Yi Feng

Fast radio bursts (FRBs) are millisecond-duration radio transients. As they propagate through the interstellar medium, they interact with free electrons, resulting in dispersion. The corresponding dispersion measure (DM) is referred to as the real DM (DM _real ). In practice, however, the DM derived from modeling (DM _model ) is often contaminated by intrinsic burst morphology, giving rise to a pseudo DM component (DM _pseudo = DM _model − DM _real ). In this work, we focus on the highly active repeating FRB 20220912A and utilize its microshots—extremely short-duration (typically tens of microseconds), broadband emissions—to investigate its DM _real and DM _pseudo . We adopt two assumptions: first, that FRB 20220912A resides in a nonmagnetoionic environment and that its DM _real variation is smaller than 10 ⁻² pc cm ⁻³ over a few years; and second, that microshots have a negligible intrinsic morphological time delay. By identifying two new microshots and combining them with previously reported ones, we find that all four microshots exhibit remarkably consistent DM values over a 1 month timescale, with an average of 219.380 ± 0.004 pc cm ⁻³ . We define this value as the DM _real of FRB 20220912A. We further show that bright, narrow bursts with a width of less than 2 ms also yield DM estimates consistent with the microshot-based DM _real . A survey of five repeating FRBs reveals that DM _pseudo is a common phenomenon, with variations typically spanning a range of approximately 10 pc cm ⁻³ at 1.2 GHz. These findings highlight the importance of accounting for morphological contributions in DM interpretation and demonstrate that microshots and narrow bursts are powerful tools for probing DM _real .

Fast radio bursts (FRBs) are millisecond-duration extragalactic radio transients, some of which are associated with compact persistent radio sources (PRSs), hinting at a physical connection. While several models have been proposed to explain PRSs and their connection to FRBs, direct observational tests remain limited. Here, we report for the first time a correlated trend between the long-term variation of the PRS flux density and the burst energetics of FRB 20190520B and FRB 20240114A, suggesting a physical coupling between the PRS and FRB activity. We further examine additional repeaters with compact PRSs and find no clear correlation between PRS luminosity and burst activity, likely due to the limited observations. These results are consistent with scenarios in which both the PRS and FRB activity may be powered by a common energy reservoir, such as the magnetic or rotational energy of a magnetar.

We present phase-connected timing solutions, total-intensity and polarized profiles for two newly discovered pulsars, PSRs J1929+4105 and J2359+4239, identified by the Five-hundred-meter Aperture Spherical radio Telescope (FAST) in the Commensal Radio Astronomy FAST Survey (or CRAFTS). PSR J1929+4105 has a spin period of 41.59 ms and is in a 63 days orbit with an eccentricity of 0.00042. PSR J2359+4239, with a spin period of 60.53 ms, is in a 4 days orbit with an eccentricity of 0.0003. Since no post-Keplerian parameters have been measured for either system, we can only constrain the companion masses based on the mass function. The minimum companion masses for PSRs J1929+4105 and J2359+4239 are 0.72 M _⊙ and 0.467 M _⊙ , respectively. Given the small eccentricities of these systems, which rule out the possibility of double neutron star systems, we propose that both systems consist of a pulsar and a massive white dwarf companion. Simulations of the formation processes for these two systems were conducted using the rapid binary population synthesis code COMPAS. We estimated their distances using the NE2001 and YMW16 electron-density models, thereby contributing new data to the Galactic electron-density distribution. Motivated by the numerous discoveries of millisecond pulsar binaries by FAST, we further studied their height distribution in the Galaxy.

Rotating radio transients (RRATs) represent a critical yet enigmatic population of neutron stars whose sporadic emission is driven by highly intermittent magnetospheric states. We present a high-sensitivity, multi-epoch polarimetric study of RRAT J2237+2828 using the Five-hundred-meter Aperture Spherical radio Telescope, uncovering a complex, nonstationary magnetospheric environment. Individual bursts exhibit high linear polarization (20%–60%) and frequent orthogonal polarization modes, suggesting a fundamental coherent emission mechanism similar to canonical pulsars. However, the emission clearly shows bimodal behavior: pulses cluster into a High-energy burst state (Mode B) and a low-energy recovery state (Mode A). We identify a profound temporal asymmetry in the state-switching dynamics. While the transition from Mode B to Mode A occurs abruptly within 1–2 pulsar periods, the transition back to Mode B requires an extensive waiting time spanning hundreds to thousands of seconds. A compelling positive correlation between the inter-burst waiting time and the subsequent burst fluence ( r _s ≈ 0.45), coupled with a significant deviation from Poissonian statistics, strongly suggests that the system is regulated by an energy accumulation and release process. This long-term “system memory” and the observed timescales suggest that the intermittency is regulated by processes beyond pure magnetospheric dynamics, such as crustal stress accumulation or the evolution of multipolar magnetic fields. The stability of the rotation measure over 2 yr disfavors environmental origins and supports an intrinsic origin for the observed intermittency.

In this study, we present a transformer-based multitask model for fast radio burst (FRB) detection, signal segmentation, and parameter estimation directly from time–frequency data, without requiring computationally expensive dedispersion preprocessing. To overcome the scarcity of labeled observational data, we develop an FRB simulator and a rule-based automatic annotation pipeline, enabling training exclusively on simulated data. Evaluations on the FAST-FREX data set show that our model achieves an F1 score of 97.8%, recall of 95.7%, and precision of 100%, outperforming both conventional tools (e.g., PRESTO, Heimdall) and recent AI-based baselines (e.g., RaSPDAM, DRAFTS) in both accuracy and inference speed. The model supports pixel-level signal segmentation and yields reliable estimates for dispersion measure and time of arrival. Large-scale blind searches on CRAFTS data further demonstrate robustness, with an average false-positive rate of 0.28% and minimal human verification required. This search has already led to the identification of two pulsar candidates, both confirmed as known pulsars. Processing benchmarks indicate that the model enables real-time searches on a single consumer-grade GPU, making petabyte-scale blind searches feasible. The code is publicly available on GitHub, and the model can be easily integrated with existing tools to automate and streamline radio data analysis beyond FRB or pulsar searches.