【作者】 高峰；

【导师】 万群；

【作者基本信息】 电子科技大学 ， 工程硕士（专业学位）， 2022， 硕士

【摘要】 在非合作通信场景下,研究直接序列扩频信号的盲解扩方法,有助于保障通信安全和提高军事对抗的能力。直接序列码分多址(Direct Sequence-Code Division Multiple Access,DS-CDMA)信号盲解扩的研究内容主要涉及直扩信号的检测、特征参数的估计、用户扩频码序列的估计以及解扩解调,其中扩频码序列的估计是整个环节中最困难和最核心的部分。本文主要针对突发直扩信号,开展了以下几个方面的工作:1.研究了基于快速独立成分分析(Fast-Independent Component Analysis,Fast-ICA)的多用户扩频码序列盲估计方法,以及在失步时间已知时对同步DS-CDMA信号码序列估计的改进方法,改进后码序列估计的误码率得到了改善,且降低了信号分段的复杂度。由于盲估计的码序列中含有误码,因此本文还针对线性反馈移位寄存器(Linear Feedback Shift Register,LFSR)序列,研究了序列类型识别、生成多项式估计、初态估计以及序列重构的方法,可以在误码率低于20%时对序列进行有效的识别和重构,具有较好的性能。通过仿真实验,验证了码序列纠错对解调性能的改善。2.提出了一种基于匹配搜索的码序列快速检测方法,仿真实验表明该方法具有良好的性能,且能在已知序列生成多项式时做到实时检测用户扩频码序列。3.针对宽带码分多址(Wideband Code Division Multiple Access,WCDMA)中采用的正交可变扩频因子(Orthogonal Variable Spreading Factor,OVSF)序列,本文充分利用序列的生成特性,提出了一种OVSF序列盲估计方法,该方法以预解扩后符号平均功率和近零率为判决依据。通过仿真实验,验证了该算法对多用户场景下OVSF序列估计的有效性。4.提出了一种基于分离重构的DS-CDMA信号时频差参数估计方法,有效避免了多峰模糊的问题。通过仿真实验,表明本文所提方法得到的时频差参数具有较高的精度。5.基于对各个模块的理论研究,设计了一套同时适用于单用户和多用户的直扩信号盲解扩系统,可以选择离线检测模式或者在线检测模式,最后用实际采集的某卫星信号对整个系统的性能进行了测验。更多还原

【Abstract】 In non-cooperative communication scenarios,the study of blind despreading methods for direct sequence spread spectrum signals is helpful to ensure communication security and improve the capability of military confrontation.The research content of blind despreading of Direct Sequence-Code Division Multiple Access(DS-CDMA)signals mainly involves the detection of direct spread signals,estimation of characteristic parameters,estimation of user spreading code sequences,and despreading and despreading.The estimation of the spreading code sequence is the most difficult and core part of the whole process.Thesis mainly studies the burst direct spread signal,and has carried out the following aspects:1.The blind estimation method of multi-user spreading code sequence based on Fast-Independent Component Analysis(Fast-ICA)is studied,and the estimation method of synchronous DS-CDMA signal code sequence when the out-of-synchronization time is known is studied.The improved method improves the bit error rate of code sequence estimation and reduces the complexity of signal segmentation.Since the code sequence of blind estimation contains errors,Thesis also studies the methods of sequence type identification,generator polynomial estimation,initial state estimation and sequence reconstruction for Linear Feedback Shift Register(LFSR)sequences.The sequence can be effectively identified and reconstructed when the bit error rate is lower than 20%,and it has good performance.The improvement of demodulation performance by code sequence error correction is verified by simulation experiments.2.A fast code sequence detection method based on matching search is proposed.Simulation experiments show that the method has good performance and can detect user spreading code sequences in real time when the polynomial is known.3.For the Orthogonal Variable Spreading Factor(OVSF)sequence used in Wideband Code Division Multiple Access(WCDMA),Thesis uses the pre-despreading symbol average power and near-zero rate As the decision basis,blind estimation of the OVSF sequence is realized.Through simulation experiments,the effectiveness of the algorithm for OVSF sequence estimation in multi-user scenarios is verified.4.A method for estimating time-frequency difference parameters of DS-CDMA signal based on separation and reconstruction is proposed,which effectively avoids the problem of multi-peak ambiguity.Simulation experiments show that the time-frequency difference parameters obtained by the method proposed in thesis have high accuracy.5.Based on the theoretical research of each module,a set of blind despreading system for direct-spreading signals suitable for both single-user and multi-user is designed.The offline detection mode or online detection mode can be selected.Finally,the performance of the whole system is tested with a certain satellite signal actually collected.更多还原