Frequency Analysis and Extrasolar Planets

Abstract

The discovery [13] (and confirmation [14]) of the first extrasolar planetary system by Wolszczan & Frail (1991) around the pulsar B1257+12 began a new era in this exciting field of astronomy. A few years later, in 1995, the first extrasolar planet around a normal star was found by Queloz & Mayor [10]. This was followed by other such discoveries [9]. The common point of all these findings is that the planets were indirectly deduced from the observations of their parent stars. In the case of the PSR B1257+12 the planets were found in the timing data of the pulsar and in the case of normal stars in their Doppler velocity measuerements. The usual approach to analyzing such data is to fit by meanas of the least-squares technique a Keplerian orbit (or orbits) that would account for the observed planetary effects. In most situations this is perfectly satisfactory. However, it is possible to show examples in which this technique fails. In this paper we show a new approach to the analysis of the timing observations of pulsars, as well as the Doppler velocity measurements of normal stars, when one suspects to have planetary signatures in the data. We disscus how this method, frequency analysis, can give more insight into to the data or even properly solve the problem when the least-squares fit of Keplerian orbits is not capable to do so.