Using Genetic Algorithms to Optimize Antenna Designs for Improved Sensitivity to Ultra-High Energy Neutrinos

Genetic algorithms (GAs) are a type of computational optimization algorithms that emulate natural selection to “evolve" candidate solutions to a given problem. The GENETIS collaboration applies GAs to experimental design to efficiently optimize for improved performance. To this end, GENETIS has begun by designing GAs for the evolution of vertically polarized (VPol) antennas used in ultra-high energy (UHE) neutrino observatories. Due to the low flux of UHE neutrinos, as well as their small cross sections, it is essential to maximize the sensitivity of neutrino observatories at every step of the experiment. Neutrino observatories make use of radio signals produced by UHE neutrino interactions by observing large volumes of ice. The Askaryan Radio Array (ARA) achieves this by distributing stations of antennas across vast areas near the South Pole. Experiments like ARA measure their expected performance using simulation software that incorporates properties of the experiment and the physics of neutrino interactions in ice. The Physical Antenna Evolutionary Algorithm (PAEA) evolves the geometric properties of antennas within the physical constraints of specific UHE neutrino experiments and simulates their responses using EM simulation software XFdtd. To measure the performance of antenna designs, PAEA uses neutrino observatories’ simulation software with the evolved antennas’ responses included. This proceeding will discuss GENETIS’ evolution of VPol antenna designs for ARA and upcoming work on evolving more antenna designs for ARA and the Payload for Ultrahigh Energy Observations (PUEO). New efforts to capitalize on the birefringent properties of Antarctic ice to evolve experimental design will also be discussed.