eduzhai > Applied Sciences > Engineering >

Deep Neural Network based Distance Estimation for Geometry Calibration in Acoustic Sensor Networks

  • Save

... pages left unread,continue reading

Document pages: 5 pages

Abstract: We present an approach to deep neural network based (DNN-based) distanceestimation in reverberant rooms for supporting geometry calibration tasks inwireless acoustic sensor networks. Signal diffuseness information from acousticsignals is aggregated via the coherent-to-diffuse power ratio to obtain adistance-related feature, which is mapped to a source-to-microphone distanceestimate by means of a DNN. This information is then combined withdirection-of-arrival estimates from compact microphone arrays to infer thegeometry of the sensor network. Unlike many other approaches to geometrycalibration, the proposed scheme does only require that the sampling clocks ofthe sensor nodes are roughly synchronized. In simulations we show that theproposed DNN-based distance estimator generalizes to unseen acousticenvironments and that precise estimates of the sensor node positions areobtained.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×