Blockchain | Privacy-preserving IoT security | Focuses on privacy/integrity; omits zero-day generation and adaptive modelling |
| Xie et al. (2024) | PoisonedFL attack | Demonstrates model poisoning attack | Highlights vulnerability but does not offer a defensive framework |
| Our work | FTL + GenAI + Blockchain | Privacy-preserving, adaptive and robust IDS | Handles non-IID data, zero-day threats and model poisoning through integrated approach |
| Study | Methodology | Objective | Limitations addressed by our work |
|---|---|---|---|
| Survey on | Reviews | Theoretical survey; does not propose or validate an integrated solution | |
| Intrusion detection for IoT | Lacks mechanism to ensure node trust or address zero-day threats | ||
| Secure data sharing | Does not leverage GenAI or | ||
| Unified defense against utility and privacy attacks | Focuses on cryptographic defenses; lacks GenAI for synthetic data or | ||
| Privacy-preserving IoT security | Focuses on privacy/integrity; omits zero-day generation and adaptive modelling | ||
| PoisonedFL attack | Demonstrates model poisoning attack | Highlights vulnerability but does not offer a defensive framework | |
Sharing content requires targeting cookies to be enabled. Please update your cookie preferences to use this feature.