Speaker
Description
The proposed signal processing architecture for the next phase of the MWA consists of number of improvements including a new over-sampled coarse PFB and an FFT based ultra-fine channeliser. This architecture is designed to eliminate aliasing at the coarse channel edges, a problem that continues to plague MWA data to this day. Strict requirements, particularly those for EoR science, impose limits on the extent that systematic errors can contaminate the spectrum after 100 hours of integration. Therefore, we conducted an experiment that emulates the full signal chain within the lab and integrated the data for a duration of 100 hours. I present the findings from this experiment and our progress towards verifying the new architecture can meet these requirements.
