Fourier Transforms of Color Glass Condensate Multi-Wilson-Line Correlators via Filon Quadrature

At the heart of modern physics lies a quest to understand how matter behaves when squeezed to its absolute limit. Inside protons and atomic nuclei, particles called gluons carry the force that binds quarks together. When these particles are smashed together at incredibly high speeds, the density of gluons becomes so immense that they begin to overlap and interact in a chaotic, collective state known as the Color Glass Condensate. This state is not a solid or a liquid, but a unique regime of matter where the rules of quantum mechanics and high-energy collisions merge. To predict what happens when scientists collide these particles in machines like the Large Hadron Collider, physicists must calculate the probability of specific outcomes, such as the creation of pairs of particle jets. These calculations require translating information from a map of positions into a map of speeds, a mathematical process that involves summing up countless tiny contributions that oscillate and cancel each other out in complex ways.

For decades, calculating these probabilities for the Color Glass Condensate has been a bottleneck. The mathematics required to move from the position of particles to their momentum involves a type of integral that is notoriously difficult to solve on a computer. The numbers involved oscillate so rapidly that standard calculation methods would need to take billions of tiny steps to get a single accurate answer, a task that would take a supercomputer days or even weeks to complete for just one scenario. This limitation has forced researchers to rely on simplified approximations that ignore the intricate connections between particles, potentially missing the very physics that explains how matter saturates and stabilizes at high energies. Without a way to perform these calculations quickly and accurately, comparing theoretical predictions with experimental data has remained a slow and often imprecise endeavor.

In a new study, a team of researchers from Central China Normal University has developed a powerful new method to break this deadlock. They have created a specialized algorithm that can perform these difficult calculations in a fraction of the time previously thought possible. Instead of trying to resolve every single rapid oscillation step-by-step, their approach treats the oscillating parts of the equation as a known pattern. By pre-calculating the behavior of these patterns, the team turned a massive, slow calculation into a series of simple, fast multiplications. This technique, known as Filon quadrature, allows the computer to skip the tedious work of resolving every tiny wiggle in the data, effectively compressing hours of processing time into mere minutes.

The researchers tested this new method on a specific, high-stakes problem: calculating the production of particle pairs in deep inelastic scattering, a process where a high-energy electron smashes into a proton. In this scenario, the mathematics involves a four-point connection between particles, which is far more complex than the simpler two-point connections often used in approximations. The team first had to solve a numerical instability that arises when the standard formulas for these connections approach zero, a situation that previously caused computers to crash or produce garbage results. They devised a mathematical reformulation that smoothed out this singularity, ensuring the calculations remained stable even in the most extreme conditions. Once this hurdle was cleared, they applied their new quadrature method to the problem.

The results were dramatic. On a standard multi-core computer processor, the calculation of a single particle configuration took several hours. By porting the algorithm to a Graphics Processing Unit (GPU)—a type of chip designed for handling massive amounts of parallel data—the team reduced this time to approximately two minutes. This speedup was not just a matter of using faster hardware; it was a fundamental change in how the math was structured. The researchers built custom software that fused the calculation of the particle connections directly with the final transformation step, eliminating the need to store massive intermediate tables of data in memory. This "fused" approach meant the computer could process the data as it was generated, discarding it immediately after use, which saved both time and memory.

To ensure their new method was not just fast but also correct, the team subjected it to rigorous testing. They compared their results against a highly accurate, slow reference calculation running on a traditional processor and found that the new method matched the old one perfectly, even in regions where the numbers were extremely small and prone to error. They also pushed the method to its limits, testing it with higher resolutions to see if it could handle the most difficult cases where previous methods had failed. In these tests, the new algorithm successfully recovered stable, positive results where lower-resolution attempts had produced nonsensical negative values, proving that the previous failures were due to insufficient detail rather than a flaw in the theory itself.

The team did not stop at the two-dimensional case relevant to electron-proton collisions. They generalized their algorithm to handle three sequential transformations, a requirement for more complex collisions involving atomic nuclei. Since the data for these three-dimensional cases would be too large to store in memory, they adapted their "fused" streaming approach to handle the data in real-time. They validated this generalized version against a family of mathematical functions where the answer is known exactly, confirming that their method could accurately reconstruct the final result from the raw data. This six-dimensional transform capability opens the door to calculating next-to-leading-order corrections for proton-nucleus and electron-ion collisions, processes that are currently beyond the reach of standard computational methods.

The significance of this work extends beyond a single calculation. By making these complex transforms fast and reliable, the researchers have provided a tool that can be used to interpret data from current and future particle colliders. The code they developed is publicly available, allowing other scientists to apply these techniques to their own problems without needing to rebuild the complex machinery from scratch. This advancement means that physicists can now explore the full range of particle interactions without relying on simplifying assumptions that might hide the true nature of the Color Glass Condensate. The ability to compute these probabilities in minutes rather than hours transforms what was once a theoretical curiosity into a practical tool for understanding the fundamental structure of matter.

The study concludes that while the specific application to nuclear collisions is still in the future, the foundation is now solid. The researchers have demonstrated that the difficult mathematical barriers preventing a full understanding of high-energy gluon saturation can be overcome with clever algorithmic design and modern hardware. Their work bridges the gap between abstract theory and experimental reality, offering a clear path forward for analyzing the next generation of high-energy physics data. By turning a prohibitively expensive calculation into a routine task, they have enabled a new level of precision in the study of the universe's most extreme states of matter.