低信噪比下直扩信号盲检测技术研究

发布时间：2018-04-21 00:13

本文选题：直扩通信系统 + 信号检测　；参考：《哈尔滨工业大学》2017年硕士论文

【摘要】：随着信息时代的来临,人类日常生活以及在军事行动中对通信的依赖性越来越大。直扩通信的抗干扰性非常好,并且信息不容易被截获到,因此在如今得到了非常广泛的应用。在非协作通信中直扩信号的盲检测是直扩信号参数估计和扩频序列盲估计的基础,因此直扩信号的盲检测受到了很大的重视。然而以往的很多检测算法有的是半盲检测,有的检测算法所需的信噪比远高于直扩信号工作时信噪比,不能达到实际应用的要求。本文在深入研究与分析了直扩信号的基础上,提出了两种针对于低信噪比环境下的直扩信号盲检测算法。首先,本文在研究分析了时域相关检测算法的基础上改进了基于预测的时域相关(Estimation-based Time-domain Sliding Correlating Accumulation,ETSCA)算法,该算法通过估计扩频码与估计数据相互更新极大的抑制了带内噪声。仿真分析表明该算法对扩频码长为31位的直扩信号可以在信噪比为-15d B时检测出直扩信号,并且随着检测使用的数据长度的增加性能会进一步提升。并且采用矢量信号源生成的直扩信号对该算法进行了验证,结果表明该算法检测性能良好。其次,本文在研究了特征值分解算法后提出了针对中频信号的基于自相关的矩阵分析(Autocorrelation-based Matrix Analysis,ACMA)算法。仿真结果表明在同等条件下该算法的检测性能要比ETSCA算法提高2d B左右。通过理论推导了直扩信号同步偏移量对ACMA算法性能的影响。该算法在同步情况下检测性能最好,而在归一化同步偏移量为1/2时性能最差,并且通过仿真验证了理论推导的结果。最后,把两种检测算法结合起来提出了基于估计的自相关矩阵分析算法(Estimation-based Autocorrelation Matrix Analysis,EACMA)。该算法解决了ACMA算法检测性能随着同步偏移量的变化而波动的缺点,提升了算法的检测性能。该算法良好的性能是在提高了算法复杂度的基础上实现的,针对该算法的高复杂度本文还提出了该算法的步进快速搜索方案,该方案可以在牺牲很小检测性能的情况下使检测时间缩小到原有检测时间的几分之一,步长与复杂度成线性关系,当步长增加时,检测所需要的时间降低,但检测性能下降。
[Abstract]:With the advent of the information age, people depend more and more on communication in their daily life and military operations. Direct-spread-sequence communication (DSSS) is widely used because of its good anti-interference and the information is not easily intercepted. Blind detection of DSSS signals in non-cooperative communication is the basis of parameter estimation and blind estimation of spread spectrum sequences, so blind detection of DSSS signals is paid great attention to. However, some of the previous detection algorithms are semi-blind, and some of them require a higher SNR than the DSSS signal to noise ratio (SNR), which can not meet the requirements of practical applications. On the basis of deep research and analysis of DSSS signals, two blind detection algorithms for DSSS signals in low SNR environment are proposed in this paper. Firstly, based on the analysis of the time-domain correlation detection algorithm, this paper improves the prediction-based Time-domain Sliding Correlating acceptance algorithm, which greatly reduces the in-band noise by updating the spread spectrum code and the estimated data. The simulation results show that the proposed algorithm can detect the DSSS signal when the SNR is -15dB for the 31-bit DSSS signal, and the performance will be further improved with the increase of the data length used in the detection. The DSSS signal generated by the vector signal source is used to verify the algorithm, and the results show that the algorithm has good detection performance. Secondly, after studying the eigenvalue decomposition algorithm, an autocorrelation-based Matrix analysis algorithm for intermediate frequency signals is proposed. Simulation results show that the detection performance of this algorithm is about 2 dB higher than that of ETSCA algorithm under the same conditions. The influence of synchronous offset of DSSS signal on the performance of ACMA algorithm is deduced theoretically. The performance of the algorithm is the best in the case of synchronization, but the worst when the normalized synchronization offset is 1 / 2, and the theoretical results are verified by simulation. Finally, an estimation based Autocorrelation Matrix analysis algorithm based on estimation is proposed by combining the two detection algorithms. The algorithm solves the shortcoming that the detection performance of ACMA algorithm fluctuates with the change of synchronous offset, and improves the detection performance of the algorithm. The good performance of the algorithm is realized on the basis of increasing the complexity of the algorithm. In view of the high complexity of the algorithm, this paper also proposes a step by step fast search scheme for the algorithm. The proposed scheme can reduce the detection time to a fraction of the original detection time at the expense of very small detection performance. The step size is linearly related to the complexity. When the step size increases, the detection time will decrease, but the detection performance will decline.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN914.42

【参考文献】