Title :
A novel multi-user detection method based on IEA-RBF neural network in CDMA
Author :
Yang, Shuyuan ; Jiao, Licheng ; Liu, Fang
Author_Institution :
Key Lab for Radar Signal Process., Xidian Univ., Xi´´an, China
Abstract :
The radial basis function neural network (RBFNN) is a feedforward neural network; it has great advantages over the MLP (multilayer perceptron) in the approximation and classification. However it is difficult to construct a simple and efficient NN with learning algorithm, that is, there are so many parameters to determine. A novel IEA-RBFNN multiuser detector is presented based on the immune theory in biology. The goal of the algorithm is to solve the contradiction of the complexity and the performance of the RBF-NN in the application in CDMA. Simulations show that the algorithm can converge rapidly and effectively eliminate multi-access interference, so it is near-far resistant.
Keywords :
code division multiple access; computational complexity; convergence; evolutionary computation; interference suppression; learning (artificial intelligence); mobile radio; multiuser detection; radial basis function networks; CDMA; IEA-RBFNN multiuser detector; complexity; convergence; feedforward neural network; immune evolutionary algorithm; immune theory; learning algorithm; multi-access interference elimination; near-far resistance; performance; radial basis function neural network; Biological system modeling; Detectors; Feedforward neural networks; Immune system; Multi-layer neural network; Multiaccess communication; Multilayer perceptrons; Multiuser detection; Neural networks; Radial basis function networks;
Conference_Titel :
Signal Processing, 2002 6th International Conference on
Print_ISBN :
0-7803-7488-6
DOI :
10.1109/ICOSP.2002.1180984