Researchers find pulsar in Fermi data using new search method


The team found PSR J1622−7204 in about 10 CPU hours; an earlier method failed after approximately 600 CPU years. Researchers plan to extend the search to pulsars with companion stars.
Researchers have found a previously unknown pulsar in 15 years of data from NASA’s Fermi Gamma-ray Space Telescope, using a search method that turns the hunt into a geometry problem. The discovery is the first gamma-ray pulsar found using an extension of the method, the team said in a study published 19 November in Physical Review Letters.
The isolated pulsar, PSR J1622−7204, spins 261 times per second. A pulsar is a rapidly rotating neutron star whose beams of radiation sweep across space.
The successful search used about 10 hours of computing time on a standard processor. The previous best method had failed to find the same pulsar after a calculation equivalent to about 600 years on a standard processor. NASA says the new method can find pulsars in its public data about a million times faster than before; the comparison is not between two successful detections.
Fermi picks up only a few photons, or particles of light, from a given pulsar amid signals from other sources. Researchers must search for an unknown spin rate, the rate at which it slows and a position in the sky. Testing every possible combination would require enormous computing power.
Dotan Gazith and Barak Zackay of Israel’s Weizmann Institute of Science developed an approach that recasts the search as finding a short vector in a high-dimensional lattice, a regularly spaced array of points. It uses algorithms developed by cryptographers trying to break encryption designed to resist future quantum computers, rather than testing every possible periodic signal. Gazith said the pulsar analysis itself is not cryptography. Daniel Pook-Kolb, a scientist at Germany’s , led the study.

After the method identified a candidate, the team checked it against the full Fermi data set. NASA said the resulting signal was too strong to be a chance occurrence.
In an earlier proof of concept, Gazith, Aaron B. Pearlman and Zackay recovered the timing pattern of PSR J0318+0253, a gamma-ray pulsar already discovered by another search, in a few CPU minutes. The new study extends that work to a discovery.
Earlier searches split Fermi data into smaller pieces and used volunteers' spare computing power. Einstein@Home discoveries of four young gamma-ray pulsars in Fermi data were published in 2013.
NASA launched Fermi in 2008 and made its data public after an initial one-year calibration period. New data continue to be released as soon as they are processed. The open archive let researchers outside the mission try their approach, while NASA’s Fermitools helped them prepare the data. The study also provides an example implementation and material for reproducing its timing solution.
Co-author Colin Clark, an astrophysicist at the Max Planck Institute, said the team aims to combine the algorithm with Einstein@Home volunteers' computing power. The goal is to search for pulsars with companion stars, whose gravity distorts their signals. The study says such searches have been intractable without detailed prior knowledge of orbital parameters. Clark said there are probably tens or hundreds of such pulsars in the data that researchers have not yet found.
Researchers estimated that monthly dispensing rose from 28 to 67 prescriptions per 1,000 women aged 40–74, as increases that began in 2024 continued into 2026.
2 min

A study published November 17 also found that memory decline became less steep across successive generations, adding evidence about changes in memory as people age.
2 min

Étienne Artigau’s team says the system, 28 light-years from Earth, is the nearest known triple-star system with two white dwarfs. The researchers are investigating how its red dwarf survived its companion’s expansion.
2 min