摘要
针对现有半实值化MVDR(Semi-Real-Valued Minimum Variance Distortionless Response,SRV-MVDR)算法存在波达角(Direction of Arrival,DOA)模糊的缺点(无法区分半谱内的真实源与镜面辐射源),提出一种改进的SRV-MVDR方法.为了对SRV-MVDR算法解模糊,首先引入能量的思想,通过利用信号的特征向量和特征值构建一种能量谱函数,所构建的能量谱函数包含了完整的声源目标方位信息,利用这一特性将所构建的能量谱函数与现有的SRV-MVDR算法相结合.由于SRV-MVDR真实源谱峰会被能量谱函数谱峰放大形成新的主峰,而镜面辐射源得不到能量谱函数谱峰的放大,从而形成伪峰,即在半谱范围内,主峰便对应真实源位置,伪峰对应镜面辐射源位置,从而解决了SRV-MVDR算法DOA模糊的问题.最后仿真实验表明:所提算法不仅能够解决SRV-MVDR波达角模糊的弊端,且在抗噪性能上有一个较大的提升,这是由于本文算法充分利用信号信息,从而算法更加适用于低信噪比的工程环境.
An improved SRV-MVDR （Semi-Real-Valued Minimum Variance Distortionless Response）has been proposed to overcome the shortcomings of the SRV-MVDR algorithm for Direction of Arrival（DOA）ambiguity（una- ble to distinguish between real and mirror radiation sources in the half spectrum）.In order to solve the problem of D0A ambiguity in SRV-MVDR algorithm, the eigenveetors and eigenvalues of the signal have been introduced to construct an energy spectrum function. The constructed energy spectrum function contains the complete sound source target DOA information. Using this property,the energy spectrum function constructed in this paper has been combined with the existing SRV-MVDR algorithm. The SRV-MVDR true source peak has been amplified by the energy spectrum function peak to form a new main peak.The mirror radiation source can not be amplified by the peak of the energy spectrum function, thereby forming a pseudo peak.Therefore, in the semi-spectral range, the main peak corresponds to the true source＇s DOA position and the pseudo-peak corresponds to the mirror radiation source＇ s DOA position. Therefore, the problem of SRV-MVDR algorithm DOA ambiguity has been solved. Finally, computer simulation experiments have shown that the improved SRV MVDR algorithm can not only overcome the shortcomings of SRV MVDR Dirrection of Arrival ambiguity, but also have a big improvement in anti-noise performance compared to SRV-MVDR algorithm. What＇s more,the performance of the algorithm is robust. Ali of these benefit from the full use of the signal information. The algorithm can be used in the harsh conditions of low signal to noise ratio （SNR）, and is more suitable for the actual engineering environment. Through these experiments, the effectiveness and superiority of the proposed algorithm have been proved.
作者
陈峰
王彪
陈迎春
莫世奇
何呈
Chen Feng;Wang Biao;Chen Yingchun;Mo Shiqi;He Cheng(School of Electronic and Information,Jiangsu University of Science and Technology,Zhenjiang,212003,China;College of Underwater Acoustic Engineering,Harbin Engineering University,Harbin,150001,China)
出处
《南京大学学报（自然科学版）》
CAS
CSCD
北大核心
2018年第5期937-943,共7页
Journal of Nanjing University（Natural Science）
基金
国家自然科学基金(11574120
U1636117)
江苏省自然科学基金(BK20161359)
江苏省研究生科研与实践创新计划项目(SJCX17_0604)
