二维多快拍无网格压缩波束形成声源识别理论及算法研究

Conventional grid-based compressive beamforming acoustic source identification technology suffers from the basis mismatch, which makes the results inaccurate. The two-dimensional grid-free compressive beamforming developed by the applicant and the members of this project can overcome the limit fundamentally for the measurements with planar microphone arrays. It is based on the single-snapshot data model and perfectly suitable for identifying transient or moving sources. The existing researches on the one-dimensional compressive beamforming and the two-dimensional grid-based compressive beamforming have proved that for stationary sources, the identification performance with multiple-snapshot data is far superior to the one with single-snapshot data. Motivated to identify stationary sources accurately, this project will devote itself to building a complete two-dimensional multiple-snapshot grid-free compressive beamforming theory and forming a numerical algorithm. First, a denoising mathematic model is built for planar microphone arrays under the multiple-snapshot grid-free framework. Then, fast, accurate and robust mathematic model solving methods are explored, as well as two-dimensional direction-of-arrival estimation and source strength quantification methods. Next, the influence rules and mechanisms of typical factors on the identification performance are revealed. Both example experiments and engineering application experiments will be utilized to examine the correctness, effectiveness, and superiority of the built theory and algorithm. The research achievements will provide an innovative approach for the accurate identification of stationary acoustic sources, help to perfect the acoustic source identification function of compressive beamforming, and ultimately make positive contributions to the development of the noise test, analysis and control discipline.

传统有网格压缩波束形成声源识别技术存在基不匹配问题，声源识别结果不准确。申请人及项目组成员为平面传声器阵列测量发展的二维无网格压缩波束形成能从根本上克服该问题，其以单快拍数据模型为基础，特别适合识别瞬态及运动声源。关于一维压缩波束形成及二维有网格压缩波束形成的已有研究证明：对稳态声源，采用多快拍数据的识别性能远优于单快拍数据。以准确识别稳态声源为目标，本项目在多快拍无网格框架下，建立平面传声器阵列接收声信号去噪声数学模型、探索准确快速稳健的数学模型求解方法和二维声源波达方向估计及强度量化方法、揭示典型因素对声源识别性能的影响规律、探明影响机理，最终建立完整的二维多快拍无网格压缩波束形成声源识别理论、形成数值算法，并基于算例试验及工程应用试验检验其正确性、有效性和优越性。研究成果将为准确识别稳态声源提供新途径，有助于完善压缩波束形成声源识别功能，对噪声测试、分析及控制学科的发展做出积极贡献。

传统有网格压缩波束形成声源识别技术存在基不匹配问题，声源识别结果不准确。基于平面传声器阵列测量发展的二维无网格压缩波束形成能从根本上克服该问题，其以单快拍数据模型为基础，特别适合识别瞬态及运动声源。本项目以进一步提升稳态声源的识别性能为目标，在多快拍无网格框架下，建立了平面传声器阵列接收声信号去噪声数学模型、探索了准确快速稳健的数学模型求解方法和二维声源波达方向估计及强度量化方法、揭示了典型因素对声源识别性能的影响规律并探明了影响机理，最终建立了完整的二维多快拍无网格压缩波束形成声源识别理论、形成了数值算法，并基于算例试验及工程应用试验检验了其正确性、有效性和优越性。研究成果为准确识别稳态声源提供新途径，有助于完善压缩波束形成声源识别功能，对噪声测试、分析及控制学科的发展做出了积极贡献。