摘要
介绍了粒子滤波的基本思想和具体算法实现步骤,并将粒子滤波算法应用在声源定位中,解决了在高斯噪声环境下的声源定位问题。所提出的基于粒子滤波的声源定位方法,在高斯噪声情况下,甚至在低信噪比(SNR<-20dB)情况下,定位的均方根误差RMSE值均小于0.2m。
PF（Particle Filter） is a new filtering method based on Bayesian estimation and Monte Carlo method, and can effectively cope with complicated nonlinear and/or non-Gaussian problems. The basic idea and algorithm description of particle filter are presented. Then, the PF is introduced into acoustic source localization. The rootmean-square error of this method is less than 0.2 meters in Gaussian noisy environment even in the environment of low SNR.
出处
《电声技术》
2009年第10期52-55,共4页
Audio Engineering
基金
广西自然科学基金(0832007Z)
关键词
高斯噪声
粒子滤波
到达时间差
声源定位
均方根误差
Gaussian noise
PF
time delay of arrival
acoustic source localization
root-mean-square error
