Gamma-ray astronomy

Abstract:

The identification, and subsequent discovery, of fast radio transients through blind-search surveys requires a large amount of processing power, in worst cases scaling as $\mathcal{O}(N^3)$. For this reason, survey data are generally processed offline, using high-performance computing architectures or hardware-based designs. In recent years, graphics processing units have been extensively used for numerical analysis and scientific simulations, especially after the introduction of new high-level application programming interfaces. Here we show how GPUs can be used for fast transient discovery in real-time. We present a solution to the problem of de-dispersion, providing performance comparisons with a typical computing machine and traditional pulsar processing software. We describe the architecture of a real-time, GPU-based transient search machine. In terms of performance, our GPU solution provides a speed-up factor of between 50 and 200, depending on the parameters of the search.

Abstract:

One of the tenets of the radio pulsar observational picture is that the integrated pulse profiles are constant with time. This assumption underpins much of the fantastic science made possible via pulsar timing. Over the past few years, however, this assumption has come under question with a number of pulsars showing pulse shape changes on a range of timescales. Here, we show the dramatic appearance of a bright component in the pulse profile of PSR J0738-4042 (B0736-40). The component arises on the leading edge of the profile. It was not present in 2004 but strongly present in 2006 and all observations thereafter. A subsequent search through the literature shows the additional component varies in flux density over timescales of decades. We show that the polarization properties of the transient component are consistent with the picture of competing orthogonal polarization modes. Faced with the general problem of identifying and characterising average profile changes, we outline and apply a statistical technique based on a Hidden Markov Model. The value of this technique is established through simulations, and is shown to work successfully in the case of low signal-to-noise profiles.