
A newly discovered star in the southern sky rotates more than 260 times per second, sending regular pulses of energy toward Earth. Known as a pulsar, this type of star offers scientists valuable opportunities to study general relativity, gravitational waves, and even space navigation.
However, locating pulsars across the vast universe is extraordinarily demanding. An earlier attempt to detect this particular pulsar failed despite consuming computing time equivalent to 600 years on a single processor core.
Researchers at the Weizmann Institute of Science have now developed an algorithm that enabled them, alongside collaborators at the Max Planck Institute for Gravitational Physics in Hannover, Germany, and Chalmers University of Technology in Gothenburg, Sweden, to identify the star in just 100 minutes.
The discovery of the pulsar, named PSR J1622-7204, is described in a study published in Physical Review Letters. The new method is already helping researchers discover additional pulsars.
Pulsars are ultra-dense, rapidly rotating neutron stars that can form when stars eight to 20 times the mass of the Sun exhaust their nuclear fuel and collapse. They produce magnetic fields trillions of times stronger than Earth's, accelerating charged particles and generating powerful radiation beams.
As a pulsar spins, its radiation beam sweeps across space, briefly pointing toward Earth during each rotation. This creates the appearance of regularly timed energy pulses. Over time, pulsars lose energy, causing their rotation and pulse rates to gradually decrease.
The first pulsar was discovered in 1967 through its radio emissions, and nearly all pulsars identified over the following four decades were detected using radio waves. NASA's Fermi Gamma-ray Space Telescope, launched in 2008, opened new possibilities by observing high-energy radiation.
During its 18 years of operation, Fermi has identified nearly 1,000 regions emitting gamma rays with characteristics associated with pulsars. Yet confirming a pulsar requires identifying its precise location and rotational properties, including its spin frequency and the rate at which it slows.
"Trying to find the correct combination among all possibilities without a smart strategy forces us to test an astronomical number of guesses," explains Dr. Barak Zackay of Weizmann's Particle Physics and Astrophysics Department.
"Another challenge is that although a pulse of energy reaches Earth with every rotation of a pulsar, the Fermi telescope only manages to detect a single packet of energy - a photon - roughly once every 100 million rotations. In addition, the telescope itself is moving, orbiting Earth as Earth circles the Sun. All these factors make the computation required to find a pulsar extremely demanding."
A geometric solution to a physics problem
In 2025, Zackay's research group demonstrated a method for dramatically simplifying pulsar searches.
Their approach relies on the fact that each detected pulse marks a complete rotation of the star, returning its radiation beam to an orientation facing Earth. However, because Fermi detects only a tiny fraction of these pulses, astronomers cannot directly determine how many rotations occurred between observations.
The new algorithm addresses this missing information by transforming the problem into a mathematical structure called a lattice.
A lattice can be visualized as a grid in which movement is restricted to whole-number steps. Each intersection represents a particular combination of integers.
Researchers enter the arrival times of several dozen photons suspected of coming from an unidentified pulsar. The algorithm then constructs a lattice in which each point represents a possible number of rotations between observations, corresponding to a specific rotational frequency.
Incorrect hypotheses produce non-integer rotation counts when applied to the observed intervals. The algorithm seeks a solution that minimizes these deviations, using a mathematical approach known as the shortest vector problem.
By converting the physical challenge into a geometric one, researchers can combine, add, and subtract previous hypotheses to generate new possibilities more efficiently.
Although the calculation remains complex, established mathematical techniques can solve such problems. The Weizmann team adapted one of the most powerful available algorithms for this purpose.
A new lighthouse in the southern universe
Late last year, Zackay's laboratory began collaborating with the research group of Prof. Bruce Allen at the Max Planck Institute in Hannover to apply the method to promising pulsar candidates.
"We found the PSR J1622-7204 in a very short time," says Zackay.
"We ran the new algorithm 20 times, each time using a slightly different set of 80 photons that we suspected originated from the same pulsar, and one of those runs produced the right answer. Each run took just five minutes on a single CPU core, allowing us to determine the star's properties in less than two hours. We were somewhat lucky, however, since theoretical calculations showed it should take 300 runs to achieve a 95 percent probability of success."
The researchers also measured the pulsar's rotation rate and its gradual slowdown with high precision.
"We were able to calculate the pulsar's rotation rate with great precision and determine the rate at which it slows over time," says Dotan Gazith of Zackay's laboratory, who co-led the study with Dr. Daniel Pook-Kolb of Allen's group.
"Our calculations show that in about 100 million years, the star will complete roughly five fewer rotations per second than it does today. We have since expanded the search, and we are already discovering additional pulsars using this new method."
Often described as the universe's lighthouses, pulsars may also provide highly accurate navigation systems.
In 2018, a NASA experiment demonstrated that spacecraft could determine their position in space within a few kilometers by measuring the arrival times of pulses from different pulsars.
Extremely fast pulsars, rotating hundreds of times per second, are particularly valuable because they can function as exceptionally precise clocks.
Scientists worldwide are developing pulsar-based timing systems for navigation and for testing Einstein's general theory of relativity, introduced more than a century ago.
General relativity describes the universe as four-dimensional spacetime, consisting of length, width, height, and time. Massive objects distort this spacetime, while accelerating masses generate ripples known as gravitational waves.
When gravitational waves pass through Earth, they slightly distort local spacetime, potentially causing measurable changes in the arrival times of pulsar signals.
An international research effort is therefore monitoring a network of exceptionally stable pulsars to detect low-frequency gravitational waves through minute timing deviations.
Although high-frequency gravitational waves were first detected through other methods in 2015, scientists predict that many low-frequency waves remain to be observed.
Detecting them could reveal extraordinary cosmic events, including mergers of supermassive black holes, each potentially a billion times more massive than the Sun.
A distorted relationship
The newly discovered pulsar is isolated, but many of the fastest known pulsars belong to binary systems, in which a pulsar and a companion star orbit a shared center of mass.
These systems provide exceptional opportunities to test general relativity.
The companion star's gravity curves the surrounding spacetime, affecting the path of the pulsar's radiation. Scientists on Earth may therefore observe slight delays in the arrival of its signals.
Measuring these delays can reveal important characteristics of the companion star, including its mass, while allowing researchers to investigate additional relativistic effects.
Binary systems may also help explain the origins of heavy elements such as gold and uranium. Some theories suggest that processes ejecting matter from neutron stars in binary systems distribute these elements throughout the universe.
Until now, identifying pulsars in binary systems has been especially challenging because researchers generally needed information about the companion star before analyzing the pulsar.
"What has slowed the discovery of pulsars in binary systems until now is that we had to characterize the companion star before studying the pulsar itself," Zackay explains.
"The new, efficient method will allow us to find pulsars in binary systems blindly, without prior information about the companion, and even to learn about the companion's properties through observations of the pulsar. Estimates suggest that hundreds of such pulsars are waiting to be discovered."
Additional participants in the study included Dr. Colin J. Clark of the Max Planck Institute for Gravitational Physics in Hannover, Germany, and Prof. Erik Agrell of Chalmers University of Technology in Gothenburg, Sweden.
Dr. Barak Zackay's research is supported by the Shimon and Golde Picker - Weizmann Annual Grant.
