Protection technologies, security, situational awareness
Due to the emerging threat represented by UAVs (Unmanned Aerial Vehicles) in the civil and military applications, ISL started to develop a compact numerical MEMs microphone antenna adapted to the acoustic localization and classification of these devices. This antenna is designed to detect, localize and classify a potential target.
A prototype of this compact microphones array has been developed along with a new localization technique. First, in order to enhance the detection capability of the array, the signals measured by the 32 digital MEMS microphones arranged in the 2D plane are used for spatial filtering using differential beamforming in four directions. Then, the RANSAC algorithm is used in the time domain to estimate the source direction from estimates of the pressure and two components of the particle velocity at the center of the sensor. Finally, the localized sources are enhanced by DOA informed spatial filtering techniques and identification is performed on these signals.
The goal of the present thesis is to go a step further and initiate new evolution of this compact microphone array based on the perspective raised during the initial steps of the study. It will consist in:
- Study new evolutions for integration of simultaneous multi-target detection and localization. Time domain and spectral domain techniques will be studied that will take advantage of the spatial filtering techniques developed for the microphones array.
- Develop data fusion techniques in the presence of multiple compact MEMs microphone arrays:
- First, outlier rejection techniques will have to be studied, integrated in the individual localization process, and then generalized to the multi-array scenario.
- In a second step, it will be necessary to integrate predictive filtering methods in order to be able to anticipate the movements of the detected targets and feed other types of sensors with sufficiently precise and reliable estimates of the actual position of the threat.
The student will have the opportunity to develop and test the proposed solutions on ISL’s vehicle sounds database integrating various types of targets (UAVs, light/heavy terrestrial vehicles, etc.). In order to complete the database, field experiments can be organized.