Intelligent Noise Removal from EMG Signal Using Focused Time-Lagged Recurrent Neural Network

Table 5

Variable parameters of FTLRNN model.

S.N.	Parameter	Typical range	Optimal value
1	Learning rule	Momentum, Conjugate Gradient, Levenberg Marquardt (LM), Quick propagation (qp), step, delta bar delta (dbd)	Momentum
2	Transfer function in output layer	Linear, lineartanh, Tanh	Linear
3	Memory structure	TDNNAxon, Gamma, Laguarre	Laguarre
4	Depth in samples	2 to 12	4
5	Length of trajectory	10 to 60	50