Pulsar Candidates Research Articles

A hybrid ensemble method for pulsar candidate classification

In this paper, three ensemble methods: Random Forest, XGBoost, and a Hybrid Ensemble method were implemented to classify imbalanced pulsar candidates. To assist these methods, tree models were used to select features among 30 features of pulsar candidates from references. The skewness of the integrated pulse profile, chi-squared value for sine-squared fit to amended profile and best S/N value play important roles in Random Forest, while the skewness of the integrated pulse profile is one of the most significant features in XGBoost. More than 20 features were selected by their relative scores and then applied in three ensemble methods. In the Hybrid Ensemble method, we combined Random Forest and XGBoost with EasyEnsemble. By changing thresholds, we tried to make a trade-off between Recall and Precision to make them approximately equal and as high as possible. Experiments on HTRU 1 and HTRU 2 datasets show that the Hybrid Ensemble method achieves higher Recall than the other two algorithms. In HTRU 1 dataset, Recall, Precision, and F-Score of the Hybrid Ensemble method are $0.967$, $0.971$, and $0.969$, respectively. In HTRU 2 dataset, the three values of that are $0.920$, $0.917$, and $0.918$, respectively.

A processing pipeline for high volume pulsar candidate data streams

Abstract Pulsar candidate analysis pipelines have historically been comprised of bespoke software systems, supporting the off-line study of data. However modern data acquisition systems are making off-line analyses impractical, as increasing candidate volumes become prohibitively expensive to retain. To maintain processing capabilities when off-line analysis becomes infeasible due to cost, requires a shift to on-line data processing. This paper makes four contributions facilitating this shift relevant to the search for radio pulsars: (i) it characterises the key components of a pulsar search candidate processing pipeline, (ii) it examines the feasibility of implementing on-line candidate filtering via existing tools, (iii) explores the problems preventing an easy transition to on-line filtering, and finally (iv) presents a new prototype filtering pipeline capable of overcoming such problems. Realised using Commercial off-the-shelf (COTS) software components, the deployable system is open source, simple, scalable, and cheap to produce. It has the potential to achieve candidate filtering design requirements for the Square Kilometre Array (SKA), illustrated via testing under simulated SKA loads.

Pulsar candidate selection using ensemble networks for FAST drift-scan survey

The Commensal Radio Astronomy Five-hundred-meter Aperture Spherical radio Telescope (FAST) Survey (CRAFTS) utilizes the novel drift-scan commensal survey mode of FAST and can generate billions of pulsar candidate signals. The human experts are not likely to thoroughly examine these signals, and various machine sorting methods are used to aid the classification of the FAST candidates. In this study, we propose a new ensemble classification system for pulsar candidates. This system denotes the further development of the pulsar image-based classification system (PICS), which was used in the Arecibo Telescope pulsar survey, and has been retrained and customized for the FAST drift-scan survey. In this study, we designed a residual network model comprising 15 layers to replace the convolutional neural networks (CNNs) in PICS. The results of this study demonstrate that the new model can sort >96% of real pulsars to belong the top 1% of all candidates and classify >1.6 million candidates per day using a dual--GPU and 24--core computer. This increased speed and efficiency can help to facilitate real-time or quasi-real-time processing of the pulsar-search data stream obtained from CRAFTS. In addition, we have published the labeled FAST data used in this study online, which can aid in the development of new deep learning techniques for performing pulsar searches.

The first pulsar discovered by FAST

To assist with the commissioning (Jiang et al. 2019) of the Five-hundred-meter Aperture Spherical radio Telescope (FAST), we performed a pulsar search, with the primary goal of developing and testing the pulsar data acquisition and processing pipelines. We tested and used three pipelines, two (P1 and P2 hereafter) searched for the periodic signature of pulsars whereas the other one was used to search for bright single pulses (P3 hereafter). A pulsar candidate was discovered in the observation on the 22nd August, 2017, and later confirmed by the Parkes radio telescope on the 10th September, 2017. This pulsar, named PSR J1900-0134, was the first pulsar discovered by FAST. The pulsar has a pulse period of 1.8 s and a dispersion measure (DM) of 188\,pc\,cm$^{-3}$.

A LOFAR search for steep-spectrum pulsars in supernova remnants and pulsar wind nebulae

Pinpointing a pulsar in its parent supernova remnant (SNR) or resulting pulsar wind nebula (PWN) is key to understanding its formation history and the pulsar wind mechanism, yet only about half the SNRs and PWNe appear associated with a pulsar. Our aim was to find the pulsars in a sample of eight known and new SNRs and PWNe. Using the LOFAR radio telescope at 150 MHz, each source was observed for 3 h. We covered the entire remnants where needed, by employing many tied-array beams to tile out even the largest objects. For objects with a confirmed point source or PWN we constrained our search to those lines of sight. We identified a promising radio pulsar candidate towards PWN G141.2+5.0. The candidate, PSR J0337+61, has a period of 94 ms and a DM of 226 pc cm−3. We re-observed the source twice with increased sensitivities of 30% and 50%, but did not re-detect it. It thus remains unconfirmed. For our other sources we obtain very stringent upper limits of 0.8 − 3.1 mJy at 150 MHz. Generally, we can rule out that the pulsars travelled out of the remnant. From these strict limits we conclude our non-detections towards point sources and PWNe are the result of beaming and propagation effects. Some of the remaining SNRs should host a black hole rather than a neutron star.

Prolonged sub-luminous state of the new transitional pulsar candidate CXOU J110926.4−650224

We report on a multi-wavelength study of the unclassified X-ray source CXOU J110926.4−650224 (J1109). We identified the optical counterpart as a blue star with a magnitude of ∼20.1 (3300–10500 Å). The optical emission was variable on timescales from hundreds to thousands of seconds. The spectrum showed prominent emission lines with variable profiles at different epochs. Simultaneous XMM-Newton and NuSTAR observations revealed a bimodal distribution of the X-ray count rates on timescales as short as tens of seconds, as well as sporadic flaring activity. The average broad-band (0.3–79 keV) spectrum was adequately described by an absorbed power law model with photon index of Γ = 1.63 ± 0.01 (at 1σ c.l.), and the X-ray luminosity was (2.16 ± 0.04) × 1034 erg s−1 for a distance of 4 kpc. Based on observations with different instruments, the X-ray luminosity has remained relatively steady over the past ∼15 years. J1109 is spatially associated with the gamma-ray source FL8Y J1109.8−6500, which was detected with Fermi at an average luminosity of (1.5 ± 0.2) × 1034 erg s−1 (assuming the distance of J1109) over the 0.1–300 GeV energy band between 2008 and 2016. The source was undetected during ATCA radio observations that were simultaneous with NuSTAR, down to a 3σ flux upper limit of 18 μJy beam−1 (at 7.25 GHz). We show that the phenomenological properties of J1109 point to a binary transitional pulsar candidate currently in a sub-luminous accretion disk state, and that the upper limits derived for the radio emission are consistent with the expected radio luminosity for accreting neutron stars at similar X-ray luminosities.

Constraints on the number of X‐ray pulsars in IC 10 from a deep XMM‐Newton observation

We report the most sensitive search yet for X‐ray pulsars in the dwarf starburst galaxy IC 10, which is known to contain a population of young, high‐mass X‐ray binaries. We searched for pulsations in 207 point‐like X‐ray sources in the direction of IC 10 by a 2012 XMM‐Newton observation with a total exposure time of 134.5 ks. Pulsation searches in faint objects can be sensitive to the energy bands of the light curves, as well as the source and background extraction areas. We analyzed separately the PN and metal–oxide–semiconductor barycenter corrected 0.2–12 keV data, with good time interval filtering. Different schemes for source and background extraction were compared, and the search was repeated in the narrower 0.5–8 keV energy band to increase the signal‐to‐noise ratio. For the most conservative parameters, five point sources produced significant peaks in the Lomb–Scargle periodogram (99% significance, single trial, assuming white noise). A similar number of different candidates result from alternative analyses. A ∼4,100 s period seen in all three instruments for the black hole (BH) + Wolf–Rayet binary IC 10 X‐1 is probably due to red noise of astrophysical origin. Considering the periods, luminosities, and spatial distribution of the pulsar candidates in the direction of IC 10, they do not belong to the same distribution as the ones in the Magellanic Clouds and Milky Way. This result holds even if the candidates are spurious, since if the Small Magellanic Cloud were placed at the distance of IC 10, we would expect to see nearly five pulsars at Lx > 1036 erg s–1 inside the D25 contour, and their periods would be of order 100 s, rather than the mostly ∼1 s periods for the candidates reported here, which lie outside the main body of the galaxy.

Pulsar candidate recognition with deep learning

In this paper, we present a deep learning-based recognition algorithm to identify pulsars by observing data containing millions of candidates including radio frequency interference and noise sources. The dataset is obtained from the High Time Resolution Universe survey created and updated by the Parkes telescope. We investigate several effective single and combined features via simple logistic regression. To deal with the imbalanced dataset, we oversimplify the original dataset at different sampling rates, which is also one of the learning parameters. After training the pre-processed dataset via a convolutional neural network, we provide a cross-validated evaluation of all candidates. Results show that the deep-learning based recognition algorithm can identify the pulsar and radio frequency interference signals with high accuracy. The precision and recall of radio frequency interference are both 100%, and those of pulsars are 91% and 94%, respectively.

Pulsar Identification of Background Spectra of LIGO Data

By choosing the metric (called physical metric) in general relativity as the exact solution to the Einstein equation that fits the time delay data, one can determine the size and gravitational redshift on the surface of compact objects (neutron stars and black holes). The author shows that the physical metric is invariant by rotation. As a result, the frequencies of gravitational waves from pulsars are represented as n * f / for pulsar frequency f and harmonics n. Based on this result, the author has identified potential pulsar candidates with gravitational wave spectra. This result will be critical in the study of gravitational redshift of compact objects. PACS numbers: 04.20.-q, 04.20.Jb, 04.40.Nr, 98.80.-k

Comment on “Characterizing the population of pulsars in the Galactic bulge with the Fermi large area telescope” [arXiv:1705.00009v1

Abstract The Fermi-LAT Collaboration recently presented a new catalog of gamma-ray sources located within the 40 ° × 40 ° region around the Galactic Center Ajello et al. (2017) — the Second Fermi Inner Galaxy (2FIG) catalog. Utilizing this catalog, they analyzed models for the spatial distribution and luminosity function of sources with a pulsar-like gamma-ray spectrum. Ajello et al. (2017) v1 also claimed to detect, in addition to a disk-like population of pulsar-like sources, an approximately 7 σ preference for an additional centrally concentrated population of pulsar-like sources, which they referred to as a “Galactic Bulge” population. Such a population would be of great interest, as it would support a pulsar interpretation of the gamma-ray excess that has long been observed in this region. In an effort to further explore the implications of this new source catalog, we attempted to reproduce the results presented by the Fermi-LAT Collaboration, but failed to do so. Mimicking as closely as possible the analysis techniques undertaken in Ajello et al. (2017), we instead find that our likelihood analysis favors a very different spatial distribution and luminosity function for these sources. Most notably, our results do not exhibit a strong preference for a “Galactic Bulge” population of pulsars. Furthermore, we find that masking the regions immediately surrounding each of the 2FIG pulsar candidates does not significantly impact the spectrum or intensity of the Galactic Center gamma-ray excess. Although these results refute the claim of strong evidence for a centrally concentrated pulsar population presented in Ajello et al. (2017), they neither rule out nor provide support for the possibility that the Galactic Center excess is generated by a population of low-luminosity and currently largely unobserved pulsars. In a spirit of maximal openness and transparency, we have made our analysis code available at https://github.com/bsafdi/GCE-2FIG .

Characterising the VHE diffuse emission in the central 200 parsecs of our Galaxy with H.E.S.S.

The diffuse very high-energy (VHE; >100 GeV) γ-ray emission observed in the central 200 pc of the Milky Way by H.E.S.S. was found to follow dense matter distribution in the central molecular zone (CMZ) up to a longitudinal distance of about 130 pc to the Galactic centre (GC), where the flux rapidly decreases. This was initially interpreted as the result of a burst-like injection of energetic particles 104 yr ago, but a recent more sensitive H.E.S.S. analysis revealed that the cosmic-ray (CR) density profile drops with the distance to the centre, making data compatible with a steady cosmic PeVatron at the GC. In this paper, we extend this analysis to obtain, for the first time, a detailed characterisation of the correlation with matter and to search for additional features and individual γ-ray sources in the inner 200 pc. Taking advantage of 250 h of H.E.S.S. data and improved analysis techniques, we perform a detailed morphology study of the diffuse VHE emission observed from the GC ridge and reconstruct its total spectrum. To test the various contributions to the total γ-ray emission, we used an iterative 2D maximum-likelihood approach that allows us to build a phenomenological model of the emission by summing a number of different spatial components. We show that the emission correlated with dense matter covers the full CMZ and that its flux is about half the total diffuse emission flux. We also detect some emission at higher latitude that is likely produced by hadronic collisions of CRs in less dense regions of the GC interstellar medium. We detect an additional emission component centred on the GC and extending over about 15 pc that is consistent with the existence of a strong CR density gradient and confirms the presence of a CR accelerator at the very centre of our Galaxy. We show that the spectrum of full ridge diffuse emission is compatible with that previously derived from the central regions, suggesting that a single population of particles fills the entire CMZ. Finally, we report the discovery of a VHE γ-ray source near the GC radio arc and argue that it is produced by the pulsar wind nebula candidate G0.13−0.11.

The population of TeV pulsar wind nebulae in the H.E.S.S. Galactic Plane Survey

The nine-year H.E.S.S. Galactic Plane Survey (HGPS) has yielded the most uniform observation scan of the inner Milky Way in the TeV gamma-ray band to date. The sky maps and source catalogue of the HGPS allow for a systematic study of the population of TeV pulsar wind nebulae found throughout the last decade. To investigate the nature and evolution of pulsar wind nebulae, for the first time we also present several upper limits for regions around pulsars without a detected TeV wind nebula. Our data exhibit a correlation of TeV surface brightness with pulsar spin-down power Ė. This seems to be caused both by an increase of extension with decreasing Ė, and hence with time, compatible with a power law RPWN(Ė) ~Ė−0.65±0.20, and by a mild decrease of TeV gamma-ray luminosity with decreasing Ė, compatible with L1−10 TeV ~Ė0.59±0.21. We also find that the offsets of pulsars with respect to the wind nebula centre with ages around 10 kyr are frequently larger than can be plausibly explained by pulsar proper motion and could be due to an asymmetric environment. In the present data, it seems that a large pulsar offset is correlated with a high apparent TeV efficiency L1−10 TeV∕Ė. In addition to 14 HGPS sources considered firmly identified pulsar wind nebulae and 5 additional pulsar wind nebulae taken from literature, we find 10 HGPS sources that are likely TeV pulsar wind nebula candidates. Using a model that subsumes the present common understanding of the very high-energy radiative evolution of pulsar wind nebulae, we find that the trends and variations of the TeV observables and limits can be reproduced to a good level, drawing a consistent picture of present-day TeV data and theory.

Space Navigation by X-Ray Pulsars

This review considers the problem of autonomously determining the position of a spacecraft in space based on the analysis of pulses emitted by X-ray pulsars. The characteristics of the prospective equipment and lists of pulsar candidates for reference sources are given. The navigation algorithm and resulting accuracy characteristics are substantiated.

The POlarised GLEAM Survey (POGS) I: First results from a low-frequency radio linear polarisation survey of the southern sky

AbstractThe low-frequency polarisation properties of radio sources are poorly studied, particularly in statistical samples. However, the new generation of low-frequency telescopes, such as the Murchison Widefield Array (the precursor for the low-frequency component of the Square Kilometre Array) offers an opportunity to probe the physics of radio sources at very low radio frequencies. In this paper, we present a catalogue of linearly polarised sources detected at 216 MHz, using data from the Galactic and Extragalactic All-sky Murchison Widefield Array survey. Our catalogue covers the Declination range –17° to –37° and 24 h in Right Ascension, at a resolution of around 3 arcminutes. We detect 81 sources (including both a known pulsar and a new pulsar candidate) with linearly polarised flux densities in excess of 18 mJy across a survey area of approximately 6 400 deg2, corresponding to a surface density of 1 source per 79 deg2. The level of Faraday rotation measured for our sources is broadly consistent with those recovered at higher frequencies, with typically more than an order of magnitude improvement in the uncertainty compared to higher-frequency measurements. However, our catalogue is likely incomplete at low Faraday rotation measures, due to our practice of excluding sources in the region where instrumental leakage appears. The majority of sources exhibit significant depolarisation compared to higher frequencies; however, a small sub-sample repolarise at 216 MHz. We also discuss the polarisation properties of four nearby, large-angular-scale radio galaxies, with a particular focus on the giant radio galaxy ESO 422–G028, in order to explain the striking differences in polarised morphology between 216 MHz and 1.4 GHz.

An image-based search for pulsars among Fermi unassociated LAT sources

We describe an image-based method that uses two radio criteria, compactness and spectral index, to identify promising pulsar candidates among Fermi Large Area Telescope (LAT) unassociated sources. These criteria are applied to those radio sources from the Giant Metrewave Radio Telescope all-sky survey at 150 MHz (TGSS ADR1) found within the error ellipses of unassociated sources from the 3FGL catalog and a preliminary source list based on 7 years of LAT data. After follow-up interferometric observations to identify extended or variable sources, a list of 16 compact, steep-spectrum candidates is generated. An ongoing search for pulsations in these candidates, in gamma rays and radio, has found six millisecond pulsars and one normal pulsar. A comparison of this method with existing selection criteria based on gamma-ray spectral and variability properties suggests that the pulsar discovery space using Fermi may be larger than previously thought. Radio imaging is a hitherto underutilized source selection method that can be used, as with other multi-wavelength techniques, in the search for Fermi pulsars.

Search for and detection of pulsars inmonitoring observations at 111 MHz

Pulsars with periods more than 0.4 seconds in the declination range -9o < decj < 42o and in the right ascension range 0h < r.a.< 24h were searched in parallel with the program of interplanetary scintillations monitoring of a large number of sources with the radiotelescope LPA LPI. Four-year observations carried out at the central frequency 110.25 MHz in the band 2.5 MHz and in six frequency channels with the time resolution 100 ms were used for the search. Initial detections of candidates for pulsars were based on the Fourier power spectra averaged over the whole observations period. The candidates for pulsars were checked in more details from observations analysis with higher time-frequency resolution: 32 frequency channels with time resolution of 12.5 ms. 18 new pulsars was discovered, their main characteristics are presented.

Machine Learning for Pulsar Detection

AbstractThe next generation of radio telescopes will have unprecedented sensitivity and time-resolution offering exciting new capabilities in time-domain science. However, this will result in very large numbers of pulsar and transient event candidates and the associated data rates will be technically challenging in terms of data storage and signal processing. Automated detection and classification techniques are therefore required and must be optimized to allow high-throughput data processing in real time. In this paper we provide a summary of the emerging machine learning techniques being applied to this problem.

Searching for pulsars associated with the Fermi GeV excess

The Fermi Large Area Telescope has detected an extended region of GeV emission toward the Galactic Center that is currently thought to be powered by dark matter annihilation or a population of young and/or millisecond pulsars. In a test of the pulsar hypothesis, we have carried out an initial search of a 20 deg**2 area centered on the peak of the galactic center GeV excess. Candidate pulsars were identified as a compact, steep spectrum continuum radio source on interferometric images and followed with targeted single-dish pulsation searches. We report the discovery of the recycled pulsar PSR 1751-2737 with a spin period of 2.23 ms. PSR 1751-2737 appears to be an isolated recycled pulsar located within the disk of our Galaxy, and it is not part of the putative bulge population of pulsars that are thought to be responsible for the excess GeV emission. However, our initial success in this small pilot survey suggests that this hybrid method (i.e. wide-field interferometric imaging followed up with single dish pulsation searches) may be an efficient alternative strategy for testing whether a putative bulge population of pulsars is responsible for the GeV excess.

A millisecond pulsar candidate in a 21-h orbit: 3FGL J0212.1+5320

We present the discovery of a variable optical counterpart to the unidentified gamma-ray source 3FGL J0212.1+5320 and argue that this is a new compact binary millisecond pulsar (MSP) candidate. We show 3FGL J0212.1+5320 hosts a semidetached binary with a 0.869 55 ± 0.000 15 d orbital period and an F6-type companion star at an estimated distance of D = 1.1 ± 0.2 kpc, with a radial velocity curve semi-amplitude K2 = 214.1 ± 5.0 km s−1 and a projected rotational velocity of V sin (i) = 73.2 ± 1.6 km s−1. We find a hard X-ray source at the same location with a 0.5–10 keV luminosity LX = 2.6 × 1032 (D/1.1 kpc)2 erg s−1, which strengthens the MSP identification. Our results imply a mass ratio q = M2/M1 = 0.26|$^{+0.02}_{-0.03}$| if the companion star fills its Roche lobe, and q ≳ 0.26 in any case. This classifies 3FGL J0212.1+5320 as a ‘redback’ binary MSP; if its MSP nature is confirmed, this will be the brightest compact binary MSP in the optical band (r″ ≃ 14.3 mag) and will have the longest orbital period among Galactic field systems (nearly 21 h). Based on the light curve peak-to-peak amplitude (Δr = 0.19 mag), we further suggest that the orbital inclination is high and the putative pulsar mass is close to canonical (M1 ≃ 1.3–1.6 M⊙). Finally, we discuss the lack of heating signatures and asymmetric optical light curves in the context of other redback MSPs.

Radio-interferometric and near-IR observations of PSR J1357—6429

PSR J1357—6429 is a young radio pulsar detected in X-rays and γ-rays. We performed radio-interferometric and high spatial resolution near-infrared observations of the pulsar. Based on the obtained radio data, we estimated the most accurate pulsar position, RA = 13:57:02.525(14) and Dec = 64:29:29.89(15). Using the new and archival radio data, we obtained the 90 per cent pulsar proper motion upper limit μ < 106 mas yr-1, which corresponds to the pulsar transverse velocity vtr < 1300 km s-1. The near-infrared imaging revealed a point source which we propose as a pulsar near-infrared counterpart candidate. It is confidently detected in the J and Ks bands, with J = 23.51±0.24 and Ks = 21.82±0.25. The respective dereddened near-infrared fluxes are compatible with the long-wavelength extrapolation of the pulsar X-ray spectrum. If the candidate is the true counterpart, by this property PSR J1357—6429 would be similar to the nearby middle-age pulsar PSR B0656+14. In this case, both pulsars demonstrate an unusually high near-infrared efficiency relative to the X-ray efficiency as compared to other pulsars detected in both ranges.