Astrophysics Seminar

Modern radio interferometers typically consist of hundreds to thousands of antenna elements and employ substantial digital signal processing to handle large operating bandwidths of a few tens to hundreds of MHz. Conventionally, FX correlators are used as the primary signal processing unit of the interferometer and these become computationally complex and inefficient for large-N (>1000) arrays. For this reason, a generic fast imaging back-end called the E-field Parallel Imaging Correlator (EPIC) was recently deployed on the Long Wavelength Array station at the Sevilleta National Wildlife Refuge (LWA-SV) in New Mexico. EPIC uses a novel architecture that has the capability to produce electric field or intensity images of the sky at the angular resolution of the array with full or partial polarization and spectral resolution of the channelizer at a very high cadence–on the order of tens of milliseconds. By eliminating the intermediate cross-correlation data products, the computational costs are significantly lowered in comparison to a conventional FX or XF correlator from   ̴N 2ant to   ̴ Nant log Nant for large and dense, but otherwise arbitrary array layouts. This substantially lowers the output data rates while yielding two-dimensional polarimetric image products for science analysis. EPIC is now implemented on a GPU-accelerated hardware and commissioned, at the Long Wavelength Array station located at the Seviletta National Refuge (LWA-SV) in New Mexico (USA), as a commensal transient imaging back-end that can potentially detect and localize sources of impulsive radio emission on millisecond timescales. In this talk, I will discuss the architecture of EPIC, its GPU implementation describing code optimizations that improve performance. I will present the initial validations from commissioning observations of EPIC and simultaneous beam-formed observations of bright sources in the field.