Researchers used a lattice method borrowed from cryptography to find a new pulsar in NASA Fermi data in 10 hours of computing, where older searches failed.
An international team of astronomers has found a way to spot hidden pulsars in NASA gamma-ray data roughly a million times faster than before, according to a paper published on Thursday, 19 November, in Physical Review Letters. Working with 15 years of observations from NASA's Fermi Gamma-ray Space Telescope, the researchers identified PSR J1622−7204, an isolated pulsar that spins 261 times per second. NASA says the search used the equivalent of 10 hours of computing time on a standard processor. The previous best method, given enough computing power to keep a single processor busy for 600 years, failed to find the same object. The study is led by Daniel Pook-Kolb of the Max Planck Institute for Gravitational Physics in Germany, with Dotan Gazith and Barak Zackay of Israel's Weizmann Institute of Science.
A pulsar is a rapidly spinning neutron star, the collapsed core of a dead star, that sweeps beams of radiation across space like a lighthouse. Finding one in gamma-ray data is hard because Fermi catches only a handful of photons from each source, buried in background noise, and astronomers must guess its spin rate, how fast that spin is slowing and its exact position all at once. Until now, teams split the data into chunks and relied on volunteers' home computers through the Einstein@Home project, often spending hundreds of computer-years per discovery. Gazith and Zackay recast the search as a geometry puzzle: finding the shortest vector in a lattice, a regular grid of points in many dimensions. The algorithms come from cryptographers testing encryption meant to resist future quantum computers.
Each trial run took about five minutes on one processor core, so the method can run on an ordinary laptop, NASA says. Once it flagged a candidate, the team confirmed the signal using the full Fermi dataset. The researchers now want to go after pulsars in binary systems, where a companion star's gravity distorts the signal and has so far made them nearly impossible to detect. Co-author Colin Clark, of the Max Planck Institute, estimates that tens or perhaps hundreds of such pulsars are hiding in the existing data, and the team plans to pair the new algorithm with Einstein@Home volunteers. Fermi, launched in 2008, scans the whole sky every three hours, and all of its data is public, which let a team from outside the mission try a new approach.



