在浅海环境中，由于受到混响的影响，主动声呐接收到的信号混淆不清。针对上述问题，提出一种基于稀疏表示的非负矩阵分解(non-negative matrix factorization, NMF)抗混响方法。利用稀疏表示方法处理主动声呐回波信号，然后根据信号的稀疏性，构建基于非负矩阵分解的Kullback-Leibler（KL）问题，通过梯度下降法给出迭代规则，进而得到了目标信号矩阵中的协方差估计。仿真结果表明，相对其他去混响方法，该方法能够有效抑制混响，提高对水下目标的识别率。
In the shallow sea environment, due to the influence of the reverberation, the signal received by the active sound was confusing. In response to the above issues, the article proposes a method of non-negative matrix decomposition based on sparsely expressed sparse expressions (NMF). The above method uses the sparse representation method to process the active sound echo signal, and then builds the Kullback-Leibler (KL) problem based on the sparseness of the signal. The collaborative difference estimation in the signal matrix. The simulation results show that relative to other reverberation methods, this method can effectively suppress the reverberation and improve the recognition rate of underwater targets.
2023,45(17): 129-134 收稿日期：2022-06-09
DOI：10.3404/j.issn.1672-7649.2023.17.025
分类号：U666.7
基金项目：国家自然科学基金资助项目(61901079)
作者简介：李青(1998-),女,硕士研究生,研究方向为水下目标探测与识别
参考文献：
[1] 李玉强, 徐燕, 周胜增. 海洋混响特性分析与建模仿真研究[J]. 舰船电子工程, 2018, 38(11): 86–88+182
[2] SHU Xianglan, SUN Rongguang, MA Xin. Echo detection of LFM signal under strong reverberation background[J]. Telecommunication engineering., 2016, 56(1): 82–87
[3] 陈益明. 混响背景下混合块归一化匹配滤波算法[C]//. 2013中国西部声学学术交流会论文集(下), 2013: 202-204.
[4] 许策, 章新华, 钱海民. 主动声呐中混响干扰的一种抑制方法[C]//. 2009年全国水声学学术交流暨水声学分会换届改选会议论文集. , 2009: 243-245.
[5] 刘贯领, 凌国民, 严琪. 主动声呐检测技术的回顾与展望[J]. 声学技术, 2007(2): 335–340
[6] ZHU G, SONG Z, YIN J, et al. Extracting target highlight feature based on low-rank matrix recovery in reverberation background[J]. Acta Acustica, 2019, 4: 471–479
[7] WRIGHT J, GANESH A, RAO S, et al. Robust principal component analysis: Exact recovery of corrupted low-rank matrices via convex optimization[J]. Advances in neural information processing systems, 2009: 22
[8] QAYYUM A, MALIK A S, SAAD M N M, et al. Evaluation of overcomplete dictionaries for image inpainting[J]. Science International, 2016, 28(4).
[9] LI Junlin, WANG Huaqing, SONG Liuyang, et al. A novel feature extraction method for roller bearing using sparse decomposition based on self-Adaptive complete dictionary[J]. Measurement, 2019, 148(C).
[10] VESHKI F G, VOROBYOV S A, A fast dictionary learning method for coupled feature space learning. [J]. CoRR, 2019, abs/1904.06968.
[11] MA Xiaoyu, AHENG Jinsheng, LI Ting, et al. Super-Resolution geomagnetic reference map reconstruction based on dictionary learning and sparse representation[J]. IEEE Access, 2020: 8
[12] 何冲, 王冬霞, 王旭东, 等. 一种基于正交非负矩阵分解的多通道线性预测语音去混响方法[J]. 声学技术, 2018, 37(5): 468–474
[13] 高馨. 基于NMF的SAR图像目标识别方法研究[D]. 成都; 电子科技大学, 2013.
[14] 何天远. 基于信号稀疏表示的轴承故障诊断方法研究[D]. 石家庄: 石家庄铁道大学, 2020.000317.
[15] 李振, 李伟光, 陈辉, 等. 基于匹配追踪的特征频率提取算法及其应用[J]. 振动与冲击, 2019, 38(19): 7–13
[16] 王林. 非凸正则化稀疏表示方法及其旋转机械故障诊断应用研究[D]. 苏州: 苏州大学, 2019.
[17] 韩卫丽, 邹建成, 李建伟. 一种基于信号稀疏表示的语音去噪新方法[J]. 北方工业大学学报, 2013, 25(3): 6–11
[18] 韩卫丽. 信号冗余表示理论在声音修复中的应用研究[D]. 北京: 北方工业大学, 2014.
[19] MICHAL A, MICHEAL E, ALFRED B. K-SVD: an algorithm for designing overcomplete dictionaries for sparse representation[J]. IEEE Transactions on Signal Processing: A publication of the IEEE Signal Processing Society, 2006, 54(11).
