Search NASA⌕ Search

SEARCH · Search NASA

Results for “Generalization error”

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 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
NASA NTRS1974

A direct approach to the design of linear multivariable systems

Design of multivariable systems is considered and design procedures are formulated in the light of the most recent work on model matching. The word model matching is used exclusively to mean matching the input-output behavior of two systems. The term is used in the frequency domain to indicate the comparison of two transfer matrices containing transfer functions as elements. Design methods where non-interaction is not used as a criteria were studied. Two design methods are considered. The first method of design is based solely upon the specification of generalized error coefficients for each individual transfer function of the overall system transfer matrix. The second design method is called the pole fixing method because all the system poles are fixed at preassigned positions. The zeros of terms either above or below the diagonal are partially fixed via steady state error coefficients. The advantages and disadvantages of each method are discussed and an example is worked to demonstrate their uses. The special cases of triangular decoupling and minimum constraints are discussed.

Agrawal, B. L.↗
NASA NTRS1987

Nonuniform sampling techniques for antenna applications

A two-dimensional sampling technique, which can employ irregularly spaced samples (amplitude and phase) in order to generate the complete far-field patterns is presented. The technique implements a matrix inversion algorithm, which depends only on the nonuniform sampled data point locations and with no dependence on the actual field values at these points. A powerful simulation algorithm is presented to allow a real-life simulation of many reflector/feed configurations and to determine the usefulness of the nonuniform sampling technique for the copolar and cross-polar patterns. Additionally, an overlapped window concept and a generalized error simulation model are discussed to identify the stability of the technique for recovering the field data among the nonuniform sampled data. Numerical results are tailored for the pattern reconstruction of a 20-m offset reflector antenna operating at L-band. This reflector is planned to be used in a proposed measurement concept of large antenna aboard the Space Shuttle, whereby it would be almost impractical to accurately control the movement of the Shuttle with respect to the RF source in prescribed directions in order to generate uniform sampled points. Also, application of the nonuniform sampling technique to patterns obtained using near-field measured data is demonstrated. Finally, results of an actual far-field measurement are presented for the construction of patterns of a reflector antenna from a set of nonuniformly distributed measured amplitude and phase data.

Rahmat-Samii, Yahya↗
NASA NTRS1990

Cloud optical parameters as derived from the multispectral cloud radiometer

Here, researchers analyze the data taken in the marine stratocumulus intensive field observation component of the First ISCCP Regional Experiment (FIRE), conducted off the California coast in July, 1987 using the Multispectral Cloud Radiometer (MCR). Researchers installed an efficient retrieval scheme for deriving the cloud optical thickness and droplet mode radius, and have described the behavior of the retrieval error. Generally the scheme can retrieve the optical thickness and mode radius adequately unless they are too small; the use of the forward scattering region is more sound than the use of the backscattering portion. Applying the method to real data, researchers derived the two-dimensional distribution of optical thickness and mode radius for a portion of one of the FIRE marine stratocumulus missions. In this case study, the droplet size showed a more uniform distribution than optical thickness with some correlation between large droplet size and small optical thickness. Although researchers can find microphysical reasons for these tendencies, they suspect that the remotely sensed droplet size may be overestimated. As a future problem, they will compare results with in situ data of the droplet size distribution. Also it will be very important to check several reasons why the droplet radius might be overestimated, e.g., soot contamination, effect of escaping photons from the lateral sides of broken clouds, and so on.

Nakajima, Eruyuki↗
NASA NTRS1992

Land-mobile-satellite fade measurements in Australia

Attenuation measurements were implemented at L-band (1.5 GHz) in southeastern Australia during an 11-day period in October 1988 as part of a continuing examination of the propagation effects due to roadside trees and terrain for mobile-satellite service. Beacon transmissions from the geostationary ETS-V and IPORS satellites were observed. The Australian campaign expanded to another continent our Mobile Satellite Service data base of measurements executed in the eastern and southwestern United States regions. An empirical fade distribution model based on U.S. data predicted the Australian results with errors generally less than 1 dB in the 1-20 percent probability region. Directive antennas are shown to suffer deeper fades under severe shadowing conditions (3 dB excess at 4 percent), the equal-probability isolation between co- and cross-polarized transmissions deteriorated to 10 dB at the 5 dB fade level, and antenna diversity reception may reduce unavailability of the system by a factor of 2-8.

Vogel, Wolfhard J.↗
NASA NTRS1993

Atmospheric moisture fields derived by satellite observations over the tropical Pacific Ocean

Values of precipitable water are retrieved over the tropical and subtropical Pacific Ocean from TOVS infrared and microwave channel brightness temperature and OLR observations by means of stepwise linear regression. The most useful temperature and moisture sensing channels are pre-selected from sensitivity tests of a radiative transfer model. Numerous models are developed and tested against collocated radiosonde observations and Nimbus-7 SMMR precipitable water estimates. For RAOB comparisons, the best estimator used 15 TOVS predictors and captured 71.1 deg percent of the variance (+0.62 g/sq cm standard error) for column precipitable water; for precipitable water of 700-500 mb bulk layer, these values were 71.7 percent and +/- 0.17 g/sq cm. Little skill of estimated precipitable water was obtained for moisture above 500 mb. Regressions were less skillful against SMMR, unless collocation parameters were tightly controlled; SMMR was less acceptable than RAOB's because of observational drift and errors. Generally, the most skillful predictors were boundary layer brightness temperatures of TOVS channels and satellite estimated stability indices. 'Moisture channels' were hardly useful except for estimating middle and upper tropospheric moisture. Additional regression models were constructed testing the sensitivity to different observational and meteorological characteristics. Models which used some in situ observations surface observations or stabilities calculated from RAOB, were the most successful. The best of these explained 87.5 percent of the variance but the regression selected almost no TOVS channels, relying instead on conventional RAOB and surface observations. A set of four regression models were developed, stratifying atmospheric characteristics on the basis of collocated OLR values. These models improved the variance explained by 5.0 percent; the model associated with the highest OLR values (275 W/sq m less than or equal to OLR; that is, no cloud) showed only marginal skill. Precipitable water fields were generated from the best TOVS-only model for seven days in January 1983 and compared with SMMR-estimated fields. OLR fields and ECMWF precipitable water analysis. The TOVS regression model compared favorably to the SMMR analysis in amplitudes and features. It revealed more evolving synoptic signal than the ECMWF analysis. In synoptically active regions, it differed with respect to the OLR analysis, primarily because of actual differences in the vertical distribution of water vapor.

Chung, Hyosang↗
NASA NTRS1998

Advanced Water Vapor Lidar Detection System

In the present water vapor lidar system, the detected signal is sent over long cables to a waveform digitizer in a CAMAC crate. This has the disadvantage of transmitting analog signals for a relatively long distance, which is subjected to pickup noise, leading to a decrease in the signal to noise ratio. Generally, errors in the measurement of water vapor with the DIAL method arise from both random and systematic sources. Systematic errors in DIAL measurements are caused by both atmospheric and instrumentation effects. The selection of the on-line alexandrite laser with a narrow linewidth, suitable intensity and high spectral purity, and its operation at the center of the water vapor lines, ensures minimum influence in the DIAL measurement that are caused by the laser spectral distribution and avoid system overloads. Random errors are caused by noise in the detected signal. Variability of the photon statistics in the lidar return signal, noise resulting from detector dark current, and noise in the background signal are the main sources of random error. This type of error can be minimized by maximizing the signal to noise ratio. The increase in the signal to noise ratio can be achieved by several ways. One way is to increase the laser pulse energy, by increasing its amplitude or the pulse repetition rate. Another way, is to use a detector system with higher quantum efficiency and lower noise, on the other hand, the selection of a narrow band optical filter that rejects most of the day background light and retains high optical efficiency is an important issue. Following acquisition of the lidar data, we minimize random errors in the DIAL measurement by averaging the data, but this will result in the reduction of the vertical and horizontal resolutions. Thus, a trade off is necessary to achieve a balance between the spatial resolution and the measurement precision. Therefore, the main goal of this research effort is to increase the signal to noise ratio by a factor of 10 over the current system, using a newly evaluated, very low noise avalanche photo diode detector and constructing a 10 MHz waveform digitizer which will replace the current CAMAC system.

Elsayed-Ali, Hani↗
NASA NTRS2013

The Role of GRAIL Orbit Determination in Preprocessing of Gravity Science Measurements

The Gravity Recovery And Interior Laboratory (GRAIL) mission has constructed a lunar gravity field with unprecedented uniform accuracy on the farside and nearside of the Moon. GRAIL lunar gravity field determination begins with preprocessing of the gravity science measurements by applying corrections for time tag error, general relativity, measurement noise and biases. Gravity field determination requires the generation of spacecraft ephemerides of an accuracy not attainable with the pre-GRAIL lunar gravity fields. Therefore, a bootstrapping strategy was developed, iterating between science data preprocessing and lunar gravity field estimation in order to construct sufficiently accurate orbit ephemerides.This paper describes the GRAIL measurements, their dependence on the spacecraft ephemerides and the role of orbit determination in the bootstrapping strategy. Simulation results will be presented that validate the bootstrapping strategy followed by bootstrapping results for flight data, which have led to the latest GRAIL lunar gravity fields.

science preprocessing↗
NASA NTRS1973

On the error probability of general tree and trellis codes with applications to sequential decoding

An upper bound on the average error probability for maximum-likelihood decoding of the ensemble of random binary tree codes is derived and shown to be independent of the length of the tree. An upper bound on the average error probability for maximum-likelihood decoding of the ensemble of random L-branch binary trellis codes of rate R = 1/n is derived which separates the effects of the tail length T and the memory length M of the code. It is shown that the bound is independent of the length L of the information sequence. This implication is investigated by computer simulations of sequential decoding utilizing the stack algorithm. These simulations confirm the implication and further suggest an empirical formula for the true undetected decoding error probability with sequential decoding.

Johannesson, R.↗
NASA NTRS1977

On the error probability of general trellis codes with applications to sequential decoding

An upper bound on the average probability of error for maximum-likelihood decoding of the ensemble of random L-branch binary trellis codes of rate R = 1/n with distinction between memory length and tail length is given. It is shown that the bound is independent of the length L of the information sequence if the memory length exceeds the tail length by a specified amount that depends on L. Sequential decoding simulations using the stack algorithm were conducted to test the dependence of the undetected error probability on tail length and memory length, and the results corroborated the theory.

Johannesson, R.↗
NASA NTRS1993

Analysis of open-loop conical scan pointing error and variance estimators

General pointing error and variance estimators for an open-loop conical scan (conscan) system are derived and analyzed. The conscan algorithm is modeled as a weighted least-squares estimator whose inputs are samples of receiver carrier power and its associated measurement uncertainty. When the assumptions of constant measurement noise and zero pointing error estimation are applied, the variance equation is then strictly a function of the carrier power to uncertainty ratio and the operator selectable radius and period input to the algorithm. The performance equation is applied to a 34-m mirror-based beam-waveguide conscan system interfaced with the Block V Receiver Subsystem tracking a Ka-band (32-GHz) downlink. It is shown that for a carrier-to-noise power ratio greater than or equal to 30 dB-Hz, the conscan period for Ka-band operation may be chosen well below the current DSN minimum of 32 sec. The analysis presented forms the basis of future conscan work in both research and development as well as for the upcoming DSN antenna controller upgrade for the new DSS-24 34-m beam-waveguide antenna.

Alvarez, L. S.↗
NASA NTRS1996

Computerized Design and Generation of Gear Drives With a Localized Bearing Contact and a Low Level of Transmission Errors

A general approach developed for the computerized simulation of loaded gear drives is presented. In this paper the methodology used to localize the bearing contact, provide a parabolic function of transmission errors, and simulate meshing and contact of unloaded gear drives is developed. The approach developed is applied to spur and helical gears, spiral bevel gears, face-gear drives, and worm-gear drives with cylindrical worms.

Litvin, F.↗
NASA NTRS1989

Dynamical predictability in a simple general circulation model - Average error growth

Average predictability and error growth in a simple realistic two-level general circulation model (GCM) were investigated using a series of Monte Carlo experiments for fixed external forcing (perpetual winter in the Northern Hemisphere). It was found that, for realistic initial errors, the dependence of the limit of dynamic predictability on total wavenumber was similar to that found for the ECMWF model for the 1980/1981 winter conditions, with the lowest wavenumbers showing significant skill for forecast ranges of more than 1 month. On the other hand, for very small amplitude errors distributed according to the climate spectrum, the total error growth was superexponential, reaching a maximum growth rate (2-day doubling time) in about 1 week. A simple empirical model of error variance, which involved two broad wavenumber bands and incorporating a 3/2 power saturation term, was found to provide an excellent fit to the GCM error growth behavior.

Schubert, Siegfried D.↗
NASA NTRSDate not supplied

Avoiding Selection Bias in Generating Examples of Plans in the Presence of Heuristic Error

It is generally understood that heuristic error hurts the performance of search algorithms, measured in terms of search effort. Hence there is an interest in understanding how to reduce heuristic error. One way to do this is to learn a heuristic from a set of examples of plans generated offline, e.g. bootstrapping methods. In this paper, we consider how some methods for generating examples of plans may skew the training set in the presence of heuristic errors. Initial theoretical results show that duplicate detection is one source of selection bias in the canonical A* algorithm. We introduce a duplicate selection scheme for A* that avoids selection bias in generating cost-optimal examples, without compromising memory efficiency, and develop ideas in the satisficing setting. We evaluate our approach on n x m grids with multiple cost-optimal solutions and synthetic heuristic error. Finally, we attempt to extend these ideas to the problem of generating extreme examples of plans.

Alison S Paredes↗
NASA NTRS1980

Electron Capture into Rydberg States According to the Born Approximation: Correction of an Error

A general expression is derived for the Born amplitude for electron capture into high principal, and arbitrary angular momentum, quantum numbers, n and l, by bare nuclei from hydrogenic atom targets. Cross sections for capture into high n and l = 0,1 are given in analytic forms. The limiting values of the ratios of the Born over OBK cross sections for n >> 1 and l = 0,1 are found, and shown that they confirm the empirical law of Jackson and Schiff only approximately. For high angular momenta, it is shown that the Born cross section approaches the OBK cross section. An error in a previous publication by the author has been corrected.

Omidvar, Kazem↗
NASA NTRS2014

Combined Uncertainty and A-Posteriori Error Bound Estimates for General CFD Calculations: Theory and Software Implementation

This workshop presentation discusses the design and implementation of numerical methods for the quantification of statistical uncertainty, including a-posteriori error bounds, for output quantities computed using CFD methods. Hydrodynamic realizations often contain numerical error arising from finite-dimensional approximation (e.g. numerical methods using grids, basis functions, particles) and statistical uncertainty arising from incomplete information and/or statistical characterization of model parameters and random fields. The first task at hand is to derive formal error bounds for statistics given realizations containing finite-dimensional numerical error [1]. The error in computed output statistics contains contributions from both realization error and the error resulting from the calculation of statistics integrals using a numerical method. A second task is to devise computable a-posteriori error bounds by numerically approximating all terms arising in the error bound estimates. For the same reason that CFD calculations including error bounds but omitting uncertainty modeling are only of limited value, CFD calculations including uncertainty modeling but omitting error bounds are only of limited value. To gain maximum value from CFD calculations, a general software package for uncertainty quantification with quantified error bounds has been developed at NASA. The package provides implementations for a suite of numerical methods used in uncertainty quantification: Dense tensorization basis methods [3] and a subscale recovery variant [1] for non-smooth data, Sparse tensorization methods[2] utilizing node-nested hierarchies, Sampling methods[4] for high-dimensional random variable spaces.

CFD↗
NASA NTRS1992

Formulation and error analysis for a generalized image point correspondence algorithm

A Generalized Image Point Correspondence (GIPC) algorithm, which enables the determination of 3-D motion parameters of an object in a configuration where both the object and the camera are moving, is discussed. A detailed error analysis of this algorithm has been carried out. Furthermore, the algorithm was tested on both simulated and video-acquired data, and its accuracy was determined.

Shapiro, Linda↗