|Supervisor:||Dr.-Ing. Robert Aichner|
|Faculty:||Prof. Dr.-Ing. Walter Kellermann|
|Student:||Fei (Felix) Yan|
Blind source separation (BSS) addresses the problem to separate sources from a set of linear mixtures. In the acoustical scenario this corresponds to multiple speech signals acquired by a microphone array. Thus the mixing system is convolutive, i.e., it consists of room impulse responses which can be modelled by finite impulse response (FIR) filters. To accomplish separation we have to assume that the source signals are mutually independent, which is in general true for speech signals. The term "blind" stresses that no additional a priori knowledge about the source signals, mixing conditions or the sensor array configuration is necessary.
In this master thesis a real-time demonstrator suitable for current state-of-the-art BSS algorithms should be developed under Linux in C++. This will be followed by experiments using a real-time demonstrator in combination with a two-microphone array to verify the separation capability of the BSS algorithm in various environments.
Further work will include the examination of the algorithm behaviour when increasing the number of microphones and number of source signals. Possible problems should be identified and several modifications which may lead to a more robust algorithm should be investigated.