Title :
RBF growing algorithm applied to the equalization and co-channel interference rejection problem
Author :
Howitt, Ivan ; Reed, Jeffrey H. ; Vemuri, V. ; Hsia, T.C.
Author_Institution :
Dept. of Electr. & Comput. Eng., California Univ., Davis, CA, USA
fDate :
27 Jun- 2 Jul 1994
Abstract :
Recently the RBF network has been successfully applied to a number of communication signal processing problems, including equalization and co-channel interference rejection. Determining the number and locations of the centers is an important but usually neglected issue. A growing RBF algorithm for determining the number and location of the centers is presented and the performance is compared to a RBF with prescribed centers. The growing algorithm´s advantages are its limited need for prior information and its ability to better utilize the allocated centers. As an example, the growing algorithm reduced the number of centers by more than a factor of 231 over the prescribed method to achieve a desired performance for an equalization and interference rejection application
Keywords :
cochannel interference; equalisers; feedforward neural nets; interference suppression; signal processing; RBF growing algorithm; co-channel interference rejection; communication signal processing; equalization; neural nets; radial basis function network; Clustering algorithms; Computer networks; Feeds; Interchannel interference; Neural networks; Parameter estimation; Radial basis function networks; Signal processing algorithms; State-space methods; Vector quantization;
Conference_Titel :
Neural Networks, 1994. IEEE World Congress on Computational Intelligence., 1994 IEEE International Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-1901-X
DOI :
10.1109/ICNN.1994.374910