基于压缩感知的跳频信号检测与参数估计技术
发布时间：2018-09-17 13:39
【摘要】：随着军事通信技术的快速发展,跳频通信信号趋向于大带宽、高载波,导致针对其的电子侦察面临着高采样率,海量数据存储及处理的压力,因此传统的基于奈奎斯特采样定理的跳频信号检测与估计技术面对这类问题时有着很大的缺陷,以现有的器件水平在具体的实现中会很困难或不可实现。压缩感知(compressed sensing)是一种利用信号稀疏特性而可进行欠采样观测的信号处理方式,通过它可以以接近信号携带有用信息的采样频率完成对宽带稀疏信号的采样,相对Nyquist采样频率,此时的采样频率会远远低于它。本文针对跳频信号大带宽范围采样及数据处理的问题,将压缩感知应用到跳频信号的检测与估计技术上,详细分析了针对非合作宽带高速跳频信号的欠采样处理以及在此基础上后续的信号检测与参数技术,给出了完整的系统设计方案、原理分析和仿真验证。本文的主要内容有以下几个方面:(1)以宽带调制转换器为基础,完成了针对非合作宽带跳频信号的欠采样接收技术研究,分析了跳频信号的分段稀疏表示,欠采样系统、重构系统,然后在原MWC的基础上,提出了一种基于带通采样模型的宽带调制转换器,其相对原来的基于低通模型的宽带调制转换器,有效地减少了系统中伪随机码片序列的符号交替速率,降低了系统实现的难度。最后从通道数、码片序列的随机特性以及重构算法三个方面分析了它们对系统性能的影响。(2)使用系统在欠采样条件下完成了对跳频信号的检测,并且在有定频干扰信号存在的情况下使用一种子空间投影法抑制住定频干扰;在欠采样后重构跳频信号与非重构跳频信号两种情况下,研究了针对跳频信号的参数估计技术,为欠采样后非重构跳频参数估计提供了新的研究思路。
[Abstract]:With the rapid development of military communication technology, frequency-hopping communication signals tend to have large bandwidth and high carrier, which leads to the pressure of high sampling rate, mass data storage and processing for electronic reconnaissance. Therefore, the traditional frequency-hopping signal detection and estimation technology based on Nyquist sampling theorem has great defects in the face of this kind of problem, and it will be difficult or impossible to implement in concrete implementation with the existing device level. Compressed sensing (compressed sensing) is a signal processing method which can be used for under-sampling and observation by using the sparse characteristic of signal. It can sample wideband sparse signal at the sampling frequency close to the signal carrying useful information, relative to the sampling frequency of Nyquist. At this point, the sampling frequency will be much lower than it. In order to solve the problem of sampling and data processing of frequency-hopping signals in large bandwidth, the compressed sensing is applied to the detection and estimation of frequency-hopping signals. The under-sampling processing of the non-cooperative broadband high-speed frequency hopping signal and the subsequent signal detection and parameter technology are analyzed in detail. The complete system design scheme, principle analysis and simulation verification are given. The main contents of this paper are as follows: (1) based on the broadband modulation converter, the undersampling and receiving techniques for the non-cooperative broadband frequency-hopping signals are studied, and the piecewise sparse representation and under-sampling systems of the frequency-hopping signals are analyzed. Then, based on the original MWC, a broadband modulation converter based on bandpass sampling model is proposed, which is compared with the original broadband modulation converter based on low-pass model. The symbol alternation rate of pseudorandom sequence in the system is reduced effectively, and the difficulty of system implementation is reduced. Finally, the effects of channel number, random characteristics of code sequence and reconstruction algorithm on the system performance are analyzed. (2) the frequency hopping signal detection is completed under the condition of under-sampling. In the presence of fixed frequency interference signal, a subspace projection method is used to suppress the fixed frequency interference, and the parameter estimation technique for frequency hopping signal is studied in the case of reconstructing frequency hopping signal and non reconfigurable frequency hopping signal after under-sampling. It provides a new research idea for the estimation of non-reconfigurable FH parameters after undersampling.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN911.23
本文编号：2246091
[Abstract]:With the rapid development of military communication technology, frequency-hopping communication signals tend to have large bandwidth and high carrier, which leads to the pressure of high sampling rate, mass data storage and processing for electronic reconnaissance. Therefore, the traditional frequency-hopping signal detection and estimation technology based on Nyquist sampling theorem has great defects in the face of this kind of problem, and it will be difficult or impossible to implement in concrete implementation with the existing device level. Compressed sensing (compressed sensing) is a signal processing method which can be used for under-sampling and observation by using the sparse characteristic of signal. It can sample wideband sparse signal at the sampling frequency close to the signal carrying useful information, relative to the sampling frequency of Nyquist. At this point, the sampling frequency will be much lower than it. In order to solve the problem of sampling and data processing of frequency-hopping signals in large bandwidth, the compressed sensing is applied to the detection and estimation of frequency-hopping signals. The under-sampling processing of the non-cooperative broadband high-speed frequency hopping signal and the subsequent signal detection and parameter technology are analyzed in detail. The complete system design scheme, principle analysis and simulation verification are given. The main contents of this paper are as follows: (1) based on the broadband modulation converter, the undersampling and receiving techniques for the non-cooperative broadband frequency-hopping signals are studied, and the piecewise sparse representation and under-sampling systems of the frequency-hopping signals are analyzed. Then, based on the original MWC, a broadband modulation converter based on bandpass sampling model is proposed, which is compared with the original broadband modulation converter based on low-pass model. The symbol alternation rate of pseudorandom sequence in the system is reduced effectively, and the difficulty of system implementation is reduced. Finally, the effects of channel number, random characteristics of code sequence and reconstruction algorithm on the system performance are analyzed. (2) the frequency hopping signal detection is completed under the condition of under-sampling. In the presence of fixed frequency interference signal, a subspace projection method is used to suppress the fixed frequency interference, and the parameter estimation technique for frequency hopping signal is studied in the case of reconstructing frequency hopping signal and non reconfigurable frequency hopping signal after under-sampling. It provides a new research idea for the estimation of non-reconfigurable FH parameters after undersampling.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN911.23
【参考文献】
相关期刊论文 前2条
1 王丰华;沙志超;刘章孟;黄知涛;;基于贝叶斯稀疏学习的多跳频信号频率跟踪方法[J];电子与信息学报;2013年06期
2 王建雄;张立民;张媛;夏沭涛;;基于循环平稳特性的跳频信号盲检测算法[J];舰船电子工程;2011年09期
，本文编号：2246091
本文链接：https://www.wllwen.com/kejilunwen/wltx/2246091.html
