Title :
Noise tolerance of output-coded neural net
Author :
Al-Mashouq, Khalid A.
Author_Institution :
Dept. of Electr. Eng., King Saud Univ., Riyadh, Saudi Arabia
Abstract :
Error correcting codes were used previously to encode the output of feed-forward neural nets. We study the effect of additive noise on the performance of a coded net and compare it to an uncoded net. Some necessary analytical tools are developed to estimate the performance of a neural net in the presence of noise. Simulation examples (isolated word utterances recognition) are also included to show the advantage of coding in reducing the probability of classification error due to noise. In addition we point the use of the estimated performance as a lower limit to the performance of a multilayer neural net
Keywords :
Gaussian noise; coding errors; error correction codes; feedforward neural nets; multilayer perceptrons; probability; speech recognition; white noise; AWGN; additive noise; analytical tools; classification error probability; error correcting codes; feedforward neural nets; lower limit; multilayer neural net; neural network performance; noise tolerance; output coded neural net; simulation; speech recognition; uncoded net; AWGN; Additive white noise; Decoding; Error correction; Error correction codes; Feedforward neural networks; Feedforward systems; Gaussian noise; Hamming distance; Neural networks;
Conference_Titel :
Digital Signal Processing Workshop Proceedings, 1996., IEEE
Conference_Location :
Loen
Print_ISBN :
0-7803-3629-1
DOI :
10.1109/DSPWS.1996.555557
