A Post-correlation Beamformer for Time-domain Studies of Pulsars and Transients

Abstract

We present a detailed analysis of post-correlation beamforming (i.e. beamforming which involves only phased sums of the correlation of the voltages of different antennas in an array), and compare it with the traditionally used incoherent and phased beamforming techniques. Using data from the GMRT we show that post-correlation beamformation results in a many-folds increase in the signal-to-noise for periodic signals from pulsars and several order of magnitude reduction in the number of false triggers from single pulse events like fast radio bursts (FRBs). This difference arises primarily because the post-correlation beam contains less red-noise, as well as less radio frequency interference. The post-correlation beam can also be more easily calibrated than the incoherent or phased array beams. We also discuss two different modes of post-correlation beamformation, viz. (1) by subtracting the incoherent beam from the coherent beam and (2) by phased addition of the visibilities. The computational costs for both these beamformation techniques as well as their suitability for studies of pulsars and FRBs are discussed. Techniques discussed here would be of interest for all upcoming surveys with interferometric arrays. Finally, we describe a timedomain survey with the GMRT using the post-correlation beamformation as a case study. We find that post-correlation beamforming will improve the current GMRT time-domain survey sensitivity by ∼ 2 times for pulsars with periods of few 100s of millisecond and by many-folds for even slower pulsars, making it

Reduced-resolution beamforming: Lowering the computational cost for pulsar and technosignature surveys

“…This conclusion is consistent with Roy et al (2018), which considers post-correlation beamforming techniques for pulsar studies with the Giant Metrewave Radio Telescope. Roy et al (2018) also concludes that post-correlation beamforming is computationally cheaper for a large number of beams at low time resolution, and regular beamforming is cheaper for a small number of high time-resolution beams.…”

“…Note that in synthesis imaging, autocorrelations are generally not included; that is, the weights tensor W pq = 0 if p = q. From a power beam interpretation, this is equivalent to subtracting the incoherently summed beam from the tied-array power beam (Roy et al 2018).…”

“…From a power beam interpretation, this is equivalent to subtracting the incoherently summed beam from the tied-array power beam (Roy et al. 2018).…”

Deep Synoptic Array Science: Polarimetry of 25 New Fast Radio Bursts Provides Insights into Their Origins

“…The voltage dumps are processed using a custom fullpolarization beamforming pipeline (dsa110-bbproc). Postcorrelation beamforming is implemented (Roy et al 2018) with the addition of both auto-and cross-correlation terms to mitigate quantization noise in applying per-antenna beamforming weights. For FRB data, the weights are determined from visibility data on bright bandpass calibrators typically within 12 hr of detection.…”

Wide field beamformed observation with MeerKAT

“…The visibility-based approach has used in Roy et al (2018), where it is described by the following expression:…”