基于神经网络均衡器的小波包多载波扩频系统

Wavelet packet spread spectrum system based on neural network equalizers

在多径衰落信道下,本文提出了一种基于神经网络均衡器的小波包多载波扩频系统(Neural Network Equalizers Wavelet Packet Spread Spectrum,NNE-WPSS)。本文利用基于最小均方算法(Least Mean Square algorithm,LMS)的复径向基函数神经网络均衡器(Complex Radial Basis Function Network Equalizers,CRBF)来抑制由多径衰落信道引起的码间干扰(Inter-Symbol Interference,ISI)并且利用最大似然算法对解调后的码元进行检测。在多径衰落信道和白高斯噪声情况下,本文分析了基于神经网络均衡器的新型小波包多载波扩频系统的传输性能。仿真结果表明,本文所提出的基于神经网络均衡器的新型小波包多载波扩频系统的性能要优于传统的基于正交频分复用(Orthogonal Frequency Division Multiplexing,OFDM)多载波扩频系统;本文提出的复径向基函数神经网络均衡器(CRBF)的性能也优于迫零均衡器(Zero-Forcing Equalizer,ZFE)。

A novel wavelet packet spread spectrum （WPSS） system based on neural network equalizers （NNE） is studied over multipath fading channels. The proposed complex radial basis function （CRBF） network equalizers based on least mean square （LMS） algorithm can mitigate the detrimental effects of the inter-symbol interference （ISI） caused by multipath fading channels. The demodulated symbols are detected by maximum likelihood （ML） algorithm. The transmission performance of the proposed scheme is analyzed in multipath fading channels with Additive Gaussian White Noise （AWGN）. It is demonstrated by simulation results that the proposed WPSS scheme can provide greater immunity to multipath fading channels and AWGN than the traditional orthogonal frequency division multiplexing spread spectrum （OFDM-SS） scheme. The performance of the proposed CRBF neural network equalizer is better than that of the Zero-Forcing equalizer （ZFE）