Search NASA⌕ Search

SEARCH · Search NASA

Results for “Astronomy data analysis”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 631 records · Page 35

CARBON-ENHANCED METAL-POOR STARS IN SDSS/SEGUE. I. CARBON ABUNDANCE ESTIMATION AND FREQUENCY OF CEMP STARS

We describe a method for the determination of stellar [C/Fe] abundance ratios using low-resolution (R = 2000) stellar spectra from the Sloan Digital Sky Survey (SDSS) and its Galactic sub-survey, the Sloan Extension for Galactic Understanding and Exploration (SEGUE). By means of a star-by-star comparison with a set of SDSS/SEGUE spectra with available estimates of [C/Fe] based on published high-resolution analyses, we demonstrate that we can measure [C/Fe] from SDSS/SEGUE spectra with S/N ≥ 15 Å{sup –1} to a precision better than 0.35 dex for stars with atmospheric parameters in the range T {sub eff} = [4400, 6700] K, log g = [1.0, 5.0], [Fe/H] = [–4.0, +0.5], and [C/Fe] = [–0.25, +3.5]. Using the measured carbon-to-iron abundance ratios obtained by this technique, we derive the frequency of carbon-enhanced stars ([C/Fe] ≥ +0.7) as a function of [Fe/H], for both the SDSS/SEGUE stars and other samples from the literature. We find that the differential frequency slowly rises from almost zero to about 14% at [Fe/H] ∼ –2.4, followed by a sudden increase, by about a factor of three, to 39% from [Fe/H] ∼ –2.4 to [Fe/H] ∼ –3.7. Although the number of stars known with [Fe/H] < –4.0 remains small, the frequency of carbon-enhanced metal-poor (CEMP) stars below this value is around 75%. We also examine how the cumulative frequency of CEMP stars varies across different luminosity classes. The giant sample exhibits a cumulative CEMP frequency of 32% for [Fe/H] ≤ –2.5, 31% for [Fe/H] ≤ –3.0, and 33% for [Fe/H] ≤ –3.5; a roughly constant value. For the main-sequence turnoff stars, we obtain a lower cumulative CEMP frequency, around 10% for [Fe/H] ≤ –2.5, presumably due to the difficulty of identifying CEMP stars among warmer turnoff stars with weak CH G-bands. The dwarf population displays a large change in the cumulative frequency for CEMP stars below [Fe/H] = –2.5, jumping from 15% for [Fe/H] ≤ –2.5 to about 75% for [Fe/H] ≤ –3.0. When we impose a restriction with respect to distance from the Galactic mid-plane (|Z| < 5 kpc), the frequency of the CEMP giants does not increase at low metallicity ([Fe/H] < –2.5), but rather decreases due to the dilution of C-rich material in stars that have undergone mixing with CNO-processed material from their interiors. The frequency of CEMP stars near the main-sequence turnoff, which are not expected to have experienced mixing, increases for [Fe/H] ≤ –3.0. The general rise in the global CEMP frequency at low metallicity is likely due to the transition from the inner-halo to the outer-halo stellar populations with declining metallicity and increasing distance from the plane.

79 ASTRONOMY AND ASTROPHYSICS↗

Demographics of Sloan digital sky survey galaxies along the Hubble sequence

We present the statistical properties of a volume-limited sample of 7429 nearby (z = 0.033-0.044) galaxies from the Sloan Digital Sky Survey Data Release 7. Our database includes morphology distribution as well as the structural and spectroscopic properties of each morphology type based on the recent remeasurements of spectral line strengths by Oh and collaborators. Our database does not include galaxies that are apparently smaller and flatter because morphology classification of them turned out to be difficult. Our statistics confirmed the up-to-date knowledge of galaxy populations, e.g., correlations between morphology and line strengths as well as the derived ages. We hope that this database will be useful as a reference.

79 ASTRONOMY AND ASTROPHYSICS↗

SELF-CONSISTENT ORBITS AND PHYSICAL PROPERTIES FOR EIGHT SINGLE-LINED SPECTROSCOPIC BINARIES

For single-lined spectroscopic binaries (SB1s), self-consistent orbits can be determined via a simultaneous fit including both radial velocity data and the Hipparcos Intermediate Astrometric Data. Using the ratio λ of the semimajor axes of the photocenter to the primary orbits, the present paper develops an iterative fitting method. First, an initial orbit of an SB1 is determined with the assumption that λ equals 1. On the basis of this orbit, we use a stellar evolutionary model to distribute the total mass and luminosity to the primary and the secondary and update λ from the component mass and luminosity ratios. Then the orbit is updated using the updated value of λ, which completes a step of the iterative process. If this process is convergent, the resulting orbit is compatible with the physical properties of component stars. By using this method, the orbital solutions of eight SB1s as well as the physical properties are determined. We find that the component magnitude differences of these systems are very large, except for HIP 7143 and 45333. This suggests that only the secondaries of the two systems can be resolved with the present-day observational techniques, which is in agreement with recent observations.

79 ASTRONOMY AND ASTROPHYSICS↗

PRECISION MEASUREMENTS OF THE CLUSTER RED SEQUENCE USING AN ERROR-CORRECTED GAUSSIAN MIXTURE MODEL

The red sequence is an important feature of galaxy clusters and plays a crucial role in optical cluster detection. Measurement of the slope and scatter of the red sequence are affected both by selection of red sequence galaxies and measurement errors. In this paper, we describe a new error-corrected Gaussian Mixture Model for red sequence galaxy identification. Using this technique, we can remove the effects of measurement error and extract unbiased information about the intrinsic properties of the red sequence. We use this method to select red sequence galaxies in each of the 13,823 clusters in the maxBCG catalog, and measure the red sequence ridgeline location and scatter of each. These measurements provide precise constraints on the variation of the average red galaxy populations in the observed frame with redshift. We find that the scatter of the red sequence ridgeline increases mildly with redshift, and that the slope decreases with redshift. We also observe that the slope does not strongly depend on cluster richness. Using similar methods, we show that this behavior is mirrored in a spectroscopic sample of field galaxies, further emphasizing that ridgeline properties are independent of environment. These precise measurements serve as an important observational check on simulations and mock galaxy catalogs. The observed trends in the slope and scatter of the red sequence ridgeline with redshift are clues to possible intrinsic evolution of the cluster red sequence itself. Most importantly, the methods presented in this work lay the groundwork for further improvements in optically based cluster cosmology.

79 ASTRONOMY AND ASTROPHYSICS↗

A GROUP FINDING ALGORITHM FOR MULTIDIMENSIONAL DATA SETS

We describe a density-based hierarchical group finding algorithm capable of identifying structures and substructures of any shape and density in multidimensional data sets where each dimension can be a numeric attribute with arbitrary measurement scale. This has applications in a wide variety of fields from finding structures in galaxy redshift surveys, to identifying halos and subhalos in N-body simulations and group finding in Local Group chemodynamical data sets. In general, clustering schemes require an a priori definition of a metric (a non-negative function that gives the distance between two points in a space) and the quality of clustering depends upon this choice. The general practice is to use a constant global metric which is optimal only if the clusters in the data are self-similar. For complex data configurations even the most finely tuned constant global metric turns out to be suboptimal. Moreover, the correct choice of metric also becomes increasingly important as the number of dimensions increase. To address these problems, we present an entropy-based binary space partitioning algorithm which uses a locally adaptive metric for each data point. The metric is employed to calculate the density at each point and a list of its nearest neighbors, and this information is then used to form a hierarchy of groups. Finally, the ratio of maximum to minimum density of points in a group is used to estimate the significance of the groups. Setting a threshold on this significance can effectively screen out groups arising due to Poisson noise and helps organize the groups into meaningful clusters. For a data set of N points, the algorithm requires only O(N) space and O(N(log N){sup 3}) time which makes it ideally suitable for analyzing large data sets. As an example, we apply the algorithm to identify structures in a simulated stellar halo using the full six-dimensional phase space coordinates.

79 ASTRONOMY AND ASTROPHYSICS↗

IMPROVEMENT OF THE RICHNESS ESTIMATES OF maxBCG CLUSTERS

Minimizing the scatter between cluster mass and accessible observables is an important goal for cluster cosmology. In this work, we introduce a new matched filter richness estimator, and test its performance using the maxBCG cluster catalog. Our new estimator significantly reduces the variance in the L{sub X} -richness relation, from sigma{sub lnLx}{sup 2} = (0.86+-0.02){sup 2} to sigma{sub lnLx}{sup 2} = (0.69+-0.02){sup 2}. Relative to the maxBCG richness estimate, it also removes the strong redshift dependence of the L{sub X} -richness scaling relations, and is significantly more robust to photometric and redshift errors. These improvements are largely due to the better treatment of galaxy color data. We also demonstrate the scatter in the L{sub X} -richness relation depends on the aperture used to estimate cluster richness, and introduce a novel approach for optimizing said aperture which can easily be generalized to other mass tracers.

79 ASTRONOMY AND ASTROPHYSICS↗

IMPACTS OF SOURCE PROPERTIES ON STRONG LENSING BY RICH GALAXY CLUSTERS

We use a high-resolution N-body simulation to investigate the influence of background galaxy properties, including redshift, size, shape, and clustering, on the efficiency of forming giant arcs by gravitational lensing of rich galaxy clusters. Two large sets of ray-tracing simulations are carried out for 10 massive clusters at two redshifts, i.e., z {sub l} approx 0.2 and 0.3. The virial mass (M {sub vir}) of the simulated lens clusters at z approx 0.2 ranges from 6.8 x 10{sup 14} h {sup -1} M {sub sun} to 1.1 x 10{sup 15} h {sup -1} M {sub sun}. The information of background galaxies brighter than 25 mag in the I-band is taken from the Cosmological Evolution Survey (COSMOS) imaging data. Around 1.7 x 10{sup 5} strong lensing realizations with these images as background galaxies have been performed for each set. We find that the efficiency for forming giant arcs for z {sub l} = 0.2 clusters is broadly consistent with observations. Our study on control source samples shows that the number of giant arcs is decreased by a factor of 1.05 and 1.61 when the COSMOS redshift distribution of galaxies is adopted, compared to the cases where all the galaxies were assumed to be in a single source plane at z {sub l} = 1.0 and z {sub l} = 1.5, respectively. We find that the efficiency of producing giant arcs by rich clusters is weakly dependent on the source size and clustering. Our principal finding is that a small proportion (approx1/3) of galaxies with elongated shapes (e.g., ellipticity epsilon = 1 - b/a > 0.5) can boost the number of giant arcs substantially. Compared with recent studies where a uniform ellipticity distribution from 0 to 0.5 is used for the sources, the adoption of directly observed shape distribution increases the number of giant arcs by a factor of approx2. Our results indicate that it is necessary to account for source information and survey parameters (such as point-spread function, seeing) to make correct predictions of giant arcs and further to constrain the cosmological parameters.

79 ASTRONOMY AND ASTROPHYSICS↗

SIGNIFICANT FOREGROUND UNRELATED NON-ACOUSTIC ANISOTROPY ON THE 1 DEGREE SCALE IN WILKINSON MICROWAVE ANISOTROPY PROBE 5-YEAR OBSERVATIONS

The spectral variation of the cosmic microwave background (CMB) as observed by WMAP was tested using foreground reduced WMAP5 data, by producing subtraction maps at the 1 deg. angular resolution between the two cosmological bands of V and W, for masked sky areas that avoid the Galactic disk. The resulting V - W map revealed a non-acoustic signal over and above the WMAP5 pixel noise, with two main properties. First, it possesses quadrupole power at the approx1 muK level which may be attributed to foreground residuals. Second, it fluctuates also at all values of l> 2, especially on the 1 deg. scale (200 approx< l approx< 300). The behavior is random and symmetrical about zero temperature with an rms approx7 muK, or 10% of the maximum CMB anisotropy, which would require a 'cosmic conspiracy' among the foreground components if it is a consequence of their existence. Both anomalies must be properly diagnosed and corrected if 'precision' cosmology is the claim. The second anomaly is, however, more interesting because it opens the question on whether the CMB anisotropy genuinely represents primordial density seeds.

79 ASTRONOMY AND ASTROPHYSICS↗

RANDOM FORESTS FOR PHOTOMETRIC REDSHIFTS

The main challenge today in photometric redshift estimation is not in the accuracy but in understanding the uncertainties. We introduce an empirical method based on Random Forests to address these issues. The training algorithm builds a set of optimal decision trees on subsets of the available spectroscopic sample, which provide independent constraints on the redshift of each galaxy. The combined forest estimates have intriguing statistical properties, notable among which are Gaussian errors. We demonstrate the power of our approach on multi-color measurements of the Sloan Digital Sky Survey.

79 ASTRONOMY AND ASTROPHYSICS↗

A NEW GALACTIC EXTINCTION MAP OF THE CYGNUS REGION

We have made a Galactic extinction map of the Cygnus region with 5' spatial resolution. The selected area is 80{sup 0} to 90{sup 0} in the Galactic longitude and -4{sup 0} to 8{sup 0} in the Galactic latitude. The intensity at 140 {mu}m is derived from the intensities at 60 and 100 {mu}m of the IRAS data using the tight correlation between 60, 100, and 140 {mu}m found in the Galactic plane. The dust temperature and optical depth are calculated with 5' resolution from the 140 and 100 {mu}m intensity, and A{sub V} is calculated from the optical depth. In the selected area, the mean dust temperature is 17 K, the minimum is 16 K, and the maximum is 30 K. The mean A{sub V} is 6.5 mag, the minimum is 0.5 mag, and the maximum is 11 mag. The dust temperature distribution shows significant spatial variation on smaller scales down to 5'. Because the present study can trace the 5'-scale spatial variation of the extinction, it has an advantage over the previous studies, such as the one by Schlegel, Finkbeiner, and Davis, who used the COBE/DIRBE data to derive the dust temperature distribution with a spatial resolution of 1{sup 0}. The difference of A{sub V} between our map and Schlegel et al.'s is {+-} 3 mag. A new extinction map of the entire sky can be produced by applying the present method.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

THE ALLEN TELESCOPE ARRAY FLY'S EYE SURVEY FOR FAST RADIO TRANSIENTS

The relatively unexplored fast radio transient parameter space is known to be home to a variety of interesting sources, including pulsars, pulsar giant pulses, and non-thermal emission from planetary magnetospheres. In addition, a variety of hypothesized but as-yet-unobserved phenomena such as primordial black hole evaporation and prompt emission associated with coalescing massive objects have been suggested. The 2007 announcement by Lorimer et al. of the detection of a bright (30 Jy) radio pulse that was inferred to be of extragalactic origin and the subsequent consternation have demonstrated both the need for wide-field surveys characterizing the fast-transient parameter space and the potential utility of bright radio pulses as probes of the interstellar medium and intergalactic medium. Here we present results from the 450 hr, 150 deg{sup 2} Fly's Eye survey for bright dispersed radio pulses at the Allen Telescope Array (ATA). The Fly's Eye Spectrometer produces 128 channel power spectra over a 209 MHz bandwidth, centered at 1430 MHz, on 44 independent signal paths originating with 30 independent ATA antennae. Data were dedispersed between 0 and 2000 pc cm{sup -3} and searched for pulses with dispersion measures greater than 50 pc cm{sup -3} between 625 {mu}s and 5 s in duration. No pulses were detected in the survey, implying a limiting rate of less than 2 sky{sup -1} hr{sup -1} for 10 ms duration pulses having apparent energy densities greater than 440 kJy {mu}s, or mean flux densities greater than 44 Jy. Here we present details of the instrument, experiment, and observations, including a discussion of our results in light of other single pulse searches.

79 ASTRONOMY AND ASTROPHYSICS↗

ACTIVE LEARNING TO OVERCOME SAMPLE SELECTION BIAS: APPLICATION TO PHOTOMETRIC VARIABLE STAR CLASSIFICATION

Despite the great promise of machine-learning algorithms to classify and predict astrophysical parameters for the vast numbers of astrophysical sources and transients observed in large-scale surveys, the peculiarities of the training data often manifest as strongly biased predictions on the data of interest. Typically, training sets are derived from historical surveys of brighter, more nearby objects than those from more extensive, deeper surveys (testing data). This sample selection bias can cause catastrophic errors in predictions on the testing data because (1) standard assumptions for machine-learned model selection procedures break down and (2) dense regions of testing space might be completely devoid of training data. We explore possible remedies to sample selection bias, including importance weighting, co-training, and active learning (AL). We argue that AL-where the data whose inclusion in the training set would most improve predictions on the testing set are queried for manual follow-up-is an effective approach and is appropriate for many astronomical applications. For a variable star classification problem on a well-studied set of stars from Hipparcos and Optical Gravitational Lensing Experiment, AL is the optimal method in terms of error rate on the testing data, beating the off-the-shelf classifier by 3.4% and the other proposed methods by at least 3.0%. To aid with manual labeling of variable stars, we developed a Web interface which allows for easy light curve visualization and querying of external databases. Finally, we apply AL to classify variable stars in the All Sky Automated Survey, finding dramatic improvement in our agreement with the ASAS Catalog of Variable Stars, from 65.5% to 79.5%, and a significant increase in the classifier's average confidence for the testing set, from 14.6% to 42.9%, after a few AL iterations.

79 ASTRONOMY AND ASTROPHYSICS↗

DECORRELATION TIMES OF PHOTOSPHERIC FIELDS AND FLOWS

We use autocorrelation to investigate evolution in flow fields inferred by applying Fourier local correlation tracking (FLCT) to a sequence of high-resolution (0.''3), high-cadence ({approx_equal} 2 minute) line-of-sight magnetograms of NOAA active region (AR) 10930 recorded by the narrowband filter imager of the Solar Optical Telescope aboard the Hinode satellite over 2006 December 12 and 13. To baseline the timescales of flow evolution, we also autocorrelated the magnetograms, at several spatial binnings, to characterize the lifetimes of active region magnetic structures versus spatial scale. Autocorrelation of flow maps can be used to optimize tracking parameters, to understand tracking algorithms' susceptibility to noise, and to estimate flow lifetimes. Tracking parameters varied include: time interval {Delta}t between magnetogram pairs tracked, spatial binning applied to the magnetograms, and windowing parameter {sigma} used in FLCT. Flow structures vary over a range of spatial and temporal scales (including unresolved scales), so tracked flows represent a local average of the flow over a particular range of space and time. We define flow lifetime to be the flow decorrelation time, {tau}. For {Delta}t > {tau}, tracking results represent the average velocity over one or more flow lifetimes. We analyze lifetimes of flow components, divergences, and curls as functions of magnetic field strength and spatial scale. We find a significant trend of increasing lifetimes of flow components, divergences, and curls with field strength, consistent with Lorentz forces partially governing flows in the active photosphere, as well as strong trends of increasing flow lifetime and decreasing magnitudes with increases in both spatial scale and {Delta}t.

79 ASTRONOMY AND ASTROPHYSICS↗

THE CFHTLS-STRONG LENSING LEGACY SURVEY (SL2S): INVESTIGATING THE GROUP-SCALE LENSES WITH THE SARCS SAMPLE

We present the Strong Lensing Legacy Survey-ARCS (SARCS) sample compiled from the final T0006 data release of the Canada-France-Hawaii Telescope Legacy Survey (CFHTLS) covering a total non-overlapping area of 159 deg{sup 2}. We adopt a semi-automatic method to find gravitational arcs in the survey that makes use of an arc-finding algorithm. The candidate list is pruned by visual inspection and ranking to form the final SARCS sample. This list also includes some serendipitously discovered lens candidates which the automated algorithm did not detect. The SARCS sample consists of 127 lens candidates which span arc radii {approx}2''-18'' within the unmasked area of {approx}150 deg{sup 2}. Within the sample, 54 systems are promising lenses among which, we find 12 giant arcs (length-to-width ratio {>=}8). We also find two radial arc candidates in SL2SJ141447+544704. From our sample, we detect a systematic alignment of the giant arcs with the major axis of the baryonic component of the putative lens in concordance with previous studies. This alignment is also observed for all arcs in the sample and does not vary significantly with increasing arc radius. The mean values of the photometric redshift distributions of lenses corresponding to the giant arcs and all arcs sample are at z {approx} 0.6. Owing to the large area and depth of the CFHTLS, we find the largest sample of lenses probing mass scales that are intermediate to cluster and galaxy lenses for the first time. We compare the observed image separation distribution (ISD) of our arcs with theoretical models. A two-component density profile for the lenses which accounts for both the central galaxy and the dark matter component is required by the data to explain the observed ISD. Unfortunately, current levels of uncertainties and degeneracies accommodate models both with and without adiabatic contraction. We also show the effects of changing parameters of the model that predict the ISD and that a larger lens sample might constrain relations such as the concentration-mass relation, mass-luminosity relation, and the faint-end slope of the luminosity function.

79 ASTRONOMY AND ASTROPHYSICS↗

THE zCOSMOS 20k GROUP CATALOG

We present an optical group catalog between 0.1 {approx}< z {approx}< 1 based on 16,500 high-quality spectroscopic redshifts in the completed zCOSMOS-bright survey. The catalog published herein contains 1498 groups in total and 192 groups with more than five observed members. The catalog includes both group properties and the identification of the member galaxies. Based on mock catalogs, the completeness and purity of groups with three and more members should be both about 83% with respect to all groups that should have been detectable within the survey, and more than 75% of the groups should exhibit a one-to-one correspondence to the 'real' groups. Particularly at high redshift, there are apparently more galaxies in groups in the COSMOS field than expected from mock catalogs. We detect clear evidence for the growth of cosmic structure over the last seven billion years in the sense that the fraction of galaxies that are found in groups (in volume-limited samples) increases significantly with cosmic time. In the second part of the paper, we develop a method for associating galaxies that only have photo-z to our spectroscopically identified groups. We show that this leads to improved definition of group centers, improved identification of the most massive galaxies in the groups, and improved identification of central and satellite galaxies, where we define the former to be galaxies at the minimum of the gravitational potential wells. Subsamples of centrals and satellites in the groups can be defined with purities up to 80%, while a straight binary classification of all group and non-group galaxies into centrals and satellites achieves purities of 85% and 75%, respectively, for the spectroscopic sample.

79 ASTRONOMY AND ASTROPHYSICS↗

BAYESIAN CONFIDENCE LIMITS OF ELECTRON SPECTRA OBTAINED THROUGH REGULARIZED INVERSION OF SOLAR HARD X-RAY SPECTRA

Many astrophysical observations are characterized by a single, non-repeatable measurement of a source brightness or intensity, from which we are to construct estimates for the true intensity and its uncertainty. For example, the hard X-ray count spectrum from transient events such as solar flares can be observed only once, and from this single spectrum one must determine the best estimate of the underlying source spectrum I({epsilon}), and hence the form of the responsible electron spectrum F(E). Including statistical uncertainties on the measured count spectrum yields a 'confidence strip' that delineates the boundaries of electron spectra that are consistent with the observed photon spectrum. In this short article, we point out that the expectation values of the source brightness and its variance in a given photon energy bin are in general not (as has been assumed in prior works) equal to n, the number of counts observed in that energy bin. Rather, they depend both on n and on prior knowledge of the overall photon spectrum. Using Bayesian statistics, we provide an explicit procedure and formulas for determining the 'confidence strip' (Bayesian credible region) for F(E), thus providing rigorous bounds on the intensity and shape of the accelerated electron spectrum.

79 ASTRONOMY AND ASTROPHYSICS↗

DOES A ''STOCHASTIC'' BACKGROUND OF GRAVITATIONAL WAVES EXIST IN THE PULSAR TIMING BAND?

We investigate the effects of gravitational waves (GWs) from a simulated population of binary supermassive black holes (SMBHs) on pulsar timing array data sets. We construct a distribution describing the binary SMBH population from an existing semi-analytic galaxy formation model. Using realizations of the binary SMBH population generated from this distribution, we simulate pulsar timing data sets with GW-induced variations. We find that the statistics of these variations do not correspond to an isotropic, stochastic GW background. The ''Hellings and Downs'' correlations between simulated data sets for different pulsars are recovered on average, though the scatter of the correlation estimates is greater than expected for an isotropic, stochastic GW background. These results are attributable to the fact that just a few GW sources dominate the GW-induced variations in every Fourier frequency bin of a five-year data set. Current constraints on the amplitude of the GW signal from binary SMBHs will be biased. Individual binary systems are likely to be detectable in five-year pulsar timing array data sets where the noise is dominated by GW-induced variations. Searches for GWs in pulsar timing array data therefore need to account for the effects of individual sources of GWs.

79 ASTRONOMY AND ASTROPHYSICS↗

OPTIMAL CORRELATION ESTIMATORS FOR QUANTIZED SIGNALS

Using a maximum-likelihood criterion, we derive optimal correlation strategies for signals with and without digitization. We assume that the signals are drawn from zero-mean Gaussian distributions, as is expected in radio-astronomical applications, and we present correlation estimators both with and without a priori knowledge of the signal variances. We demonstrate that traditional estimators of correlation, which rely on averaging products, exhibit large and paradoxical noise when the correlation is strong. However, we also show that these estimators are fully optimal in the limit of vanishing correlation. We calculate the bias and noise in each of these estimators and discuss their suitability for implementation in modern digital correlators.

79 ASTRONOMY AND ASTROPHYSICS↗