The Northern High Time Resolution Universe pulsar survey -- II. Single-pulse search set-up and simulations

TL;DR

This work develops a dedicated single-pulse search pipeline for the HTRU-North data, augmented by FRBfaker for synthetic SP injections and RFIbye for radio-frequency interference mitigation. Through extensive injection-recovery tests across four FRB morphologies, it quantifies pipeline completeness (roughly $\mathrm{S/N} \approx 11$ for 95% recall) and a fluence sensitivity near $0.16~\mathrm{Jy\,ms}$, while also evaluating the impact of propagation effects and the detector's boxcar discretization. The results demonstrate the pipeline's readiness to process the full HTRU-North dataset, reveal RFI characteristics at Effelsberg, and uncover eight SP trains plus 141 SP candidates that may indicate new neutron stars or FRB-like phenomena. The work also highlights limitations in current FETCH models and the need for morphology-aware retraining, offering concrete paths (training data via FRBfaker, sub-banded searches, padding) to improve detection of complex, Band-limited, or highly scattered SPs with practical implications for future FRB/pulsar surveys.

Abstract

The High Time Resolution Universe (HTRU) survey is an all-sky survey looking for pulsars and other radio transients. A new single-pulse (SP) search pipeline is presented, tailored to the northern part of the HTRU survey collected with the 100m Effelsberg Radio Telescope. In a selection of this data, synthetic SPs are injected with frequency-time structures resembling those of the detected Fast Radio Burst (FRB) population and processed by the pipeline to characterize its performance. Therefore, several new software toolkits have been developed (FRBfaker and RFIbye) to enable the injection of SPs with complex frequency-time structures and cope with the Radio Frequency Interference (RFI) in the survey's data. The operation of these toolkits is described alongside the overall functionality of the SP pipeline. Qualification of the pipeline confirmed that it is ready to process all the HTRU-North data. Additionally, the survey's sensitivity to SPs, the impact of RFI thereon, the performance of the deep-learning classifier FETCH, and some insights that may be used to improve the pipeline's performance in the future are determined. Within the small data sample analysed, 21 known pulsars and a RRAT are detected. In addition, eight faint SP trains that might originate from yet undiscovered neutron stars and 141 isolated SP candidates were discovered.

Figures (15)

• Figure 1: Transient Phase Space showing the radio luminosity versus the variability time scale of several transient events observable in the radio sky. Overlaid are Effelsberg's sensitivity curves for the Galactic distances of 0.1, 1 and 10 kpc, as well as the cosmological distance of 1 Gpc, and pulse widths from 0.109 to 13.98 ms (the used search width range of the SP search). The positions of the re-detected pulsars and RRAT are indicated with the black and grey $\odot$ symbols, respectively.
• Figure 2: Recall curves for each of the four morphologies as a function of the SPs' injected S/Ns. The injected S/Ns for which the HTRU-North SP pipeline is complete are indicated by the coloured dash-dotted lines and entail: $\sim$11 for MI, $\sim$10 for MII, $\sim$14 for MIII, and $\sim$11 for MIV. The morphology averaged recall is plotted as the black dashed lines. The S/N above which the HTRU-North SP pipeline is considered to be complete, irrespective of a burst's morphology, is $\sim$11.
• Figure 3: Recall curves as a function of the SPs' intrinsic fluences per morphology. When all injections are included, these recall curves (pale dashed lines) converge around a value of 0.8 ($c$ in equation \ref{['eq:exponential']}). This convergence occurs mainly due to the applied scattering, rendering injections undetectable even for the brightest injections. By excluding all scattered injections, the solid recall curves are obtained from which pipeline completeness values follow being: $\sim$0.4, $\sim$0.6, and $\sim$0.8 Jy ms for morphologies I/II, III, and IV, respectively.
• Figure 4: Examination of the effects on the detectability of SPs when searched for with a limited set of boxcar widths. The injected S/Ns of all injected SPs are plotted against their discrete injected S/N. The width difference between the best-fitting boxcar width and the actual propagated burst width of a SP is indicated by the plotted marker sizes. The scattering time (referenced to 1 GHz) with which it is injected is represented by the marker colours and the marker symbols, which depict the number of subcomponents a SP is composed of. The red dotted lines indicate heimdall's S/N search threshold of 6.5, and the number of SPs in each quadrant of the plot is given by the red numbers in its corners.
• Figure 5: Recall curves for each of fetch's classification models per FRB morphology as a function of the recovered S/N by heimdall. The coloured numbers in the bottom-right corner of each subplot give the overall recall. For MI, the S/Ns for which the models attain a recall of 0.95 are also shown in grey, i.e., when they are complete.