论文部分内容阅读
针对现阶段国内对舱音记录仪(CVR),所记录声音的辨别和分析方法没有充分利用机舱内声音的定源特性,不能准确地分辨出飞机上特定声音,提出了一种基于独立成分分析(ICA)对舱音信号进行分离的方法,用Matlab软件进行了仿真,并通过蒙特卡洛方法(MCM)计算多组实验数据的性能指标,以验证算法的稳定性。结果表明,该方法可以比较高效地提取出飞机机舱中特定语音信号。
In view of the current domestic CVT, the method of discriminating and analyzing the recorded sound does not make full use of the fixed source characteristics of cabin sound, and can not accurately distinguish the specific sound on the aircraft. A method based on Independent Component Analysis (ICA) is used to separate the signals of cabin sound. The simulation is carried out by Matlab software. The performance of several groups of experimental data is calculated by Monte Carlo method (MCM) to verify the stability of the algorithm. The results show that this method can extract specific voice signals from aircraft cabin efficiently.