基于周期性非均匀采样的独立分量分析在宽带频谱感知中的应用研究

发布时间：2018-03-19 04:34

本文选题：认知无线电　切入点：宽带频谱感知　出处：《大连海事大学》2017年硕士论文　论文类型：学位论文

【摘要】：多功能无线电通信系统的出现与快速发展使频谱资源变得越来越紧缺。认知无线电作为一种可以实现动态通信接入的无线技术,能够持续感知周围的频谱利用情况并灵活地将频谱空穴分配给二级用户进行无线通信,既节省了开辟新的频谱资源的开销,又为缓解无线通信所面临的频谱资源压力提供了解决思路。认知无线电在进行宽带频谱感知时,由于感知频段宽且事先未知信号的具体频谱位置,如果利用均匀采样采集信号信息需要的采样率至少为频段宽度的两倍,会极大地增加硬件实现的难度。所以,如何降低系统采样率是宽带频谱感知急需处理的重要问题。利用周期性非均匀采样可以用远远低于奈奎斯特采样定理限制的采样率采集并重建原始信号信息。本文对基于周期性非均匀采样独立分量分析的宽带频谱感知问题展开研究,主要工作如下:(1)针对宽带频谱感知对采样率要求高、硬件实现困难的问题探索周期性非均匀采样应用于宽带频谱感知的现实必要性与优势。首先分析周期性非均匀采样的基本理论,详细介绍采样的实现方法和最优采样序列的选取原则。然后通过对离散的非均匀样本值进行功率谱密度估计来感知宽带范围内主用户信号的活动状态,找出频谱空穴。(2)针对周期性非均匀采样条件下发射功率各不相同的主用户信号相互混合时无法判断低功率信号频谱位置的情况,引入以独立分量分析为基础的盲源分离算法分离出混合信号中的独立成分。首先分析不同算法在周期性非均匀采样条件下的分离性能,仿真验证基于独立分量分析的宽带频谱感知设计方案,并通过非基带信号的最小能量重建判断出各主用户信号的频谱位置。(3)针对多径干扰、多普勒频移等环境下宽带频谱感知性能降低的问题,结合瑞利衰落信道模型探索基于双通道周期性非均匀采样的独立分量分析在改善信号分离效果方面的作用并进行仿真对比分析双通道采样下频谱感知的有效性。
[Abstract]:The emergence and rapid development of multifunctional radio communication systems make spectrum resources more and more scarce. Cognitive radio is a wireless technology that can realize dynamic communication access. The ability to continuously perceive the surrounding spectrum utilization and flexibly allocate spectrum holes to secondary users for wireless communication saves the overhead of opening up new spectrum resources. It also provides a solution to alleviate the pressure of spectrum resources in wireless communication. When cognitive radio is performing broadband spectrum sensing, it can perceive the frequency band wide and the specific spectrum position of the unknown signal is unknown in advance. If the sampling rate required by uniform sampling is at least twice the width of the frequency band, it will greatly increase the difficulty of hardware implementation. How to reduce the sampling rate of the system is an important problem for wideband spectrum sensing, which can be collected and reconstructed at a sampling rate far below the limit of Nyquist sampling theorem by using periodic non-uniform sampling. In this paper, the problem of wideband spectrum sensing based on periodic nonuniform sampling independent component analysis (ICA) is studied. The main work is as follows: 1) aiming at wideband spectrum sensing, high sampling rate is required. The problem of hardware implementation is difficult to explore the practical necessity and advantage of applying periodic non-uniform sampling to wideband spectrum sensing. Firstly, the basic theory of periodic non-uniform sampling is analyzed. The realization method of sampling and the selection principle of optimal sampling sequence are introduced in detail, and then the active state of the main user signal in wideband range is sensed by estimating the power spectral density of discrete non-uniform sample values. Find out the spectrum hole. 2) when the main user signals with different transmitting power are mixed with each other under the condition of periodic non-uniform sampling, it is impossible to judge the frequency spectrum position of the low power signal. A blind source separation algorithm based on independent component analysis (ICA) is introduced to separate the independent components from the mixed signals. Firstly, the separation performance of different algorithms under the condition of periodic non-uniform sampling is analyzed. Simulation verifies the design scheme of wideband spectrum sensing based on independent component analysis, and determines the spectrum position of each primary user signal by the minimum energy reconstruction of the non-baseband signal. The problem of low performance of wideband spectrum sensing in environment such as Doppler shift, Combined with Rayleigh fading channel model, this paper explores the effect of independent component analysis (ICA) based on double channel periodic nonuniform sampling on improving the signal separation effect, and compares and analyzes the effectiveness of spectrum sensing under dual channel sampling.
【学位授予单位】：大连海事大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN925

【相似文献】