A Pixel Space Method for Testing Dipole Modulation in the CMB Polarization. (arXiv:1807.02359v2 [astro-ph.CO] UPDATED)
<a href="http://arxiv.org/find/astro-ph/1/au:+Ghosh_S/0/1/0/all/0/1">Shamik Ghosh</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Jain_P/0/1/0/all/0/1">Pankaj Jain</a>

We introduce a pixel space method to detect dipole modulation or
hemispherical power asymmetry in the cosmic microwave background (CMB)
polarization. The method relies on the use of squared total polarized flux
whose ensemble average picks up a dipole due to the dipole modulation in the
CMB polarization. The method is useful since it can be applied easily to
partial sky. We define several statistics to characterize the amplitude of the
detected signal. By simulations we show that the method can be used to reliably
extract the signal at 3$sigma$ level in future CORE-like missions, assuming
that the signal is present in the CMB polarization at the level detected by the
Planck mission in the CMB temperature. An application of the method to the 2018
Planck data does not detect a significant effect. This is be due to the
presence of correlated detector noise and residual systematics in data. Using
the FFP10 we find the presence of a very strong bias which might be masking any
real effect.

We introduce a pixel space method to detect dipole modulation or
hemispherical power asymmetry in the cosmic microwave background (CMB)
polarization. The method relies on the use of squared total polarized flux
whose ensemble average picks up a dipole due to the dipole modulation in the
CMB polarization. The method is useful since it can be applied easily to
partial sky. We define several statistics to characterize the amplitude of the
detected signal. By simulations we show that the method can be used to reliably
extract the signal at 3$sigma$ level in future CORE-like missions, assuming
that the signal is present in the CMB polarization at the level detected by the
Planck mission in the CMB temperature. An application of the method to the 2018
Planck data does not detect a significant effect. This is be due to the
presence of correlated detector noise and residual systematics in data. Using
the FFP10 we find the presence of a very strong bias which might be masking any
real effect.

http://arxiv.org/icons/sfx.gif