%0 Journal Article
%J Neural Computing and Applications [Impact Factor SJR2012: 1.168]
%D 2013
%T A new solution to the hyperbolic tangent implementation in hardware: polynomial modeling of the fractional exponential part
%A Nascimento, Ivo
%A Jardim, Ricardo
%A Morgado, Dias F.
%K Algorithms implemented in hardware
%K CORDIC
%K Gate arrays
%K Hardware description languages
%K Neural networks
%X The most difficult part of an artificial neural network to implement in hardware is the nonlinear activation function. For most implementations, the function used is the hyperbolic tangent. This function has received much attention in relation to hardware implementation. Nevertheless, there is no consensus regarding the best solution. In this paper, we propose a new approach by implementing the hyperbolic tangent in hardware with a polynomial modeling of the fractional exponential part. The results in the paper then demonstrate, through the use of an example, that this solution is faster than the CORDIC algorithm, but slower than the piecewise linear solution with the same error. The advantage over the piecewise linear approach is that it uses less memory.
%B Neural Computing and Applications [Impact Factor SJR2012: 1.168]
%I Springer London
%V 23
%P 363-369
%G eng
%U http://dx.doi.org/10.1007/s00521-012-0919-0
%0 Journal Article
%J Neural Computing and Applications, [Impact Factor: 1.168]
%D 2012
%T A new solution to the hyperbolic tangent implementation in hardware: polynomial modeling of the fractional exponential part
%A Nascimento, Ivo
%A Jardim, Ricardo
%A Morgado, Dias F.
%K Artificial Neural Networks
%K Hardware description languages
%B Neural Computing and Applications, [Impact Factor: 1.168]
%I Springer-Verlag
%P 1-7
%8 04/2012
%G eng
%U http://dx.doi.org/10.1007/s00521-012-0919-0