AN AUTOMATIC SYSTEM FOR MICROPHONE SELF-LOCALIZATION USING AMBIENT SOUND
Författare
Summary, in English
In this paper, we develop a system for microphone selflocalization
based on ambient sound, without any assumptions
on the 3D locations of the microphones and sound
sources. We aim at developing a system capable of dealing
with multiple moving sound sources. We will show that this
is possible given that there are instances where there are only
one dominating sound source. In the first step of the system
we employ a feature detection and matching strategy. This
produces TDOA data, possibly with missing data and with
outliers. Then we use a robust and stratified approach for the
parameter estimation. We use robust techniques to calculate
initial estimates on the offsets parameters, followed by nonlinear
optimization based on a rank criterion. Sequentially
we use robust methods for calculating initial estimates of the
sound source positions and microphone positions, followed
by non-linear Maximum Likelihood estimation of all parameters.
The methods are tested and verified using anechoic
chamber sound recordings.
based on ambient sound, without any assumptions
on the 3D locations of the microphones and sound
sources. We aim at developing a system capable of dealing
with multiple moving sound sources. We will show that this
is possible given that there are instances where there are only
one dominating sound source. In the first step of the system
we employ a feature detection and matching strategy. This
produces TDOA data, possibly with missing data and with
outliers. Then we use a robust and stratified approach for the
parameter estimation. We use robust techniques to calculate
initial estimates on the offsets parameters, followed by nonlinear
optimization based on a rank criterion. Sequentially
we use robust methods for calculating initial estimates of the
sound source positions and microphone positions, followed
by non-linear Maximum Likelihood estimation of all parameters.
The methods are tested and verified using anechoic
chamber sound recordings.
Avdelning/ar
Publiceringsår
2014
Språk
Engelska
Publikation/Tidskrift/Serie
European Signal Processing Conference
Fulltext
Dokumenttyp
Konferensbidrag
Förlag
EURASIP
Ämne
- Mathematics
Conference name
22nd European Signal Processing Conference -
Conference date
2014-09-01 - 2014-09-05
Conference place
Lisbon, Portugal
Status
Published
ISBN/ISSN/Övrigt
- ISSN: 2219-5491