Searching for periodic sources with LIGO. II: Hierarchical searches

The detection of quasi-periodic sources of gravitational waves requires the accumulation of signal-to-noise over long observation times. If not removed, Earth-motion induced Doppler modulations, and intrinsic variations of the gravitational-wave frequency make the signals impossible to detect. These effects can be corrected (removed) using a parameterized model for the frequency evolution. We compute the number of independent corrections $N_p(\Delta T,N)$ required for incoherent search strategies which use stacked power spectra---a demodulated time series is divided into $N$ segments of length $\Delta T$, each segment is FFTed, the power is computed, and the $N$ spectra are summed up. We estimate that the sensitivity of an all-sky search that uses incoherent stacks is a factor of 2--4 better than would be achieved using coherent Fourier transforms; incoherent methods are computationally efficient at exploring large parameter spaces. A two-stage hierarchical search which yields another 20--60% improvement in sensitivity in all-sky searches for old (>= 1000 yr) slow (<= 200 Hz) pulsars, and for young (>= 40 yr) fast (<= 1000 Hz) pulsars. Assuming 10^{12} flops of effective computing power for data analysis, enhanced LIGO interferometers should be sensitive to: (i) Galactic core pulsars with gravitational ellipticities of $\epsilon\agt5\times 10^{-6}$ at 200 Hz, (ii) Gravitational waves emitted by the unstable r-modes of newborn neutron stars out to distances of ~8 Mpc, and (iii) neutron stars in LMXB's with x-ray fluxes which exceed $2 \times 10^{-8} erg/(cm^2 s)$. Moreover, gravitational waves from the neutron star in Sco X-1 should be detectable is the interferometer is operated in a signal-recycled, narrow-band configuration.