Table 3: Minimum mean accuracy comparison performance of four networks with six PSD feature extraction methods.

S.No	Features	Minimum Mean Accuracy of Neural Network Classifiers (%)
		FFNN	PNN	CFBNN	FMSVM
1	AR Burg	90.83	89.21	90.25	92.25
2	AR Yule – Walker	89.77	89.17	89.42	91.42
3	AR Covariance	88.95	89.04	89.00	89.00
4	AR Modified Covariance	89.83	88.75	89.83	89.83
5	Levinson Durbin Recursion	88.27	88.46	89.00	89.00
6	Linear Prediction Coefficient	89.83	88.08	89.00	89.00