小波神经网络在MC-CDMA多用户检测技术中应用研究

对多载波码分多址(MC-CDMA)系统的干扰受限问题,提出了利用小波变换良好的时频局部特性和神经网络的良好自学习能力,提高多用户检测性能的原理。同时为解决网络权值和参数修正进化缓慢并且容易陷入最小的问题,采用增加动量项的方法提高网络学习效率。建立了基于小波神经网络的多用户检测器并应用于MC-CDMA系统中。用MATLAB/Simulink软件搭建仿真系统,接收端采用解相关检测(MMSEC)和正交恢复(ORC)检测算法。实验表明,基于小波神经网络的多用户检测技术在误码率(BER)性能上更接近单用户的BER性能。

Interference on multi-carrier code division multiple access （MC-CDMA） system is limited problem, a wavelet transform good use when the frequency of local characteristics and good self-learning neural network capacity and improve the performance of multi-user detection principle. At the same time solve the network weights and parameters evolve slowly and fix vulnerable to the smal est of problems, to increase momentum learning methods to improve network efficiency. The establishment of a multi-user detector based on wavelet neural network and applied to MC-CDMA system. Using MATLAB /Simulink software to build simulation system, the receiving end solution using correlation detection （MMSEC） and quadrature recovery （ORC） detection algorithm. Experimental results show that multi-user detection techniques based on wavelet neural network on the bit error rate （BER） performance closer to the BER performance of single users.