| 摘要: |
| 针对多输入多输出（MIMO）系统在双选信道下信道估计问题，以及挖掘信道在时延域和角度域的联合稀疏特性，提出了一种新的基于压缩感知的联合稀疏信道估计方案.首先，基于基扩展模型，将信道估计建模为结构化压缩感知问题，随后基于压缩感知模型，提出了两种新的贪婪算法，有效地恢复了时变信道参数.其中两步同时正交匹配追踪（TS-SOMP）算法先在时延域中找到所有非零抽头位置，然后估计非零角度域系数.两环同时正交匹配追踪（TL-SOMP）算法包括内外两个循环，在外部循环中找到一个非零抽头位置后，即可直接在内部循环求解非零角度域系数.最后，给出了归一化均方误差（NMSE）的仿真曲线，验证了本算法的有效性. |
| 关键词: 信道估计 压缩感知 双选 系统 角度域 |
| DOI：10.3969/J.ISSN.1000-5137.2018.02.009 |
| 分类号:TN929.5 |
| 基金项目:国家自然科学基金（61471236，61420106008，61671295）；111计划（B07022）；上海市数字媒体处理与传输重点实验室；上海浦江人才计划（16PJD029） |
|
| Joint sparse channel estimation based on angle domain and delay domain |
|
Zhang Yueming1, Zhang Bingshan2, Gui Lin1, Qin Qibo1, Xiong Jian1
|
|
1.School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China;2.Beijing Institute of Tracking and Telecommunications Technology, Beijing 100094, China
|
| Abstract: |
| For the channel estimation problem of multiple input multiple output(MIMO) systems in doubly-selective (DS) channels and the joint sparse characteristic in the delay domain and angle domain,a new joint sparse channel estimation scheme based on compressive sensing is proposed.Firstly,based on the basic extended model,the channel estimation is modeled as a structured compressive sensing problem.Then,based on the compressed sensing model,two novel greedy algorithms are developed to effectively recover the time-varying channel parameters.The two-stage simultaneous orthogonal matching pursuit (TS-SOMP) algorithm first finds all nonzero tap positions in the delay domain and then estimates nonzero angle domain coefficients.The two-loop simultaneous orthogonal matching pursuit (TL-SOMP) algorithm includes two loops inside and outside.It estimates nonzero angle domain coefficients in the inner loop once it finds one nonzero tap position in the outer loop.Finally,a simulation curve of normalized mean squared error(NMSE) is given to verify the effectiveness of the proposed algorithm. |
| Key words: channel estimation compressive sensing doubly-selective system angle domain |
