Cellular Neural Network-Based Methods for Distributed Network Intrusion Detection

Table 8

The comparison results about DIDS with different methods.

Algorithms		DR (%)	FAR (%)
Centralized architecture DIDS	Genetic algorithm, artificial immune, and ANN [7]	84.52–97.5	2.16–3.64
	Artificial Immune System [8]	89.5–96.9	1.2–2.7
	Game-theoretical approach [9]	92.3–97.67	0.77–2.9
	Our DTCNN-based local detection method	87.83–98.16	0.71–2.94
Distributed architecture DIDS	ANN [12]	Average 96.24	None
	Signature-based multilayer IDS [13]	Average 96.7	Average 1.83
	Dynamic Election Algorithm [14]	96–97.5	0.93–1.97
	Our SCCNN-based global detection method	96.19–97.76	0.795–1.845