| |
 |
| Paper: |
Athena X-IFU Event Reconstruction: Extreme Learning Machine Approach |
| Volume: |
522, Astronomical Data Analysis Software and Systems XXVII |
| Page: |
435 |
| Authors: |
Ceballos, M. T.; Cobo, B. |
| Abstract: |
The X-ray Observatory Athena (Nandra et al. 2013) was proposed in April 2014 as the mission to implement the science theme "The Hot and Energetic Universe"
selected by ESA for L2 (the second Large-class mission in ESA's Cosmic Vision science programme).
One of the two X-ray detectors designed to be onboard Athena is the X-IFU (Barret et al. 2016), a cryogenic microcalorimeter based on Transition Edge Sensor (TES) technology
that will provide spatially resolved high-resolution spectroscopy.
X-IFU will be developed by an international consortium led by IRAP (PI), SRON (co-PI) and IAPS/INAF (co-PI) and involving ESA Member States, Japan and the United States.
The X-ray photons absorbed in X-IFU generate intensity pulses that must be detected and reconstructed onboard to recover their energy, position and arrival time.
The software prototype package (SIRENA) in development as an anticipated contribution of IFCA/Spain to X-IFU through the
Digital Readout Electronics (DRE) unit (Ravera
et al. 2014), is currently testing a set of processing algorithms with the aim of finding the best compromise between performance and
availability of on-board computing resources.
The optimal filtering has recently been chosen as the baseline algorithm but other techniques are still under study in an effort to get the best results at the
lower computing cost. In particular at IFCA, we are currently investigating the performance of the Extreme Learning Machine (ELM) algorithm for single-hidden
layer feedforward neural networks (SLFNs) for the reconstruction of the pulses in the X-IFU data signal. |
|
|
|
|
|
