Improved noise characteristics of a SAW artificial neural network RF signal processor for modulation recognition

Author

Kavalov, D. ; Kalinin, V.

Author_Institution

Sch. of Eng., Oxford Brookes Univ., Headington, UK

Volume

1

fYear

2001

fDate

2001

Firstpage

19

Abstract

A novel training algorithm for the SAW neural network (NN) digital modulation classifier is developed that allows one to improve significantly the noise performance in the case of classification of three modulation schemes. It yields high probability (90-95%) of correct recognition, reducing the required SNR at the input of the processor from 25-27 dB down to 12-15 dB, which is comparable to that of the two-signal classifier. Further improvement is achieved by adding one more layer of hidden neurons. As a result, the number of neurons in the first layer, i.e., the number of SAW filters implementing them, is reduced from seven to five, which considerably simplifies the SAW device topology

Keywords

interdigital transducers; neural nets; phase shift keying; quadrature amplitude modulation; surface acoustic wave signal processing; SAW RF signal processor; SNR; artificial neural network; device topology; digital modulation classifier; hidden neurons; modulation recognition; noise characteristics; training algorithm; Artificial neural networks; Digital modulation; Neural networks; Neurons; Programmable control; SAW filters; Signal processing; Signal to noise ratio; Surface acoustic wave devices; Surface acoustic waves;

fLanguage

English

Publisher

ieee

Conference_Titel

Ultrasonics Symposium, 2001 IEEE

Conference_Location

Atlanta, GA

Print_ISBN

0-7803-7177-1

Type

conf

DOI

10.1109/ULTSYM.2001.991569

Filename

991569