The application of a recurrent neural network structure to adaptive antenna arrays in a CDMA system is presented. This novel approach enables space and time diversity combining of the signals, and is evaluated over multipath fading channels. Computer simulation results show that the recurrent structure performs better and has a lower computational complexity than the RLS-based adaptive array
Published in:
Electronics Letters
(Volume:33
,
Issue:
25
)
Date of Publication: 4 Dec 1997