Research Article

Design of Adaptive Filter Using Jordan/Elman Neural Network in a Typical EMG Signal Noise Removal

Table 1

For hidden layers  = 2, input PE  = 1, output PE  = 1, transfer function  = TanhAxon, learning rule  = momentum, maximum epochs  = 1000, and threshold MSE  = 0.01.

Serial no.Type of ANNHidden layer variation H1 H2Correlation coefficient MSE ( sec)
TrainingCross validationTesting

01Jordan/Elman network09,060.782 021 0850.009 907 1530.017 699 7490.013 501 9877.511.3

02Jordan/Elman network07,050.775 341 1140.009 949 3770.018 167 1610.013 811 6687.510.8

03Jordan/Elman network08,050.777 408 2970.00 996 4720.017 879 2740.013 631 6491032.6