基于结构化稀疏性的直接序列扩频信号压缩感知技术

项目名称： 基于结构化稀疏性的直接序列扩频信号压缩感知技术

项目编号： No.61301088

项目类型： 青年科学基金项目

立项/批准年度： 2014

项目学科： 无线电电子学、电信技术

项目作者： 王帅

作者单位： 北京理工大学

项目金额： 24万元

中文摘要： 宽带低谱密度直接序列扩频信号具备优异的抗干扰、抗侦听和抗截获能力，对同频段内的其他无线系统影响较小，频谱兼容性强。但是在目前基于软件无线电的接收机架构中，直扩系统的带宽和功率谱密度常常受制于前端模数转换器（ADC）的采样率。申请人的前期研究表明，直扩信号是典型的稀疏信号，且其稀疏表示中的非零元素具有分组等间隔分布的特征，呈现出一种高度结构化的稀疏性。受此启发，本课题研究基于结构化稀疏性的直扩信号压缩感知技术，其目标是探索以亚Nyquist采样率接收直扩信号的新理论与新方法。课题的主要研究内容包括基于结构化最小二乘搜索的低复杂度压缩域时频双选信道估计算法，和基于重复加权提升搜索的压缩域近最优信道均衡算法；为了确定噪声折叠效应对接收机性能的影响，还将分析压缩域信道参数估计均方误差的Cramér-Rao下界以及各类压缩域均衡器的理论误码率，并以此为依据探讨测量矩阵的选择与优化。

中文关键词： 压缩感知；直接序列扩频；信道估计；均衡；

英文摘要： Wide band low power spectral density Direct Sequence Spread Spectrum (DSSS) transmission has received intensive attention due to its exellent anti-interference, interception and reconnaissance capability. Furthermore, it produces less co-channel interference against other systemsoperating in the same band, and hence exhibits satisfactory bandwidth compatibility. Under the the current Software Defined Radio (SDR) receiver framework, the difficulty in rasing the sampling rate of the Analog-to-Digital Converter (ADC) has emerged as the main bottleneck for the DSSS system to further increase its bandwidth, and to lower its PSD. Our former studies have shown that DSSS signals are actually sparse, and its sparse expression posses unique structure, which we refer to as Structured Sparsity. Motivated by these results, the aim of current proposal is to present a Compressed Sensing (CS) assisted UL-PSD DSSS receiver which is capable of operating at sub-Nyquist rate, by exploiting the so-called Structured Sparsity of the DSSS signals. The topics we are going to explore include the low-complexity identification algorithms for doubly-selctive channels in the compressed domain based on Structured Least-Squares Search (SLSS), the channel equalization strategy based on Repeated Weighted Boosting Search (RWBS), as well as the pe

英文关键词： Compressed Sensing ；Direct-Sequence Spread Spectrum；Channel identification；Equalization；