Efficient Summation of Arbitrary Masks -- ESAM

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Gupta, Vivek, Bannister, Keith, Flynn, Chris, James, Clancy
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910745215631360
author Gupta, Vivek
Bannister, Keith
Flynn, Chris
James, Clancy
author_facet Gupta, Vivek
Bannister, Keith
Flynn, Chris
James, Clancy
contents Searches for impulsive, astrophysical transients are often highly computationally demanding. A notable example is the dedispersion process required for performing blind searches for Fast Radio Bursts (FRBs) in radio telescope data. We introduce a novel approach - Efficient Summation of Arbitrary Masks (ESAM) - which efficiently computes 1-D convolution of many arbitrary 2-D masks, and can be used to carry out dedispersion over thousands of dispersion trials efficiently. Our method matches the accuracy of the traditional brute force technique in recovering the desired Signal-to-Noise ratio (S/N) while reducing computational cost by around a factor of 10. We compare its performance with existing dedispersion algorithms, such as the Fast Dispersion Measure Transform (FDMT) algorithm, and demonstrate how ESAM provides freedom to choose arbitrary masks and further optimise computational cost versus accuracy. We explore the potential applications of ESAM beyond FRB searches.
format Preprint
id arxiv_https___arxiv_org_abs_2412_10678
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Efficient Summation of Arbitrary Masks -- ESAM
Gupta, Vivek
Bannister, Keith
Flynn, Chris
James, Clancy
Instrumentation and Methods for Astrophysics
Data Analysis, Statistics and Probability
Searches for impulsive, astrophysical transients are often highly computationally demanding. A notable example is the dedispersion process required for performing blind searches for Fast Radio Bursts (FRBs) in radio telescope data. We introduce a novel approach - Efficient Summation of Arbitrary Masks (ESAM) - which efficiently computes 1-D convolution of many arbitrary 2-D masks, and can be used to carry out dedispersion over thousands of dispersion trials efficiently. Our method matches the accuracy of the traditional brute force technique in recovering the desired Signal-to-Noise ratio (S/N) while reducing computational cost by around a factor of 10. We compare its performance with existing dedispersion algorithms, such as the Fast Dispersion Measure Transform (FDMT) algorithm, and demonstrate how ESAM provides freedom to choose arbitrary masks and further optimise computational cost versus accuracy. We explore the potential applications of ESAM beyond FRB searches.
title Efficient Summation of Arbitrary Masks -- ESAM
topic Instrumentation and Methods for Astrophysics
Data Analysis, Statistics and Probability
url https://arxiv.org/abs/2412.10678