| 摘要: |
| 针对微弱直扩信号的盲检测与估计问题，在接收方未知发送方扩频序列的前提下
，提出了一种恢复直接序列扩频（DSSS）信号扩频码的方法。该方法基于反向传播（B
P）神经网络，它的输入是接收到的信号，而其期望输出是和输入相同的信号，根据误差反
向传播来有监督地调节神经网络，网络达到收敛时根据第二层权值的符号函数值即可盲估计
出扩频码序列。实验结果表明，即使是在负信噪比情况下，该方法也能得到一个很好的估计
效果。 |
| 关键词: 直接序列扩频 BP神经网络 改进BP算法 伪码序列 盲识别 |
| DOI： |
|
| 基金项目:国家自然科学基金资助项目(61071196)；国家自然科学基金-中
物院NSAF联合基金资助项目（10776040）；教育部新世纪优秀人才支持计划（NCET-10-0927
）；信号与信息处理重庆市市级重点实验室建设项目（CSTC2009CA2003）；重庆市科委自然
科学基金资助项目（CSTC2009BB2287） |
|
| DSSS Signal Spread Spectrum Codes Blind Identification Based on BP Neural Network |
| ZHAO De-fang,ZHANG Tian-qi,JIN Xiang,DU Xiao-hua |
| (Chongqing Key Lab of Signal and Information Processing, Chongqing University of Posts and Telecommunications, Chongqing 400065, China) |
| Abstract: |
| In view of blind detection and estimation problems for weak DS
signals, a method is presented to recover the spread spectrum code of direct seq
uence spread
spectrum(DSSS) signal, whereas the receiver has no knowledge of the trans
mitter’s spreading sequence. The method is based on back propagation (BP) neura
l network, whose input is the received signal, and the expected output is the sa
me as the input signal. It is supervised to adjust the neural network according
to the error back-propagation, and then it can perform blind estimation of the s
pread spectru
m code sequences according to the symbol function value of the second layer weig
hts
when adjusting the network to reach convergence. Experimental results show that
the method provides a good estimation, even when the signal power is below the n
oise power. |
| Key words: direct sequence spread spectrum(DSSS) BP (Back-Propagation) neural network imp
roved BP algorithm PN (Pseudo Noise) sequence blind identification |
《电讯技术》编辑部 