Real-time Pulse Search and Positioning Method for Baseband Data Based on Matched Filtering

Fast radio bursts (FRBs) are intense radio transients that hold significant potential for astrophysical studies.Hightime-resolution pulse searching opens new possibilities for detecting narrow and faint pulses.However, the exponential growth in data volume presents considerable challenges for real-time processing.This paper proposes a real-time pulse search and positioning method based on matched filtering for baseband data processing.The method demonstrates superior performance in both pulse detection efficiency and dispersion tolerance compared to traditional approaches.Using GPU acceleration, it achieves real-time processing of 1 ns resolution baseband data, enabling the detection of narrower and fainter pulse signals.Validation using FRB 121102 baseband data revealed our method's capability to identify all previously known pulses and discover three new pulses, demonstrating its potential in advancing astrophysical research.