An acoustic sensor network is a group of sensors that are connected via a network (usually WLAN). Each sensor has one or more microphones. As each account in the network therefore has its own module for signal acquisition, the signals are generally not digitised at the same sampling rate.

If the multi-channel signals are now processed, the different sampling rates must either be corrected or taken into account in the signal processing.

Various aspects of signal processing in distributed sensor networks are to be investigated and implemented as part of Bachelor's and Master's theses.

Please contact me for current topics and tasks.

Contact: Jörg Schmalenströer, Janek Ebbers, Tobias Gburrek

We are interested in combinations of neural network and classic signal enhancement methods. Past Bachelor and Master theses focussed on beamforming and model based separation. More recent work combined these with gradient based learning and neural networks.

If you are interested in advancing methods, unorthodox applications of neural networks and would like to apply your math and programming knowledge in these areas, please visit us in our office so that we may discuss possible topics.

Contact: Christoph Böddeker, Thilo von Neumann, Tobias Cord-Landwehr