CNN-Based Personal Identification System Using Resting State Electroencephalography

Table 2

Comparison of the performance of the proposed personal identification systems with different sliding windows.

Window (s)	Sliding ratio	Scale of training set	Training time (min)	Rank-1 (%)	FRR (%)	FAR (%)	EER (%)
0.25	0.5	41747	560	99.40	0.15	0.15	0.15
	1	20928	309	99.39	0.17	0.17	0.17
0.5	0.5	20819	273	99.51	0.06	0.07	0.06
	1	10464	136	98.89	0.19	0.19	0.19
1	0.5	10355	147	99.04	0.20	0.18	0.19
	1	5232	74	94.80	1.38	1.38	1.38
2	0.5	5123	85	92.74	1.67	1.68	1.67
	1	2616	43	60.86	10.55	10.55	10.55

Bold values indicate the best performance.