(HERA): Simulation of the chromatic effects of the antenna and

Download Report

Transcript (HERA): Simulation of the chromatic effects of the antenna and

The “Hydrogen Epoch of Reionization Array”
(HERA)
Simulation of the chromatic effects of the antenna and
impact on the detection of the EoR power spectrum
Nicolas Fagnoni – Cosmology on Safari – 14th February 2017
[email protected] – Department of Physics, University of Cambridge, UK
Summary
1. Detection of the EoR signal with HERA
2. The problematic of chromatic effects
3. Electromagnetic and electrical co-simulation of HERA
4. Design improvement
Detection of the redshifted
hydrogen 21cm signal with HERA
Detection of the redshifted hydrogen 21cm signal with HERA
• EoR signal contaminated by the foreground signal
 Galactic synchrotron emission + extra-galactic radio sources
 Foreground ~ 105 more intense than the EoR signal
• Detection of the signal:
 “Foreground subtraction” method
 Complicated, requires an excellent knowledge of the foreground properties and
chromatic effects induced by the telescope
 “Foreground avoidance” method
 “Smooth” foreground spectrum vs varying EoR spectrum
 Study of a specific region of the EoR signal not contaminated by the foreground
Power delay spectrum
• Delay spectrum
• Power delay spectrum
• Components of the wave vector k
EoR window and “wedge”
•
Delay power spectrum averaged in a
cylindrical cosmic volume
•
k⊥ = parameter associated with the
spatial scale of the observed region in
the plane of the sky
 Small k⊥ ⇒ large region probed
 Proportional to the baseline
•
k∥ = parameter associated with the time
scale of the Reionization
 Depends on the redshift and
baseline delay of the received signal
Credit: Liu, A. et al. (2014)
The problematic of chromatic
effects
The problematic of chromatic effects
•
PRISim
 Foreground sky model + EoR model
(21cmFAST)
 Convolved with antenna beam models
•
Achromatic beam
 Foreground contamination limited by
the maximum signal delay associated
with the baseline (i.e the “horizon
delay limit”)
•
Chromatic beam
 Foreground leakage at high k-modes
 Contamination of the EoR window
Thyagarajan, N., et al. (2016)
Sources of chromatic effects
Mutual coupling
Feed-vertex reflections
Crucial to understand, model and limit these chromatic effects
Electromagnetic and electrical cosimulation
Antenna model with CST
RF-front end model
• RF front-end: active balun + transmission cables + analogue receiver
• Electrical circuits simulated with Genesys
• Generation of the S-parameters
Electromagnetic / electrical co-simulation
• Antenna model + front-end S-parameters
• Simulation excited by a plane wave coming from the zenith
 Gaussian pulse centred on 150 MHz and with a bandwidth of 100 MHz
• Transient solver (time domain simulation => solution for all freq. in 1 run)
• “Finite Integration Technique”
• Hexahedral mesh (18 million cells)
• Simulation time: 1000 ns with a time step of 0.003 ns
• Simulation of the output voltage after the receiver
Antenna output signal
Main signal
Dish-feed
reflections
Cable
reflections
Antenna voltage response
Effect of the reflections on the EoR signal
• Constraints on the attenuation level of the
reflected signal as a function of the delay of
the reflections
• White area  the foreground spillover should
not impact the EoR detection
•
Ideal scenario: signal attenuated by 60 dB
after 60 ns
• k∥-mode non-detectable up to 0.2 h/Mpc
because of reflections in cables, otherwise
0.15 h/Mpc
Credit:Thyagarajan, N., et al. (2016)
Impedance mismatch
• Reflections caused by a problem of impedance mismatch between the balun and the
antenna termination
 Balun impedance: close to 65 – 30i ohm
 BUT the antenna impedance varies a lot
 Historical reason: RF front-end optimised for PAPER
Are simulations reliable?
100-ohm termination
Coupling in HERA 19
•
Antennas excited by a plane wave
 Antenna response
•
Central antenna excited
 S-parameters and beam model
Antenna response
100-ohm termination
Beam
•
With coupling
 Slighlty higher sidelobes and
lower gain
 Sidelobes not smooth at all
 Ohter effects under
investigation
At 150 MHz
Design improvement
Matching network
• New electrical matching circuit to be inserted between the antenna and
receiver
 Smooth the impedance transition
 Made up of 10 lumped elements (inductors + capacitors)
 Decrease the reflections by ~ 10 dB (k// modes above 0.1h/Mpc may be
detectable, if reflections in cable avoided)
 BUT additional losses (between 0.2 and 0.9 dB)
 Noise figure of the amplifiers modified
Central parabolic cone
• Central cone
 Flatten the antenna impedance
 Impedance matching easier (reflections decreased by 20 dB)
 BUT increase the sidelobe level by 5 – 10 dB
Development of a new Vivaldi feed
• Vivaldi feed
 Larger bandwidth: 50 – 250 MHz (z = 4.7 – 27.4)
115 million and 1.3 billion years after the Big Bang
 Null experiment at high freq.
 Probe the Cosmic Down at low freq.
• “Travelling-wave” antenna
 Impedance and beam more stable
over a large band
Conclusion
Conclusion
• The study of the EoR signal is the key to understand the birth and
evolution of the first galaxies and stars
• Astrophysical results are limited by the hardware
 Essential to properly understand and limit the impacts of the
instrument on the data
 Now it is possible to reach a good level of precision using end-toend computer simulation
 Same method applied to SKA-LA
Thank you for your attention.
Questions?