Title :
Time delay estimation in the time-frequency domain based on a line detection approach
Author :
Sandmair, Andreas ; Lietz, Mario ; Stefan, Johannes ; León, Fernando Puente
Author_Institution :
Karlsruhe Institute of Technology, Institute of Industrial Information Technology, Hertzstr. 16, 76187, Germany
Abstract :
The estimation of the time delay of arrival (TDOA) or the direction of arrival (DOA) has been an important research topic within the last few decades. In several applications, e.g., acoustic source localization or source separation, the estimation of the delay or the direction is a significant processing block. Most of the available algorithms estimate the delay in the time domain. Therefore, the estimate is an integral multiple of the sampling period. Depending on the sampling rate and other parameters, the resolution is restricted and possibly not adequate for specific applications. To overcome the resolution problem, a new method based on the phase delay between two sensors and a line detection approach commonly used in image processing is proposed. Experimental results demonstrate the advantage of the introduced approach.
Keywords :
Delay; Delay effects; Direction of arrival estimation; Estimation; Noise measurement; Sensors; Transforms; Radon transform; direction of arrival; generalized cross-correlation; resolution problem; time delay estimation;
Conference_Titel :
Acoustics, Speech and Signal Processing (ICASSP), 2011 IEEE International Conference on
Conference_Location :
Prague, Czech Republic
Print_ISBN :
978-1-4577-0538-0
Electronic_ISBN :
1520-6149
DOI :
10.1109/ICASSP.2011.5947046