Title :
Omnidirectional robotic ear based on the spatio-temporal gradient analysis with blind signal separation
Author :
Tsuruta, Kohsuke ; Teramoto, Kenbu
Author_Institution :
Saga Univ., Japan
Abstract :
This paper presents a method for a robotic ear based on the spatio-temporal gradient analysis with blind signal separation that estimates the object signals from overlapped signals. By using the spatial derivatives of mixed signals, not only the object source signals but also the arrival directions are estimated. The effectiveness of the proposed method is confirmed through the numerical experiment.
Keywords :
blind source separation; ear; gradient methods; robots; Hermite moment; blind signal separation; omnidirectional robotic ear; overlapped signals; spatio-temporal gradient analysis;
Conference_Titel :
SICE 2004 Annual Conference
Conference_Location :
Sapporo
Print_ISBN :
4-907764-22-7