HUANG Xiaojie, DING Jinhua, WANG Daqing. Spatiotemporal evolution and regulation strategies of ecological risks in green space landscape in the water network area of southern Jiangsu[J]. Journal of Zhejiang A&F University, 2024, 41(6): 1283-1292. DOI: 10.11833/j.issn.2095-0756.20240169
Citation: DING Hao, FU Donglin, ZHANG Ruizhi. Research advance on noise source identification methods of deconvolution beamforming[J]. Journal of Zhejiang A&F University, 2018, 35(2): 376-379. DOI: 10.11833/j.issn.2095-0756.2018.02.024

Research advance on noise source identification methods of deconvolution beamforming

DOI: 10.11833/j.issn.2095-0756.2018.02.024
  • Received Date: 2017-04-25
  • Rev Recd Date: 2017-09-21
  • Publish Date: 2018-04-20
  • Deconvolution beamforming as a noise source identification method based on beamforming has been widely used, especially in the identification of forestry machine noise. Compared with the conventional beamforming, deconvolution beamforming method is more effective to reduce the width of the main lobe and eliminate the side lobe, and improve the spatial resolution. This paper reviewed the recent research progress on deconvolution beamforming at home and abroad. A comparative analysis was made among several representative methods on the aspects of the side lobe suppression ability, positioning accuracy and computational efficiency. Then the characteristics and limitations of those typical deconvolution beamforming algorithms were discussed, which provided a new research idea for future improvement in order to obtain a more comprehensive deconvolution beamforming algorithm with better applicability.
  • [1] WANG Jie, HE Wenchuang, XIANG Kunli, WU Zhiqiang, GU Cuihua.  Advances in plant phylogeny in the genome era . Journal of Zhejiang A&F University, 2023, 40(1): 227-236. doi: 10.11833/j.issn.2095-0756.20220313
    [2] WANG Lu, LI Lele, LAI Mengxia, DU Changxia, FAN Huaifu.  Research progress on the causes of spatial heterogeneity of soil salinity and its effects on plants’ growth . Journal of Zhejiang A&F University, 2022, 39(6): 1369-1377. doi: 10.11833/j.issn.2095-0756.20220155
    [3] XU Sen, YANG Liting, CHEN Shuanglin, GUO Ziwu, GU Rui, ZHANG Chao.  Review on the formation of bamboo shoot palatability and its main influencing factors . Journal of Zhejiang A&F University, 2021, 38(2): 403-411. doi: 10.11833/j.issn.2095-0756.20200400
    [4] TONG Liang, LI Pingheng, ZHOU Guomo, ZHOU Yufeng, LI Chong.  A review of research about rhizome-root system in bamboo forest . Journal of Zhejiang A&F University, 2019, 36(1): 183-192. doi: 10.11833/j.issn.2095-0756.2019.01.023
    [5] ZHANG Jie, YIN Dejie, GUAN Haiyan, QU Qiqi, DONG Li.  An overview of Sedum spp. Research . Journal of Zhejiang A&F University, 2018, 35(6): 1166-1176. doi: 10.11833/j.issn.2095-0756.2018.06.022
    [6] WEI Wei, GUO Jialian, WAN Lintao, XU Linfeng, DING Mingquan, ZHOU Wei.  Research progress on molecular regulation mechanism of grain weight formation in wheat . Journal of Zhejiang A&F University, 2016, 33(2): 348-356. doi: 10.11833/j.issn.2095-0756.2016.02.022
    [7] WU Xue, DU Changxia, YANG Bingbing, FAN Huaifu.  Research progress in plant aquaporins . Journal of Zhejiang A&F University, 2015, 32(5): 789-796. doi: 10.11833/j.issn.2095-0756.2015.05.020
    [8] YE Lisha, CHEN Shuanglin, GUO Ziwu.  Research on nitrogen circulation and management of bamboo: a review . Journal of Zhejiang A&F University, 2015, 32(4): 635-642. doi: 10.11833/j.issn.2095-0756.2015.04.021
    [9] FU Jianguo, LIU Jinliang, YANG Xiaojun, AN Yulin, LUO Jiayan.  A review of wood identification based on molecular biology technologies . Journal of Zhejiang A&F University, 2013, 30(3): 438-443. doi: 10.11833/j.issn.2095-0756.2013.03.022
    [10] ZHANG Tao, LI Yongfu, JIANG Peikun, ZHOU Guomo, LIU Juan.  Research progresses in the effect of land-use change on soil carbon pools and soil respiration . Journal of Zhejiang A&F University, 2013, 30(3): 428-437. doi: 10.11833/j.issn.2095-0756.2013.03.021
    [11] CHENG Jian-zhong, YANG Ping, GUI Ren-yi.  Research progress on speciation of selenium compounds in plants . Journal of Zhejiang A&F University, 2012, 29(2): 288-395. doi: 10.11833/j.issn.2095-0756.2012.02.020
    [12] CHENG Ying, LI Gen-you, XIA Guo-hua, HUANG Shang-jue, HUANG Yu-feng.  Review on tissue culture of Aralia plants . Journal of Zhejiang A&F University, 2011, 28(6): 968-972. doi: 10.11833/j.issn.2095-0756.2011.06.022
    [13] ZHANG Hui-ling, SONG Xin-zhang, AI Jian-guo, JIANG Hong, YU Shu-quan.  A review of UV-B radiation and its influence on litter decomposition . Journal of Zhejiang A&F University, 2010, 27(1): 134-142. doi: 10.11833/j.issn.2095-0756.2010.01.022
    [14] GONG Zhi-wen, KANG Xin-gang, GU Li, ZHAO Jun-hui, ZHENG Yan-feng, YANG Hua.  Research methods on natural forest stand structure:a review . Journal of Zhejiang A&F University, 2009, 26(3): 434-443.
    [15] WANG Hang-jun, ZHANG Guang-qun, QI Heng-nian, LI Wen-zhu.  A review of research on wood recognition technology . Journal of Zhejiang A&F University, 2009, 26(6): 896-902.
    [16] LUO Xian-xian, KANG Xin-gang.  Progress in research on the comprehensive monitoring of forest resources . Journal of Zhejiang A&F University, 2008, 25(6): 803-809.
    [17] SHI Qiang, LIAO Ke, ZHONG Lin-sheng.  A review of the effects of tourists'activities on vegetation . Journal of Zhejiang A&F University, 2006, 23(2): 217-223.
    [18] DAI Jian-bing, YU Yi-wu, CAO Qun.  Summary of research on wetland protection and management . Journal of Zhejiang A&F University, 2006, 23(3): 328-333.
    [19] BAI Jiang-li, PENG Dao-li, YU Xiao-hong.  Research progress of the restoration and reconstruction of degraded ecosystems . Journal of Zhejiang A&F University, 2005, 22(4): 458-463.
    [20] WANG Xiao-de, MA Jin.  Progress and outlook of landscape cover plants . Journal of Zhejiang A&F University, 2003, 20(4): 419-423.
  • [1]
    CHEN Xinzhao. Progress of techniques for noise source identification[J]. J Hefei Univ Technol Nat Sci, 2009, 32(5):609-614.
    [2]
    YARDIBI T, BAHR C, ZAWODNY N, et al. Uncertainty analysis of the standard delay-and-sum beamformer and array calibration[J]. J Sound Vibr, 2010, 329(13):2654-2682.
    [3]
    YANG Yang, CHU Zhigang, JIANG Hong, et al. Research on DAMAS2 beamforming sound source identification[J]. Chin J Sci Instrum, 2013, 34(8):1779-1786.
    [4]
    BROOKS T F, HUMPHREYS W M. A deconvolution approach for the mapping of acoustic sources (DAMAS) determined from phased microphone arrays[J]. J Sound Vibr, 2006, 294(4):856-879.
    [5]
    YARDIBI T, ZAWODNY N S, BAHR C, et al. Comparison of microphone array processing techniques for aeroacoustic measurements[J]. Int J Aeroacoust, 2010, 9(6):733-762.
    [6]
    YANG Yang, CHU Zhigang. Weak position identification of sound insulation for car dash panel based on CLEAN-SC clearness beamforming[J]. Tech Acoust, 2015, 34(5):449-456.
    [7]
    CHU Zhigang, YANG Yang. Comparison of deconvolution methods for the visualization of acoustic sources based on cross-spectral imaging function beamforming[J]. Mech Syst Sign Proc, 2014, 48(1/2):404-422. doi:org/10.1016/j.ymssp.2014.03.012.
    [8]
    XENAKI A, JACOBSEN F, TIANA-ROIG E, et al. Improving the resolution of beamforming measurements on wind turbines[C]//ICA. Proceedings of 20th International Congress on Acoustics. Sydney: ICA, 2010: 23-27.
    [9]
    RAMACHANDRAN R C, RAMAN G, DOUGHERTY R P. Wind turbine noise measurement using a compact microphone array with advanced deconvolution algorithms[J]. J Sound Vibr, 2014, 333(14):3058-3080.
    [10]
    CHU Ning, PICHERAL J, MOHAMMAD-DJAFARI A. A robust super-resolution approach with sparsity constraint for near-field wideband acoustic imaging[C]//IEEE Computer Socirty. ISSPIT'11 Proceedings of the 2011 IEEE International Symposium on Signal Processing and Information Technology. Washington D C: IEEE Computer Society, 2011: 310-315. doi: 10.1109/ISSPIT.2011.6151579.
    [11]
    DOUGHERTY R. Extensions of DAMAS and benefits and limitations of deconvolution in beamforming[C]//AIAA Association. 11th AIAA/CEAS Aeroacoustics Conference. Monterey: AIAA Association, 2005. doi: org/10.2514/6.2005-2961.
    [12]
    EHRENFRIED K, KOOP L. Comparison of iterative deconvolution algorithms for the mapping of acoustic sources[J]. AIAA J, 2007, 45(7):1584-1595. doi:org/10.2514/1.26320.
    [13]
    YARDIBI T, LI Jian, STOICA P, et al. Sparsity constrained deconvolution approaches for acoustic source mapping[J]. J Acoust Soc Am, 2008, 123(5):2631-2642.
    [14]
    SIJTSMA P. CLEAN based on spatial source coherence[J]. Int J Aeroacoust, 2007, 6(4):357-374.
    [15]
    RÜBSAMEN M, PESAVENTO M. Steering vector non-identifiability in covariance matrix fitting based beamforming[C]//IEEE Computer Socirty. Proceedings of the IEEE 7th Sensor Array and Multichannel Signal Processing Workshop (SAM). Hoboken: IEEE Computer Socirty, 2012: 433-436.
    [16]
    LIU Peng, LIU Zhihong, LI He, et al. Study on the influencing factors of microphone array visualization[J]. J Test Measyr Technol, 2016, 30(3):231-235.
  • Created with Highcharts 5.0.7Amount of accessChart context menuAbstract Views, HTML Views, PDF Downloads StatisticsAbstract ViewsHTML ViewsPDF Downloads2024-052024-062024-072024-082024-092024-102024-112024-122025-012025-022025-032025-040102030405060Highcharts.com
    Created with Highcharts 5.0.7Chart context menuAccess Class DistributionFULLTEXT: 16.0 %FULLTEXT: 16.0 %META: 80.4 %META: 80.4 %PDF: 3.6 %PDF: 3.6 %FULLTEXTMETAPDFHighcharts.com
    Created with Highcharts 5.0.7Chart context menuAccess Area Distribution其他: 10.3 %其他: 10.3 %Macungie: 0.2 %Macungie: 0.2 %Seattle: 1.4 %Seattle: 1.4 %Wan Chai: 0.5 %Wan Chai: 0.5 %上海: 1.0 %上海: 1.0 %北京: 1.7 %北京: 1.7 %十堰: 0.5 %十堰: 0.5 %南京: 0.2 %南京: 0.2 %南充: 0.2 %南充: 0.2 %南通: 1.2 %南通: 1.2 %博阿努瓦: 0.7 %博阿努瓦: 0.7 %台州: 0.7 %台州: 0.7 %哈尔滨: 0.5 %哈尔滨: 0.5 %哥伦布: 0.2 %哥伦布: 0.2 %嘉兴: 0.5 %嘉兴: 0.5 %大连: 0.7 %大连: 0.7 %天津: 1.4 %天津: 1.4 %宣城: 0.2 %宣城: 0.2 %密蘇里城: 0.2 %密蘇里城: 0.2 %常州: 0.5 %常州: 0.5 %常德: 0.2 %常德: 0.2 %广州: 1.4 %广州: 1.4 %张家口: 4.3 %张家口: 4.3 %成都: 0.5 %成都: 0.5 %扬州: 1.7 %扬州: 1.7 %昆明: 0.2 %昆明: 0.2 %杭州: 2.6 %杭州: 2.6 %武汉: 0.2 %武汉: 0.2 %沈阳: 0.2 %沈阳: 0.2 %洛杉矶: 0.2 %洛杉矶: 0.2 %淄博: 0.2 %淄博: 0.2 %深圳: 2.4 %深圳: 2.4 %温州: 0.7 %温州: 0.7 %湖州: 1.2 %湖州: 1.2 %湘潭: 0.2 %湘潭: 0.2 %漯河: 5.0 %漯河: 5.0 %石家庄: 3.3 %石家庄: 3.3 %绍兴: 0.5 %绍兴: 0.5 %芒廷维尤: 17.9 %芒廷维尤: 17.9 %芝加哥: 3.1 %芝加哥: 3.1 %苏州: 2.6 %苏州: 2.6 %衢州: 0.2 %衢州: 0.2 %西宁: 15.0 %西宁: 15.0 %运城: 0.7 %运城: 0.7 %遵义: 0.2 %遵义: 0.2 %邯郸: 0.2 %邯郸: 0.2 %郑州: 11.0 %郑州: 11.0 %都伯林: 0.2 %都伯林: 0.2 %长沙: 0.5 %长沙: 0.5 %其他MacungieSeattleWan Chai上海北京十堰南京南充南通博阿努瓦台州哈尔滨哥伦布嘉兴大连天津宣城密蘇里城常州常德广州张家口成都扬州昆明杭州武汉沈阳洛杉矶淄博深圳温州湖州湘潭漯河石家庄绍兴芒廷维尤芝加哥苏州衢州西宁运城遵义邯郸郑州都伯林长沙Highcharts.com
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Tables(2)

Article views(3272) PDF downloads(461) Cited by()

Related
Proportional views

Research advance on noise source identification methods of deconvolution beamforming

doi: 10.11833/j.issn.2095-0756.2018.02.024

Abstract: Deconvolution beamforming as a noise source identification method based on beamforming has been widely used, especially in the identification of forestry machine noise. Compared with the conventional beamforming, deconvolution beamforming method is more effective to reduce the width of the main lobe and eliminate the side lobe, and improve the spatial resolution. This paper reviewed the recent research progress on deconvolution beamforming at home and abroad. A comparative analysis was made among several representative methods on the aspects of the side lobe suppression ability, positioning accuracy and computational efficiency. Then the characteristics and limitations of those typical deconvolution beamforming algorithms were discussed, which provided a new research idea for future improvement in order to obtain a more comprehensive deconvolution beamforming algorithm with better applicability.

HUANG Xiaojie, DING Jinhua, WANG Daqing. Spatiotemporal evolution and regulation strategies of ecological risks in green space landscape in the water network area of southern Jiangsu[J]. Journal of Zhejiang A&F University, 2024, 41(6): 1283-1292. DOI: 10.11833/j.issn.2095-0756.20240169
Citation: DING Hao, FU Donglin, ZHANG Ruizhi. Research advance on noise source identification methods of deconvolution beamforming[J]. Journal of Zhejiang A&F University, 2018, 35(2): 376-379. DOI: 10.11833/j.issn.2095-0756.2018.02.024
  • 噪声水平是衡量机械产品的重要环保指标,欧美国家建立了大量的相关技术标准作为市场准入的环保条件,这对国产林用机械产品的出口形成了无形的壁垒。为降低机械噪声水平,目前主要是通过声源识别定位方法进行噪声源识别定位,然后采取针对性手段降噪。波束形成法具有计算快、测量方便等优点,对静止或运动的中高频、远距离声源都有很好的识别能力[1]。传统波束形成法(delay and sum beamforming,DAS)不仅在真实声源位置输出具有一定宽度的主瓣,还在非声源位置输出旁瓣,为防止过大的旁瓣导致混淆或湮灭主瓣[2],如何控制旁瓣水平是当前国内外研究热点[3]。2004年,美国BROOKS等[4]首先提出了反卷积波束形成法(deconvolution approach for the mapping of acoustic sources method,DAMAS),此法显著控制了旁瓣水平,克服了传统波束形成法的局限性,近10 a取得了快速发展。本文从反卷积波束形成法理论原理及现有反卷积波束形成法的对比这2个方面来概述。

  • 反卷积波束形成法是一种逆问题求解算法。该算法以传统波束形成作为预处理,得到所有扫描点的均方声压值。假设对流层和剪切层折射并不影响噪声传播到麦克风阵列,可以得出第n个假设声源点传播到第m个麦克风的声压信号矩阵:

    式(1)中:m:n表示第n个声源点到第m个麦克风,e为导向因子。Qn为网格上第n个假设声源点的声压平方矩阵。首先计算:

    式(2)中:m′为除m以外的任意麦克风。可得第n个假设声源点的互谱矩阵:

    式(3)中:Xn=Qn*Qnm0为总麦克风数。麦克风阵列的总互谱矩阵:

    式(4)中:N为总的假设声源点数。可得扫描点的信号功率谱为:

    式(5)中:导向向量 $\hat e = {\left[ {{e_1}\; \; \; {e_2}\; \; \; \; \cdots \; \; \; \; {e_{{m_0}}}} \right]^T}$ 。将式(3)和式(4)代入式(5),可得:

    式(6)中:[ ]n是式(3)中的内容,n′表示聚焦的网格点,n表示假设声源点。简化后,可得:

    式(7)中: ${A_{nn'}} = \frac{{\hat e_n^T{{\left[ {\;} \right]}_{n'}}{{\hat e}_n}}}{{m_0^2}}$ 。令 ${Y_n}\left( {\hat e} \right) = {Y_n}$ ,即等于波束形成算法得出的均方声压值,可得:

    采用高斯-赛德尔迭代法求解此线性方程得出声源分布。在求解线性方程组时,DAMAS算法能更好地考虑不同网格位置的相互影响,使旁瓣变小甚至消失,呈现更直观的结果[3],能更准确地计算出声源位置和声源强度。

  • 在算法的旁瓣抑制能力方面,反卷积波束形成法相比于传统波束形成法具有更好的旁瓣抑制能力。目前,声源识别领域各主要算法都已经进行了大量研究对比[5, 7],研究结果表明:频率的提高可有效改善算法的旁瓣抑制能力,且传统DAS算法的旁瓣抑制能力弱于其他算法。当声源远离中心时,反卷积波束形成法2(DAMAS2)和基于傅里叶变化的非负最小二乘算法(FFT-NNLS)算法定位声源失效。这主要由于DAMAS2和FFT-NNLS都基于阵列点扩散函数位移不变性假设,当声源远离中心时,其声源位置已位于有效区域外,因此DAMAS2和FFT-NNLS不能对该类远离中心的声源点进行精准定位。XENAKI等[8]对这类算法的有效区域进行了深入研究,有效区域一般为Z轴18度角范围内。在相干声源情况下,基于空间相干的洁净算法(CLEAN-SC)不能准确识别声源,这是由于CLEAN-SC在迭代中删除与波峰相干的声源部分,当声源较靠近时,其中一个声源被当作旁瓣而被消除。杨洋等[6]也在研究中指出CLEAN-SC不适用于相干声源。

    在算法的定位精度方面,迭代算法随着迭代次数的增加,算法的精度都有不同程度的提升,分析算法迭代次数与标准差的关系[7]。在低迭代次数时洁净算法(CLEAN)和CLEAN-SC这2种算法的精度较高,具有较高的计算精度稳定性,而DAMAS2,非负最小二乘算法(NNLS)和FFT-NNLS具有较大偏差,且对迭代次数不敏感,迭代次数的增加不能大幅提升其计算精度,DAMAS在高迭代次数下声源识别性能快速迫近CLEAN和CLEAN-SC,对于迭代次数较敏感。空间分辨率的优劣也是算法精度的重要指标, 算法分辨率越高,区分较近声源的能力就越高。通过对风机的噪声分布研究表明[9]:DAMAS具有很高的分辨率,可以获得出色的噪声源分布图,即使声源距离较近,也能够很好地将它们区分,CLEAN-SC也具有很高的分辨率,但由于其算法的理论原因,较近声源,只能识别其中一个,而DAS算法分辨率较差,无法区分较近声源,定位精度降低。

    计算效率是算法能否被广泛运用的关键点,表 1是多种典型算法的计算时间,可得出DAS,DAMAS2,和CLEAN-SC具有较高的计算效率和实时性,其次为DAMAS、稀疏约束反卷积波束形成法(SC-DAMAS),但协方差矩阵拟合法(CMF)、稀疏约束的超分辨率反卷积波束形成算法(SC-RDAMAS)计算时间较长,不适用于实时测试。

    案例 t/s
    DAS DAMAS DAMAS2 SC-DAMAS SC-RDAMAS CMF CLEAN-SC
    1[5] 0.30 78.00 - 12.30 - 69.20 1.20
    2[5] 0.60 138.00 - 31.80 - 123.50 1.60
    3[7] - 240.00 2.20 - - - 4.20
    4[10] 1.39 10.28 - - 917.40 > 1 000.00 -
    说明:"-"代表该案例中未见相关结果

    Table 1.  Computation time of algorithms

    综上所述,主要代表性反卷积算法的特点如表 2所示。

    年份 算法名称 特点
    2004 反卷积波束形成法(DAMAS)[4] 高分辨率,对迭代次数敏感,旁瓣抑制能力强,计算效率较低
    2005 反卷积波束形成法2(DAMAS2)[11] 定位精度对迭代次数不敏感,计算效率高,实时性好,不适用于声源在扫描空间边缘的情况
    2007 基于傅里叶变化的非负最小二乘算法(FFT-NNLS)[12] 计算速度较快,声源定位准确,定位精度对迭代次数不敏感,不适用于声源在扫描空间边缘的情况
    2007 稀疏约束反卷积波束形成法(SC-DAMAS)[13] 利用稀疏性,计算效率较高,在噪声源定位中具有更好的准确性和鲁棒性
    2007 基于空间相干的洁净算法(CLEAN-SC)[14] 可精确提取声源强度,提高了空间分辨率和准确性,计算精度稳定性高,计算效率高,实时性好,但不适用于声源位置相距较近的相干声源
    2011 稀疏约束的超分辨率反卷积波束形成算法(SC-RDAMAS)[10] 适用于在超大分辨率下扫描大区域,计算效率低,实时性较弱,不适用于实时测试
    2012 协方差矩阵拟合法(CMF)[15] 计算效率低,实时性较弱,不适用于实时测试

    Table 2.  Main representative algorithm

  • 当前,反卷积波束形成算法在高频下都具有优秀的旁瓣抑制能力,且定位精度随着迭代次数的上升而提高,但这些算法也存在着各自的局限性,如在计算效率,实时性,扫描范围,适用声源类型等方面,因此在现有算法的基础上,需进一步提出更加全面的反卷积波束形成改进算法,使算法具有更好的普遍适用价值。同时,要提高声源识别的准确性,除了选择合适的算法,开发声源识别性能更优的传声器阵列也是一研究热点。布置形式更加优化、更加合理的传声器,使传声器阵列声源识别的空间分辨率更高,最大旁瓣水平更低,有效动态范围更大[16]

Reference (16)

Catalog

/

DownLoad:  Full-Size Img  PowerPoint
Return
Return