Digital Library
Close Browse articles from a journal
 
<< previous    next >>
     Journal description
       All volumes of the corresponding journal
         All issues of the corresponding volume
           All articles of the corresponding issues
                                       Details for article 15 of 32 found articles
 
 
  Direct Exponential Function Computer in Neural Network Application Based on Evolvable Hardware
 
 
Title: Direct Exponential Function Computer in Neural Network Application Based on Evolvable Hardware
Author: Z. Huajun
Z. Jin
Appeared in: Information technology journal
Paging: Volume 7 (2008) nr. 8 pages 1156-1162
Year: 2008
Contents: This study proposes a new direct exponential function computer based on CORDIC algorithm to solve the neuron realization problem on the Evolvable Hardware (EHW). The direct exponential function computer combines the essential character of exponential function with CORDIC algorithm and turns aside the conventional approach which computes the exponential function by adding sinh (x) and cosh (x) indirectly. Because the direct computer executes shift-add operation in allusion to exponential function directly and get rid of the computation of sinh (x) and cosh (x), it not only calculates much faster than the indirect computer, but also saves much more logic gates. In addition, through strict and integrated analysis, the direct computer is proved to own perfect convergence. It converges to the real value with arbitrary precision and operates shift-add operation without any unnecessary rotation. Since of the direct efficient shift-add operation, the direct computer is useful for the application of neural network based on the EHW.
Publisher: Asian Network for Scientific Information (provided by DOAJ)
Source file: Elektronische Wetenschappelijke Tijdschriften
 
 

                             Details for article 15 of 32 found articles
 
<< previous    next >>
 
 Koninklijke Bibliotheek - National Library of the Netherlands