International Summer Research Program in Gravitational-Wave Physics:
Research Experiences for Undergraduates around the world

Seoul National University (SNU)
  • Gravitational Waves from Pulsars after Glitches:
    Several pulsars are known to show sudden glitches in their pulse periods, but the exact mechanism is not well understood yet. At SNU, we have been carrying out numerical simulations of pulsar dynamics using an axisymmetric code. Our analysis is currently limited to axisymmetric perturbation only because of the nature of the code. We are in the process of developing a full 3-D code that is capable of carrying our more realistic simulations. As an integral part of this research effort, the student would create glitch models capable of exploring the contrast between results from the 2-D and 3-D codes, and would be responsible for carrying out simulations for such comparison tests.
    Mentor(s): Hyung Mok Lee, Jinho KIM (Post-doc), Heeil Kim (Post-doc)
    Related Project 2011: "Simulations of Pulsar Glitches for the Detection of Gravitational Waves"
    Related Project 2013: "Pulsar Glitches As A Gravitational Wave Source"
  • Glitch Identification using Machine Learning Algorithm:
    Glitches are the false signals that leads to the confusion with real signals. The exact reasons of glitches are not always well understood, and the individual identification of them is very time consuming. If the glitches are related to noise caused by the environment or any component of the system, they may be correlated with the auxiliary channels that monitor environmental parameters and various parts of the system. However, it is difficult to correlate the glitches with auxiliary channel data since there are thousands of channels in the LIGO system. We are developing glitch identification methods based on machine learning algorithms. The IREU student would collaborate with KGWG members on the development of the classifier, using machine learning algorithms in order to better understanding the glitches.
    Mentor(s): Hyung Mok Lee, Sang-Hoon Oh (Research Scientist at NIMS, Deajeon)
  • Parameter Estimation consistency between MCMC and Fisher Information Matrix:
    Gravitational wave data allows us to make accurate estimation of parameters for sources. Using the Fisher information matrix is a straightforward and fast way to evaluate the accuracy of the parameter estimation. However, the application of Fisher matrix is limited to strong sources with large signal-to-noise ratios (SNR). Moreover, the standard Fisher matrix approach is based on a simple linear correlation between the SNR and parameters of a given source. An effective Fisher matrix is more robust than the standard method, fitting the ambiguity function at the relevant S/N. The IREU student will apply the Markov Chain Monte Carlo (MCMC) method to compute the posterior functions of important source parameters for simulated gravitational-wave signals from inspirals of a neutron star and black hole binary pair. Varying SNR and total mass of the binary, the student will examine the consistency and accuracy of parameter estimation from the effective Fisher matrix approach when compared with results obtained from the MCMC method.
    Mentor(s): Hyung Mok Lee, Chung Lee Kim (Post-doc)
    Related Project 2013: "Parameter Estimation consistency between MCMC and Fisher Information Matrix"
  • SNU Project 2011: "Identifying Electromagnetic Transients Using Image Subtraction"

Past IREU Projects
Other Prior Projects