| 引用本文： |
-
秦国领,张铁茁,程艳合,等.基于稀疏系数位置和归一化残差的压缩感知信号检测[J].电讯技术，2016，56（10）： - . [点击复制]
- QIN Guoling,ZHANG Tiezhuo,CHENG Yanhe,et al.Compressed sensing signal detection based on sparse coefficient location and normalized residual[J].，2016，56（10）： - . [点击复制]
|
|
| 摘要: |
| 信号检测是压缩感知理论研究的重要内容。针对当前压缩感知信号检测算法没有充分利用稀疏系数幅值和位置信息的不足，提出了一种新的检测算法。该算法首先引入归一化残差变量，有效克服了稀疏系数幅值波动大的缺点；然后，利用不同测量矩阵确定的稀疏系数位置信息，基于正交匹配追踪（OMP）算法实现目标信号检测。实验结果表明，算法的检测性能随着信噪比的提高而增强，且与压缩比负相关，运算复杂度较正交匹配追踪算法和仅利用稀疏系数位置信息的算法相当但检测性能分别提高了4 dB和1 dB. |
| 关键词: 压缩感知 信号检测 稀疏系数特征 归一化残差 正交匹配追踪算法 |
| DOI： |
|
| 基金项目: |
|
| Compressed sensing signal detection based on sparse coefficient location and normalized residual |
| QIN Guoling,ZHANG Tiezhuo,CHENG Yanhe,WEI Shaojie |
| () |
| Abstract: |
| Signal detection is the key content of compressed sensing.Considering that current signal detection based on compressed sensing does not effectively use amplitude and location information of sparse coefficient,a new method is proposed. Based on orthogonal matching pursuit(OMP) algorithm,the algorithm introduces the normalized residual variables to overcome the shortcoming of amplitude fluctuates acutely of sparse coefficient,and then uses the different measurement matrix to determine the location information of sparse coefficient,and achieves target signal detection. Experiment results show that the detection performance of algorithm enhances with the improvement of signal-to-noise ratio(SNR),and negatively relates to compressed ratio. The computation complexity of algorithm is pretty and the detection performance improves 4 dB and 1 dB in comparison with OMP algorithm and the algorithm just using location information of sparse coefficient. |
| Key words: compressed sensing signal detection characteristic of sparse coefficient normalized residual orthogonal matching pursuit algorithm |
《电讯技术》编辑部 