Written by AZoQuantumJul 19 2019

**In the past few decades, the exponential increase in computer power and the associated increase in the quality of algorithms have facilitated particle and theoretical physicists to carry out more complicated and precise simulations of fundamental particles and their interactions.**

If the number of lattice points in a simulation is increased, it becomes tougher to tell the difference between the experimental result of the simulation and the surrounding noise. A new research by Marco Ce, a physicist based at the Helmholtz-Institut Mainz in Germany and recently published in *EPJ Plus*, illustrates a method for simulating particle ensembles that are “large” (at least by the principles of particle physics).

This enhances the signal-to-noise ratio and thus the accuracy of the simulation; importantly, it also can be used to model ensembles of baryons: a class of elementary particles that comprises the neutrons and protons that form atomic nuclei.

Ce's simulations use a Monte Carlo algorithm: a generic computational technique that depends on repetitive random sampling to acquire numerical results. These algorithms have a wide range of uses, and in mathematical physics, they are especially ideal for assessing complex integrals, and to modeling systems with several degrees of freedom.

More exactly, the type of Monte Carlo algorithm used here includes multi-level sampling. This means that the samples are taken with diverse levels of accuracy, which is less computationally costly than techniques wherein the sampling accuracy is even.

Multi-level Monte Carlo techniques have earlier been applied to ensembles of bosons (the class of particle that, clearly, includes the currently well-known Higgs particle), but not to the more complicated fermions. This latter category includes both electrons and baryons: all the key components of “everyday” matter.

Ce completes his study by remarking that there are a number of other issues in particle physics where computation is influenced by high signal-to-noise ratios, and which might gain from this method.

Source: http://www.springer.com