Generating calibration signals for an adaptive beamformer
AssigneeTelefonaktiebolaget LM Ericsson
InventorsJim Rasmusson, Ingvar Claesson, Mattias Dahl, Sven Nordholm
Life of this patent
Estimating the term…
Loading patent…
AssigneeTelefonaktiebolaget LM Ericsson
InventorsJim Rasmusson, Ingvar Claesson, Mattias Dahl, Sven Nordholm
Estimating the term…
A beamformer is calibrated for use as an acoustic echo canceler in a hands-free communications environment having a loudspeaker and a plurality of microphones. To perform the calibration, a number of adaptive filters are provided in correspondence with each of the microphones, and each of the adaptive filters is trained to model echo properties of the environment as experienced by the corresponding one of the microphones. A target source is activated, thereby generating an acoustic signal that is received by the microphones. The trained adaptive filters are then used to generate jammer signals by, for example, having each one filter a pseudo noise signal. Respective ones of the jammer signals are then combined with corresponding signals supplied by the microphones, thereby generating combination signals. The combination signals are then used to adapt the beamformer to cancel the jammer signals. In another aspect of the invention, the adaptive filters may be utilized during normal operation by having them perform an echo cancellation operation on each of the signals that is to be supplied to the calibrated beamformer.
A method of calibrating a beamformer for use in a hands-free communications environment having a loudspeaker and a plurality of microphones, the method comprising the steps of:
A method of canceling an echo generated in a hands-free communications environment having a loudspeaker and a plurality of microphones, the method comprising the steps of:
An apparatus for calibrating a beamformer for use in a hands-free communications environment having a loudspeaker and a plurality of microphones, the apparatus comprising:
An apparatus for canceling an echo generated in a hands-free communications environment having a loudspeaker and a plurality of microphones, the method comprising:
Dependencies are inferred from the claim text.