基于稀疏信号的同步与信道估计技术研究
发布时间：2018-08-24 17:21
【摘要】：稀疏表示(Sparse Representation,SR)在不同领域都有很多的运用案例,在信号处理方面最主要的应用就是从大量的繁杂信号以不同的要求提取出重要的元素。稀疏表示理论放弃了香农采样定理(Shannon’s Sampling Theorm,SST)和奈奎斯特采样准则(Nyquist Sampling Law,NSL)的原始测量,采用更有效的采样率来测量原始采样,随后采用最优的重构算法进行采样重构。在压缩感知的背景下,假设了所有的信号时稀疏或近似足够稀疏,和主信号空间相比,可靠的信号集的大小在稀疏性的约束下极可能非常小,因此,基于稀疏表示的大量算法能有效地解决信号处理领域的信号重构和恢复问题。而且稀疏表示技术能节约大量的采样时间和采样存储,具有很大的优势和潜力。全球定位系统的同步过程是为了获取信号从定位器到定位卫星的传播时间而进行定位的技术。目前来说同步过程算法比较成熟,但如何进行更简单有效的同步过程依然是一个可研究的方向,尤其是在如今的各类智能家居以及带定位功能的微型设备的普及。信道估计技术历来都是通信领域的重要研究方向之一,在现在第五代移动通信技术研究方向偏向于多天线系统的趋势下,多天线系统下的信道估计研究是研究热点之一。本论文从信号的稀疏表示特性出发,分别研究了基于稀疏傅里叶变换的快速同步方法和多天线系统下的信道估计技术。第一章对同步技术和稀疏信道估计的背景以及研究现状进行了基本的概括。第二章主要研究了用于稀疏表示的范式最小化问题的建模分析。对不同范数进行介绍,对它们之间的联系进行分析,针对范式最小化问题的解决方案,着重介绍了基于贪心策略的稀疏重构算法。第三章首先对全球定位系统(Global Positioning System,GPS)同步过程进行阐述说明,建立同步模型;对稀疏傅里叶变换也进行详细的表述。针对稀疏傅里叶变换利用混叠操作降低傅里叶变换规模的过程的启示,借用到GPS同步过程以降低快速傅里叶变换(Fast Fourier Transformation,FFT)和快速反傅里叶变换(Inverse Fast Fourier Transform,IFFT)运算的规模,达到降低整个同步过程复杂度的目的。仿真结果也验证了快速GPS的同步时间复杂度上比传统的基于FFT的同步算法要小很多,具有实际运用的前景。第四章对多天线下的信道估计技术进行了研究,首先对多天线的信道进行分析,阐述了时域信道冲击响应的抽头稀疏的特点,针对这一特性,在多天线系统中天线之间具有相关性的情景下,在正交频分复用(Orthogonal Frequency Division Multiplexing,OFDM)系统中给出相关相关性良好的先导序列代和OFDM块中的导频结合的导频设计方案,利用得到的共同稀疏支持信息可以减少导频使用,增加系统的频谱利用率。最后对多天线系统相位利用旋转导频做信道估计进行研究,针对相关发射天线的信道相关性进行时域信道重要抽头的错开移位操作,在接收端利用已知的移位因子设计两个重要抽头分类算法并进行仿真验证。第五章对全文的研究内容进行了总结,同时指出了快速同步技术和相关天线信道估计依然需要进行的研究点和研究方向。
[Abstract]:Sparse Representation (SR) has many applications in different fields. The main application in signal processing is to extract important elements from a large number of complex signals with different requirements. The original measurement of Nyquist Sampling Law (NSL) uses a more efficient sampling rate to measure the original sample, and then uses an optimal reconstruction algorithm to reconstruct the sample. The lower pole may be very small, so a large number of algorithms based on sparse representation can effectively solve the problem of signal reconstruction and recovery in the field of signal processing. At present, the algorithm of synchronization process is relatively mature, but how to carry out more simple and effective synchronization process is still a research direction, especially in today's smart homes and the popularity of micro-devices with positioning function. Channel estimation technology has always been a communication collar. One of the important research directions in the domain is that the research direction of the fifth generation mobile communication technology is inclined to multi-antenna system, and the channel estimation in multi-antenna system is one of the research hotspots. Chapter 1 summarizes the background and research status of synchronization technology and sparse channel estimation. Chapter 2 mainly studies the modeling and analysis of the normal form minimization problem for sparse representation. In the third chapter, the synchronization process of Global Positioning System (GPS) is described and the synchronization model is established. The sparse Fourier transform is also described in detail. The aliasing operation is used to reduce Fourier transform for sparse Fourier transform. The enlightenment of the process of transforming scale is borrowed from the GPS synchronization process to reduce the operation scale of Fast Fourier Transform (FFT) and Inverse Fast Fourier Transform (IFFT) to reduce the complexity of the whole synchronization process. In the fourth chapter, the channel estimation technology under multi-antenna is studied. Firstly, the channel of multi-antenna is analyzed, and the characteristics of sparse time-domain channel impulse response taps are described. In the case of correlation, the pilot sequence generation with good correlation and the pilot combination in the OFDM block are given in the Orthogonal Frequency Division Multiplexing (OFDM) system. The common sparse support information can be used to reduce the pilot usage and increase the spectrum utilization of the system. Finally, the phase estimation of multi-antenna system using rotating pilot is studied. The important taps of time-domain channel are staggered and shifted according to the channel correlation of correlated transmit antennas. Two important tap classification algorithms are designed and verified by simulation at the receiver using known shift factors. The capacitance is summarized, and the research points and directions of fast synchronization technology and related antenna channel estimation are pointed out.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN911
本文编号：2201506
[Abstract]:Sparse Representation (SR) has many applications in different fields. The main application in signal processing is to extract important elements from a large number of complex signals with different requirements. The original measurement of Nyquist Sampling Law (NSL) uses a more efficient sampling rate to measure the original sample, and then uses an optimal reconstruction algorithm to reconstruct the sample. The lower pole may be very small, so a large number of algorithms based on sparse representation can effectively solve the problem of signal reconstruction and recovery in the field of signal processing. At present, the algorithm of synchronization process is relatively mature, but how to carry out more simple and effective synchronization process is still a research direction, especially in today's smart homes and the popularity of micro-devices with positioning function. Channel estimation technology has always been a communication collar. One of the important research directions in the domain is that the research direction of the fifth generation mobile communication technology is inclined to multi-antenna system, and the channel estimation in multi-antenna system is one of the research hotspots. Chapter 1 summarizes the background and research status of synchronization technology and sparse channel estimation. Chapter 2 mainly studies the modeling and analysis of the normal form minimization problem for sparse representation. In the third chapter, the synchronization process of Global Positioning System (GPS) is described and the synchronization model is established. The sparse Fourier transform is also described in detail. The aliasing operation is used to reduce Fourier transform for sparse Fourier transform. The enlightenment of the process of transforming scale is borrowed from the GPS synchronization process to reduce the operation scale of Fast Fourier Transform (FFT) and Inverse Fast Fourier Transform (IFFT) to reduce the complexity of the whole synchronization process. In the fourth chapter, the channel estimation technology under multi-antenna is studied. Firstly, the channel of multi-antenna is analyzed, and the characteristics of sparse time-domain channel impulse response taps are described. In the case of correlation, the pilot sequence generation with good correlation and the pilot combination in the OFDM block are given in the Orthogonal Frequency Division Multiplexing (OFDM) system. The common sparse support information can be used to reduce the pilot usage and increase the spectrum utilization of the system. Finally, the phase estimation of multi-antenna system using rotating pilot is studied. The important taps of time-domain channel are staggered and shifted according to the channel correlation of correlated transmit antennas. Two important tap classification algorithms are designed and verified by simulation at the receiver using known shift factors. The capacitance is summarized, and the research points and directions of fast synchronization technology and related antenna channel estimation are pointed out.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN911
【参考文献】
相关硕士学位论文 前1条
1 王明明;GPS软件接收机基带算法研究[D];山东大学;2006年
，本文编号：2201506
本文链接：https://www.wllwen.com/kejilunwen/xinxigongchenglunwen/2201506.html
