USSF-317 · Falcon 9 Block 5Liftoff in Nov 20, 04:32 UTCWatch live ›
Hot Space News

A cryptography trick finds a hidden pulsar in Fermi data a million times faster

An international team found a pulsar spinning 261 times a second in public Fermi data with a method about a million times faster than before, and an earlier approach missed it.

A cryptography trick finds a hidden pulsar in Fermi data a million times faster

An international team has found a pulsar, a rapidly spinning dead star, in public data from NASA’s Fermi Gamma-ray Space Telescope about a million times faster than before, NASA said. The method borrows an idea from cryptography.

The pulsar, named PSR J1622-7204, is isolated and spins 261 times per second. The team found it using 15 years of Fermi data and the equivalent of 10 hours of computing time on a standard computer processor. The previous best method failed to find it, even though it used enough computing power that a standard processor would need 600 years. The results were published Thursday in Physical Review Letters.

Artist's impression of NASA's Fermi Gamma-ray Space Telescope orbiting Earth
Image: NASA's Goddard Space Flight Center/Chris Smith (USRA/GESTAR)

Why a pulsar is so hard to pick out of gamma-ray data

A pulsar is a rapidly spinning neutron star, the dense core left behind by a star that exploded. It sweeps beams of radiation toward Earth like a lighthouse. Fermi observes gamma rays, the highest-energy form of light, and detects only a tiny number of photons from any one pulsar, mixed in with background noise from other sources and the wider universe.

To pull out the signal, researchers must guess several unknowns at once: how fast the star spins, how quickly that spin is slowing, and exactly where it sits in the sky. That makes a four-dimensional grid of possibilities. “There’s just not enough computing power on Earth to do it,” said lead author Daniel Pook-Kolb of the Max Planck Institute for Gravitational Physics (Albert Einstein Institute) in Germany, describing a search of every combination.

Gamma-ray map of the whole sky built from Fermi Large Area Telescope observations, with the bright band of the Milky Way across the middle
Image: NASA/DOE/Fermi LAT Collaboration

Until now, teams have split the Fermi data into smaller chunks and shared the work among volunteers around the world, who lend their computers’ unused processing power. Even so, finding a single pulsar typically took the equivalent of hundreds of computer-years.

Turning the search into a geometry puzzle

Dotan Gazith and Barak Zackay of the Weizmann Institute of Science in Israel, co-authors of the study, saw that the search could be recast as finding the shortest vector in a high-dimensional lattice, which is a regularly spaced array of points. The lattice technique comes from algorithms that cryptographers developed while trying to break encryption designed to withstand future quantum computers. It has also been used to make satellite navigation systems more efficient.

The algorithm is light enough for everyday hardware: NASA says a single trial took roughly five minutes on one processor core, and a standard laptop has between two and eight of them. Once the method flagged a candidate, the team confirmed it with the full Fermi data set. NASA says the signal was too strong to be a chance occurrence.

The next target: Pulsars with a companion star

The team now wants to find pulsars in binary systems, where a pulsar orbits another star. The companion’s gravity distorts the pulsar’s signal, which has made these objects essentially impossible to find with traditional methods. Colin Clark, an astrophysicist at the Max Planck Institute and a co-author, said the data probably hold “tens or maybe even hundreds” of such pulsars. The plan is to combine the new algorithm with the volunteer computing of the Einstein@Home program.

Open data, and what it does to the search

The result is a clear case for open data, and the numbers around it show the potential. Fermi launched on June 11, 2008, according to NASA’s mission page, and the same page credits it with finding some 300 gamma-ray pulsars or more. A method that cuts the cost of each find so sharply could add to that tally without any new hardware in orbit.

The volunteer side is also large. Einstein@Home, the project the team plans to pair the algorithm with, says on its website that it has over 500,000 volunteers. Work that once needed all of them for hundreds of computer-years per pulsar could now, for isolated pulsars, run on a laptop. The volunteers’ power can then go toward the harder binary cases.

Fermi scans the whole sky every three hours, and its data become fully public after an initial one-year calibration period. “Any search that one team is doing, another team can just go and do that and check the search,” said Elizabeth Hays, Fermi project scientist at NASA’s Goddard Space Flight Center. The cryptography-inspired team came from outside the Fermi mission and could still apply a completely new approach, helped by NASA’s open Fermitools software for preparing the data.

TAGGED:Fermi,NASA

Leave a Reply

Your email address will not be published. Required fields are marked *

You may also like