The future engine architecture (UHBR or CROR) will be less and less noisier and it will become difficult to extract from acoustic signal measured by one or several microphones the acoustic contribution from turbulent boundary layer for example. Unfortunately, current state-of-the-art signal processing methods are not mature enough to be able to de-noise and extract accurately both tonal and broadband noise part and to separate accordingly the contribution of different sources. Three main axis will be explored:
1. Aeroacoustics sources separation by the use of spatial filtering
2. De-noising by the use of cyclostationarity
3. Acoustic sources localization by the use of Baysian approach.