Aircraft type recognition based on shortwave speech communication with PLP

This paper studies aircraft type recognition based on shortwave speech communication. Features are extracted based on Perceptual Linear Predictive (PLP) with RASTA and High-order Cumulant, considering auditory perceptual properties, which combines with physical properties of the target acoustic signal at the same time. Support vector machine(SVM) is used as classifier. The simulation results of aircraft type recognition have shown that RASTA-PLP-HOC based on PLP with RASTA, combined with HOC with physical properties of signal, can extract the robust feature of aircraft cabin background sound and identify the six kinds of aircrafts with high recognition rate. RASTA-PLP-HOC-SVM has achieved more excellent performance and higher recognition rate than RASTA-PLP -SVM and MFCC-SVM.