STATINA - Touch Interface Engine (TDOA)

Touch interface is one of the most intuitive human-computer interface methods. Although there are many existing touch interface implementation but most of them are still very costly. This project looks into developing a touch interface by using only microphones and a digital signal processing board,...

Full description

Saved in:
Bibliographic Details
Main Author: Su, Kwan Hui.
Other Authors: School of Electrical and Electronic Engineering
Format: Final Year Project
Language:English
Published: 2010
Subjects:
Online Access:http://hdl.handle.net/10356/39499
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
Description
Summary:Touch interface is one of the most intuitive human-computer interface methods. Although there are many existing touch interface implementation but most of them are still very costly. This project looks into developing a touch interface by using only microphones and a digital signal processing board, the ADSP Blackfin 561 (ADSP-BF561) by Analog Devices. The main functionality of the system developed in this project is to localize the source signal which is initiated by user. Localization of source signal is done using the information of time differences of arrival (TDOA) of the source signal as received by each the microphone receivers. The TDOA of signals as received by microphones are estimated by using the General Cross Correlation (GCC) method. Time delay between two channels is calculated as difference between time index corresponding to maximum correlation. Under real environment, a processor weighting method, namely the GCC Phase Transform (GCC-PHAT) method is used as it is robust against the multipath effect of signals and reverberation of sound, to get a better estimation of TDOA of signals. The TDOA information between each signals is a N x N matrix table, where N is the number of channels, that will be used in solving for the location of the source. From the TDOA information, the localization is done by first assuming a point.