用模糊函数表示超声信号,以解决特征提取时不同中心频率和不同到达时间的超声信号的预处理问题,并为信号分类提供时频特征。提出采用 Karhunen-Loeve(K-L)变换提取信号的模糊平面特征,并对特征的分类能力进行了分析。实验结果表明模糊函数能有效去除信号的中心频率和到达时间的差异,用 K-L 变换提取的模糊平面特征能在低维空间有效地描述超声信号,并能获得较好的分类效果。% Ambiguity function (AF) is proposed to represent ultrasonic signal to solve the preprocessing problem of feature extraction with different center frequency and different arriving time of ultrasonic signals, as well as to provide time-frequency features for signal classification. Moreover, Karhunen-Loeve (K-L) transform is used to extract ambiguity features of the ultrasonic signal, and the classification ability of the features is analyzed. Experimental results show that ambiguity function eliminates the difference of center frequency and arriving time among ultrasonic signals, and ambiguity plane features extracted by K-L transform describe signal of different class effectively in a reduced dimensional space and obtain better classification results.
展开▼