Interpretable large scale deep generative models for Dark Matter searches
AARTFAAC is an all-sky radio telescope and transient-detection facility. It piggybacks on raw data from a limited number of antennas of the LOFAR telescope. Last year, the AARTFAAC 2.0 program started, which couples a planned telescope upgrade with better transient-detection capabilities and new science cases.
This proposal, the PetaFLOP AARTFAAC Data-Reduction Engine (PADRE), aims to improve the AARTFAAC processing pipeline to achieve the following goals.
First, transients will be detected in real time with low latency, so that the raw samples of all LOFAR antennas (which are available for only seven seconds) can be saved in time for further analysis, while other instruments, observing at other wavelengths, are alerted to initiate follow-up observations immediately.
Second, the pipeline will scale to support more antennas and more bandwidth, through algorithmic optimizations and the use of new GPU technologies, which yields better images and greatly enhances the transient discovery rate.
Third, the pipeline will facilitate other, new use cases (space-wheather monitoring, ionospheric research, cosmic dawn) by providing intermediate data products like calibrated images. The pipeline that PADRE will develop is thus at the heart of a versatile, powerful instrument.