Fast radio burst detection in the presence of coloured noise

TL;DR

This paper examines how correlated noise affects fast radio burst detection, revealing increased false alarms and altered signal characteristics, and demonstrates a modelling approach using real observational data.

Contribution

It introduces a correlated noise modelling method for FRB searches and applies it to real data, improving false positive analysis in noisy environments.

Findings

Correlated noise significantly raises false alarm probability.

False positives tend to have higher S/N and wider pulse widths.

The modelling accurately reproduces candidate distributions in observational data.

Abstract

In this paper, we investigate the impact of correlated noise on fast radio burst (FRB) searching. We found that 1) the correlated noise significantly increases the false alarm probability; 2) the signal-to-noise ratios (S/N) of the false positives become higher; 3) the correlated noise also affects the pulse width distribution of false positives, and there will be more false positives with wider pulse width. We use 55-hour observation for M82 galaxy carried out at Nanshan 26m radio telescope to demonstrate the application of the correlated noise modelling. The number of candidates and parameter distribution of the false positives can be reproduced with the modelling of correlated noise. We will also discuss a low S/N candidate detected in the observation, for which we demonstrate the method to evaluate the false alarm probability in the presence of correlated noise.Possible origins of the candidate are discussed, where two possible pictures, an M82-harbored giant pulse and a cosmological FRB, are both compatible with the observation.