Xiaoqian He

Date of Graduation

2018

Document Type

Thesis

Degree Type

MS

College

Statler College of Engineering and Mineral Resources

Department

Lane Department of Computer Science and Electrical Engineering

Committee Chair

Natalia Schmid

Committee Co-Chair

Kevin Bandura

Committee Member

Maura McLaughlin

Abstract

In this thesis we propose a new blind algorithm, which we call Energy in Transformed Data (ETD), for the detection of isolated astrophysical pulses. Unlike the conventional algorithm, which requires de-dispersion, integration over frequency channels and matched filtering, the ETD algorithm detects pulses in a space reciprocal to the space of filterbank data. The ETD algorithm is implemented in three steps: (1) it applies the Spatial Fourier Transform (SFT) to filterbank data, (2) it evaluates the energy of signal in the transformed space, and (3) it compares the value of the total energy to a threshold. If the energy in transformed data exceeds the threshold, we claim that we detect a potential pulse.;To analyze the detection capabilities of the ETD algorithm, we applied it to data of ten different Rotating Radio Transients (RRATs). The performance of the algorithm was compared to the performance of two other algorithms, one of which is the conventional algorithm and the other one is an algorithm that performs matched filtering in the SFT domain. The performance of the ETD algorithm depends on the Signal-to-Noise ratio (SNR) of pulses. When the SNR threshold is set to a large value such as 8, the ETD algorithm can detect 80.5% of all pulses detected by the conventional algorithm. When the SNR threshold is set to 5, the ETD algorithm can detect only 40.4% detectable by the conventional algorithm.;The ability to detect broad and relatively weak astronomical pulses is a special feature of ETD. As we demonstrate it in this thesis, the energy of pulses with a broad pulse profile (broad compared to the time delay of the pulse arrival at the lower frequency of receiver) is concentrated around the origin in the SFT domain. This allows to easily detect broad pulses by ETD.;The main drawback of the ETD algorithm is in its inability to detect pulses which are relatively low in SNR value and have a narrow pulse profile. Analyzing the main advantages and disadvantages of ETD, we conclude that the algorithm can be used as a complementary method to the conventional algorithm, especially when the task is to detect broad pulses parameterized by a low SNR value. In addition to its use for current observations, we expect this method will be beneficial for detection of other types of isolated transients.

Download

Share

COinS