Complex neuron model with its applications to M-QAM data communications in the presence of co-channel interferences
Zengjun Xiang, Guangguo Bi · 1992
The concept of a complex neuron and its model are described. The complex neural-network-based adaptive decision feedback filter (CNNDFF) for M-QAM digital communication reception systems is put forward. The fast adaptive learning algorithm, called mixed-gradient-based fast learning algorithm with variable learning gain and selective updates, is adopted to train the CNNDFF. Experimental results indicate that the CNNDFF can simultaneously overcome the performance degradations due to multipath fading of channels and reject the non-Gaussian cochannel interferences efficiently. By using the fast learning algorithm, improved convergence and tracking ability can be obtained for the CNNDFF with moderate computational complexity.>