To address the challenges of non-cooperative and remote acoustic detection, an all-fiber laser Doppler vibrometer(LDV) is established. The all-fiber LDV system can offer the advantages of smaller size, lightweight des...To address the challenges of non-cooperative and remote acoustic detection, an all-fiber laser Doppler vibrometer(LDV) is established. The all-fiber LDV system can offer the advantages of smaller size, lightweight design and robust structure, hence it is a better fit for remote speech detection. In order to improve the performance and the efficiency of LDV for long-range hearing, the speech enhancement technology based on optimally modified log-spectral amplitude(OM-LSA) algorithm is used. The experimental results show that the comprehensible speech signals within the range of 150 m can be obtained by the proposed LDV. The signal-to-noise ratio(SNR) and mean opinion score(MOS) of the LDV speech signal can be increased by 100% and 27%, respectively, by using the speech enhancement technology. This all-fiber LDV, which combines the speech enhancement technology, can meet the practical demand in engineering.展开更多
信号检测是激光多普勒测速(LDV)系统实现高精度的关键技术,针对LDV中微弱多普勒信号的检测,本文从噪声在频域中的统计特性出发,对多普勒信号进行带阻滤波,结合雷达的恒虚警(CFAR)检测技术,设计了基于频域的单元平方和自适应阈值检测算...信号检测是激光多普勒测速(LDV)系统实现高精度的关键技术,针对LDV中微弱多普勒信号的检测,本文从噪声在频域中的统计特性出发,对多普勒信号进行带阻滤波,结合雷达的恒虚警(CFAR)检测技术,设计了基于频域的单元平方和自适应阈值检测算法以解决在低信噪比(SNR)的环境中LDV的信号检测问题,提高LDV的信号探测能力,同时降低其虚警概率。仿真与实验的结果表明:该算法与固定阈值相比可以在SNR为-12 d B时实现信号的完全检测,保持低虚警概率,运算简单,工程适用性强。展开更多
基金supported by the National Natural Science Foundation of China(No.61205143)
文摘To address the challenges of non-cooperative and remote acoustic detection, an all-fiber laser Doppler vibrometer(LDV) is established. The all-fiber LDV system can offer the advantages of smaller size, lightweight design and robust structure, hence it is a better fit for remote speech detection. In order to improve the performance and the efficiency of LDV for long-range hearing, the speech enhancement technology based on optimally modified log-spectral amplitude(OM-LSA) algorithm is used. The experimental results show that the comprehensible speech signals within the range of 150 m can be obtained by the proposed LDV. The signal-to-noise ratio(SNR) and mean opinion score(MOS) of the LDV speech signal can be increased by 100% and 27%, respectively, by using the speech enhancement technology. This all-fiber LDV, which combines the speech enhancement technology, can meet the practical demand in engineering.
文摘信号检测是激光多普勒测速(LDV)系统实现高精度的关键技术,针对LDV中微弱多普勒信号的检测,本文从噪声在频域中的统计特性出发,对多普勒信号进行带阻滤波,结合雷达的恒虚警(CFAR)检测技术,设计了基于频域的单元平方和自适应阈值检测算法以解决在低信噪比(SNR)的环境中LDV的信号检测问题,提高LDV的信号探测能力,同时降低其虚警概率。仿真与实验的结果表明:该算法与固定阈值相比可以在SNR为-12 d B时实现信号的完全检测,保持低虚警概率,运算简单,工程适用性强。
