We propose to develop a system to isolate and extract individual bioelectrical and acoustic sources from the output of sensors that are responding to multiple simultaneous sources. The purpose of the system is to enable the development of other health-related applications by allowing researchers to focus on the applications instead of the details of signal collection and analysis. Our system will """"""""clean up"""""""" live signals in real time by separating competing foreground sources, suppressing background sources, and identifying and removing echoes and similar effects from the results. It will employ multiple sensors with algorithms to extract individual sources from noisy environments, and to determine source directions and environment characteristics such as reflecting surfaces. An innovation in our system is that some sensors are used to """"""""tag"""""""" known sources. Tagging sensors are attached to significant target or masking sources that are identified to the system. Other sensors are used to pick up background noise and remote (untagged) target or masking sources. The system will provide high-level functionality through tagging sensors and simple, general information about the sources and the environment, using techniques of """"""""blind source separation"""""""". This will allow researchers to focus less on details of the data collection and coping with the environment, and more on the sources themselves or their positional and signal information. In Phase 1, we will test the separation algorithm and observe its performance with and without tagging sensors. The proposed system would be useful to researchers who need to create high-fidelity low- noise recordings in noisy environments such as MRI scanners, and who are not audio or bioelectrical-signal engineers. It would allow a user to tag the most prominent sources, record the entire """"""""signal scene"""""""", and extract the desired source signals and related location information. A second important use of our system would be as an assistive listening device for persons with mild to moderate hearing loss, allowing them to function effectively in noisy social situations such as meetings, restaurants, and conferences. With appropriate sensors, the system will be suitable for use with bioelectric signals - EEG, EMG, etc. - to allow researchers and clinicians to study fetal and maternal heartbeats separately, both for waveform patterns and for the locations of the corresponding sources.
We propose to develop a system to isolate and extract individual signal sources, whether bioelectrical (EEG, ECG) or acoustic, to enable the development of other health-related applications. An important use of our system would be as an assistive listening device for persons with mild to moderate hearing loss, allowing them to function effectively in noisy social situations such as meetings and restaurants. It would also be useful to researchers who need to create high-fidelity low-noise recordings in noisy environments such as MRI scanners. It would be equally suitable for separating bioelectrical signals such as fetal and maternal heartbeats, and providing location information for each of the sources.
