Search for continuous gravitational waves from 20 accreting millisecond x-ray pulsars in O3 LIGO data

Abstract

Results are presented of searches for continuous gravitational waves from 20 accreting millisecond x-ray pulsars with accurately measured spin frequencies and orbital parameters, using data from the third observing run of the Advanced LIGO and Advanced Virgo detectors. The search algorithm uses a hidden Markov model, where the transition probabilities allow the frequency to wander according to an unbiased random walk, while the J -statistic maximum-likelihood matched filter tracks the binary orbital phase. Three narrow subbands are searched for each target, centered on harmonics of the measured spin frequency. The search yields 16 candidates, consistent with a false alarm probability of 30% per subband and target searched. These candidates, along with one candidate from an additional target-of-opportunity search done for SAX J1808.4 − 3658, which was in outburst during one month of the observing run, cannot be confidently associated with a known noise source. Additional follow-up does not provide convincing evidence that any are a true astrophysical signal. When all candidates are assumed nonastrophysical, upper limits are set on the maximum wave strain detectable at 95% confidence, h 95% 0 . The strictest constraint is h 95% 0 ¼ 4.7 × 10 −26 from IGR J17062 − 6143. Constraints on the detectable wave strain from each target lead to constraints on neutron star ellipticity and r-mode amplitude, the strictest of which are ϵ 95% ¼ 3.1 × 10 −7 and α 95% ¼ 1.8 × 10 −5 respectively. This analysis is the most comprehensive and sensitive search of continuous gravitational waves from accreting millisecond x-ray pulsars to date.

Validating continuous gravitational-wave candidates from a semicoherent search using Doppler modulation and an effective point spread function

“…Here, δ qi (t N T ) is the likelihood of the path with the maximum probability ending in state q i (1 ≤ i ≤ N Q ) at step N T and δ q * (t N T ) is the likelihood of the optimal Viterbi path. In some applications of the HMM algorithm, e.g., [19,[22][23][24][25], the total log-likelihood along the Viterbi path, ln δ q * (t N T ), is directly used as the detection statistic. Transforming to the Viterbi score in Eq.…”

“…[21,38]). In a recent search for CWs from accreting millisecond X-ray pulsars, for each candidate that remained after the initial veto procedures, the log-likelihood is computed for a grid of off-target positions around the source's true position [24]. Every candidate whose log likelihood contours do not match simulations in Gaussian noise gets eliminated.…”

“…In this paper we focus on a computationally efficient, semicoherent search strategy based on a hidden Markov model (HMM) which is equipped to track a signal frequency that wanders stochastically and spins down secularly [14,15]. The tracking scheme has its origins in engineering applications, and has recently been used in many directed CW searches (e.g., [16][17][18][19][20][21][22][23][24][25]).…”

“…[21]. Some searches rely on an unnormalized log-likelihood instead of the Viterbi score to evaluate the significance of a candidate [19,[22][23][24][25].…”

“…If the candidates become louder, they are vetoed. Despite the wide applications of these two vetoes in existing searches [21][22][23][24], the general safety (i.e., no astrophysical signal is falsely eliminated) of these veto procedures is still being studied for various search configurations, e.g., for HMM-based searches.…”