Search NASA⌕ Search

SEARCH · Search NASA

Results for “mean squared 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 73 records · Page 4
All sourcesNASA NTRSDOE OSTI
NASA NTRS2010

Maximum Likelihood Time-of-Arrival Estimation of Optical Pulses via Photon-Counting Photodetectors

Many optical imaging, ranging, and communications systems rely on the estimation of the arrival time of an optical pulse. Recently, such systems have been increasingly employing photon-counting photodetector technology, which changes the statistics of the observed photocurrent. This requires time-of-arrival estimators to be developed and their performances characterized. The statistics of the output of an ideal photodetector, which are well modeled as a Poisson point process, were considered. An analytical model was developed for the mean-square error of the maximum likelihood (ML) estimator, demonstrating two phenomena that cause deviations from the minimum achievable error at low signal power. An approximation was derived to the threshold at which the ML estimator essentially fails to provide better than a random guess of the pulse arrival time. Comparing the analytic model performance predictions to those obtained via simulations, it was verified that the model accurately predicts the ML performance over all regimes considered. There is little prior art that attempts to understand the fundamental limitations to time-of-arrival estimation from Poisson statistics. This work establishes both a simple mathematical description of the error behavior, and the associated physical processes that yield this behavior. Previous work on mean-square error characterization for ML estimators has predominantly focused on additive Gaussian noise. This work demonstrates that the discrete nature of the Poisson noise process leads to a distinctly different error behavior.

Erkmen, Baris I.↗
NASA NTRS1973

An empirical Bayes approach for the Poisson life distribution.

A smooth empirical Bayes estimator is derived for the intensity parameter (hazard rate) in the Poisson distribution as used in life testing. The reliability function is also estimated either by using the empirical Bayes estimate of the parameter, or by obtaining the expectation of the reliability function. The behavior of the empirical Bayes procedure is studied through Monte Carlo simulation in which estimates of mean-squared errors of the empirical Bayes estimators are compared with those of conventional estimators such as minimum variance unbiased or maximum likelihood. Results indicate a significant reduction in mean-squared error of the empirical Bayes estimators over the conventional variety.

Canavos, G. C.↗
NASA NTRS1992

Convergence characteristics of the multiple input, multiple output LMS algorithm

The convergence characteristics of the multiple input, multiple output LMS algorithm, as applied to active noise and vibration control systems, are examined. The mean square error during the convergence process, as well as the final converged value, are examined analytically and in computer simulation. It is shown that the ratio of number of error sensors to number of control sources has a significant influence upon both the converging and converged value of the mean square error. Other active control system variables, such as the inherent time delays and structural/acoustic transfer functions, are also shown to have a significant influence upon the convergence process.

Snyder, Scott D.↗
NASA NTRS2019

Estimation of coarse dead wood stocks in intact and degraded forests in the Brazilian Amazon using airborne lidar

Coarse dead wood is an important component of forest carbon stocks, but it is rarely measured in Amazon forests and is typically excluded from regional forest carbon budgets. Our study is based on line intercept sampling for fallen coarse dead wood conducted along 103 transects with a total length of 48 km matched with forest inventory plots where standing coarse dead wood was measured in the footprints of larger areas of airborne lidar acquisitions. We developed models to relate lidar metrics and Landsat time series variables to coarse dead wood stocks for intact, logged, burned, or logged and burned forests. Canopy characteristics such as gap area produced significant individual relations for logged forests. For total fallen plus standing coarse dead wood (hereafter defined as total coarse dead wood), the relative root mean square error for models with only lidar metrics ranged from 33 % in logged forest to up to 36 % in burned forests. The addition of historical information improved model performance slightly for intact forests (31 % against 35 % relative root mean square error), not justifying the use of a number of disturbance events from historical satellite images (Landsat) with airborne lidar data. Lidar-derived estimates of total coarse dead wood compared favorably with independent ground-based sampling for areas up to several hundred hectares. The relations found between total coarse dead wood and variables quantifying forest structure derived from airborne lidar highlight the opportunity to quantify this important but rarely measured component of forest carbon over large areas in tropical forests.

Brazilian Amazon↗
NASA NTRS1974

Evaluation of communication systems using both time and frequency domain techniques

The present work investigates the measurement of the mean-square error of linear systems and the power in a random waveform using digital simulation. A postprocessor is described which yields all necessary information for determining input signal power, output power, average system gain, average system time delay, and system mean-square error. It is shown that a minimum of 10,000 data points must be processed if 4% accuracy with 98% confidence is required for the determination of the power in a random waveform.

Tranter, W. H.↗
NASA NTRS2014

Using a Support Vector Machine and a Land Surface Model to Estimate Large-Scale Passive Microwave Temperatures over Snow-Covered Land in North America

A support vector machine (SVM), a machine learning technique developed from statistical learning theory, is employed for the purpose of estimating passive microwave (PMW) brightness temperatures over snow-covered land in North America as observed by the Advanced Microwave Scanning Radiometer (AMSR-E) satellite sensor. The capability of the trained SVM is compared relative to the artificial neural network (ANN) estimates originally presented in [14]. The results suggest the SVM outperforms the ANN at 10.65 GHz, 18.7 GHz, and 36.5 GHz for both vertically and horizontally-polarized PMW radiation. When compared against daily AMSR-E measurements not used during the training procedure and subsequently averaged across the North American domain over the 9-year study period, the root mean squared error in the SVM output is 8 K or less while the anomaly correlation coefficient is 0.7 or greater. When compared relative to the results from the ANN at any of the six frequency and polarization combinations tested, the root mean squared error was reduced by more than 18 percent while the anomaly correlation coefficient was increased by more than 52 percent. Further, the temporal and spatial variability in the modeled brightness temperatures via the SVM more closely agrees with that found in the original AMSR-E measurements. These findings suggest the SVM is a superior alternative to the ANN for eventual use as a measurement operator within a data assimilation framework.

Modeling↗
NASA NTRS1991

Optimum data analysis procedures for Titan 4 and Space Shuttle payload acoustic measurements during lift-off

Analytical expressions have been derived to describe the mean square error in the estimation of the maximum rms value computed from a step-wise (or running) time average of a nonstationary random signal. These analytical expressions have been applied to the problem of selecting the optimum averaging times that will minimize the total mean square errors in estimates of the maximum sound pressure levels measured inside the Titan IV payload fairing (PLF) and the Space Shuttle payload bay (PLB) during lift-off. Based on evaluations of typical Titan IV and Space Shuttle launch data, it has been determined that the optimum averaging times for computing the maximum levels are (1) T (sub o) = 1.14 sec for the maximum overall level, and T(sub oi) = 4.88 f (sub i) (exp -0.2) sec for the maximum 1/3 octave band levels inside the Titan IV PLF, and (2) T (sub o) = 1.65 sec for the maximum overall level, and T (sub oi) = 7.10 f (sub i) (exp -0.2) sec for the maximum 1/3 octave band levels inside the Space Shuttle PLB, where f (sub i) is the 1/3 octave band center frequency. However, the results for both vehicles indicate that the total rms error in the maximum level estimates will be within 25 percent the minimum error for all averaging times within plus or minus 50 percent of the optimum averaging time, so a precise selection of the exact optimum averaging time is not critical. Based on these results, linear averaging times (T) are recommended for computing the maximum sound pressure level during lift-off.

Piersol, Allan G.↗
NASA NTRS1972

Reciprocal - MSE and SNR as distinct performance measures in below-threshold communication.

The possible advantages of using algorithms that are minimum mean-square-error (MSE) optimal, or other averaging-type demodulators with frequency position modulation (FPM), are investigated. An attempt is made to describe the performance of the optimal and various suboptimal demodulators in terms of three parameters which depend on SNR and bandwidth expansion. The performance measures investigated are the reciprocal of the normalized mean-square error and the output SNR, which are different in the below-threshold region. The former is compared to the rate-distortion bound. The possible advantage of averaging-type discriminators from the point of view of the threshold is discussed.

Bar-David, I.↗
NASA NTRS1973

A successive overrelaxation iterative technique for an adaptive equalizer

An adaptive strategy for the equalization of pulse-amplitude-modulated signals in the presence of intersymbol interference and additive noise is reported. The successive overrelaxation iterative technique is used as the algorithm for the iterative adjustment of the equalizer coefficents during a training period for the minimization of the mean square error. With 2-cyclic and nonnegative Jacobi matrices substantial improvement is demonstrated in the rate of convergence over the commonly used gradient techniques. The Jacobi theorems are also extended to nonpositive Jacobi matrices. Numerical examples strongly indicate that the improvements obtained for the special cases are possible for general channel characteristics. The technique is analytically demonstrated to decrease the mean square error at each iteration for a large range of parameter values for light or moderate intersymbol interference and for small intervals for general channels. Analytically, convergence of the relaxation algorithm was proven in a noisy environment and the coefficient variance was demonstrated to be bounded.

Kosovych, O. S.↗
DOE OSTI2025

Baseflow Identification via Explainable AI With Kolmogorov‐Arnold Networks

Abstract Hydrological models often involve constitutive laws that may not be optimal in every application. We propose to replace such laws with the Kolmogorov‐Arnold networks (KANs), a class of neural networks designed to identify symbolic expressions. We demonstrate KAN's potential on the problem of baseflow identification, a notoriously challenging task plagued by significant uncertainty. KAN‐derived functional dependencies of the baseflow components on the aridity index outperform their original counterparts; they demonstrate that water availability, rather than potential evapotranspiration, drives baseflow by constraining actual evapotranspiration under arid conditions. On a test set, they increase the Nash‐Sutcliffe efficiency (NSE) by 65%, decrease the root mean squared error by 29%, and increase the Kling‐Gupta efficiency by 34%. This superior performance is achieved while reducing the number of fitting parameters from three to two. Next, we use data from 378 catchments across the continental United States to refine the water‐balance equation at the mean‐annual scale. The KAN‐derived equations based on the refined water balance outperform both the current aridity index model, with up to a 105% increase in NSE, and the KAN‐derived equations based on the original water balance. While the performance of our model and tree‐based machine learning methods is similar, KANs offer the advantage of simplicity and transparency and require no specific software or computational tools. This case study focuses on the aridity index formulation, but the approach is flexible and transferable to other hydrological processes. Plain Language Summary Equations used in hydrologic model are often suboptimal, resulting in reduced prediction accuracy and efficiency. We implemented Kolmogorov‐Arnold networks (KAN), a machine learning algorithm for deriving symbolic formulations, to estimate groundwater recharge and showed that it outperforms an existing state‐of‐the‐art semi‐empirical formulation. In hydrology, Nash‐Sutcliffe efficiency (NSE), root mean squared error (RMSE), and Kling‐Gupta efficiency (KGE) are commonly used to evaluate model performance. Higher NSE and KGE values indicate better performance, while lower RMSE values are preferable. Our results show that NSE increased by 71%, RMSE decreased by 32%, and KGE improved by 25%. In addition, KAN identifies an optimal functional form and can be used to derive new analytical formulas using the prior knowledge. The KAN‐inspired equation outperformed the original formulation and reduced the fitting parameters. Furthermore, we refined the water‐balance equation at the mean‐annual scale and showed that, based on the new water‐balance equation, KAN can derive new formulations that are superior to the original aridity index formulations (up to 105% increase in NSE) and KAN‐derived equations based on the original water balance. These findings highlight the significant potential of KAN to advance the scientific understanding of a wide range of hydrologic processes. Key Points Kolmogorov‐Arnold networks (KANs) enhance interpretability of machine‐learned hydrological models KAN‐derived symbolic formulations outperform state‐of‐the‐art semi‐empirical aridity indices KAN‐identified functional form yields an analytical index with fewer fitting parameters and improved performance

baseflow↗
NASA NTRS1987

Order reduction in linear state estimation under performance constraints

The design and analysis of minimal-order state estimators for possibly time-varying linear systems, under constraints on the maximal allowable mean-square error, are considered. A global lower bound on the optimal error is derived, along with a lower bound on the minimal estimator order, needed for meeting the performance constraint. The ideal reduced-order estimator which satisfies the lower bound is derived, along with conditions for its realizability. When the ideal estimator is not realizable, its structure forms a suboptimal estimator, which maintains, in some sense, a local optimality property and is called the pseudoideal estimator. The mean-square error of the pseudoideal estimator defines upper bounds on the optimal error and on the estimator order needed for meeting the performance constraint. The lower and the upper bounds on the order define a reduced search set for the design problem. When the distance between the ideal and the pseudoideal estimators is sufficiently small in a certain numerical sense, the pseudoideal estimator may be considered optimal for practical purposes.

Baram, Yoram↗
NASA NTRS1997

Root-Raised Cosine Filter Implementation That Uses Canonical Signed Digits for High-Speed Digital Filter Applications

NASA Lewis Research Center's Space Communications Division has been investigating high-speed digital filters that can operate at a higher speed than those in current use for a digital modulator and demodulator (modem). Using the Canonical Signed Digits (CSD) number representation for filter coefficients is a very effective way to increase the filter's speed while reducing complexity in the digital filter hardware design. This approach is a good alternative to using an expensive parallel-processing design technique or custom, application-specific integrated circuits. Such integrated circuits may not be suitable for applications that require filter speeds faster than what application-specific integrated circuits digital signal processors can offer for a dedicated channel. When a communication channel is a dedicated, multiplication process--a costly, time-consuming process--it can be greatly simplified by a replacement of the filter coefficients with CSD numbers. A computer code written with the MATLAB software package runs the program and generates CSD-represented filter coefficients that are based on minimizing minimum mean square errors. Also, the Alta Group of Cadence's Signal Processing Workstation is used to simulate and analyze the CSD filter responses. The impulse response of the root-raised cosine filter that is used as a base model is defined. From this filter, a set of coefficients is sampled and stored in a file. For the all coefficients, the optimal CSD number for each coefficient is searched on the basis of the minimum-mean-square-errors criterion. Because the distribution of CSD numbers is not uniform, quantization errors tend to be bigger for coefficients greater than 1/2. To offset errors that occur in a region of coefficients between 1/2 to 1 and to better represent fractions with CSD numbers, an extra nonzero digit is allowed for any coefficients exceeding 1/2. This will greatly improve frequency response as well as intersymbol interference at the receiver. The frequency response of a set of collected CSD-represented filter coefficients was compared with the same filter that was conventionally implemented. Analyses show CSD-implemented filters perform as well as conventional filters. Comparison of eye diagrams and bit-error-rate curves between CSD filters and traditionally implemented filters are almost indistinguishable. However, filter complexity was reduced from almost 3.5 to 1 for CSD filters. Complete computer simulation results are available. In the near future, work will focus on building actual working digital filter hardware in a field programmable gate array (FPGA).

Kim, Heechul↗
DOE OSTI2024

Leveraging design of experiments to build chemometric models for the quantification of uranium (VI) and HNO3 by Raman spectroscopy

Partial least squares regression (PLSR) and support vector regression (SVR) models were optimized for the quantification of U(VI) (10–320 g L −1 ) and HNO 3 (0.6–6 M) by Raman spectroscopy with optimized calibration sets chosen by optimal design of experiments. The designed approach effectively minimized the number of samples in the calibration set for PLSR and SVR by selecting sample concentrations with a quadratic process model, despite complex confounding and covarying spectral features in the spectra. The top PLS2 model resulted in percent root mean square errors of prediction for U(VI), HNO 3 , and NO 3 − of 3.7%, 3.6%, and 2.9%, respectively. PLS1 models performed similarly despite modeling an analyte with a majority linear response (i.e., uranyl symmetric stretch) and another with more covarying vibrational modes (i.e., HNO 3 ). Partial least squares (PLS) model loadings and regression coefficients were evaluated to better understand the relationship between weaker Raman bands and covarying spectral features. Support vector machine models outperformed PLS1 models, resulting in percent root mean square error of prediction values for U(VI) and HNO 3 of 1.5% and 3.1%, respectively. The optimal nonlinear SVR model was trained using a similar number of samples (11) compared with the PLSR model, even though PLS is a linear modeling approach. The generic D-optimal design presented in this work provides a robust statistical framework for selecting training set samples in disparate two-factor systems. This approach reinforces Raman spectroscopy for the quantification of species relevant to the nuclear fuel cycle and provides a robust chemometric modeling approach to bolster online monitoring in challenging process environments.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

Analyzing and Exploring Training Recipes for Large-Scale Transformer-Based Weather Prediction

Abstract The rapid rise of deep learning (DL) in numerical weather prediction (NWP) has led to a proliferation of models which forecast atmospheric variables with comparable or superior skill than traditional physics-based NWP. However, among these leading DL models, there is a wide variance in both the training settings and architecture used. Further, the lack of thorough ablation studies makes it hard to discern which components are most critical to success. In this work, we show that it is possible to attain high forecast skill even with relatively off-the-shelf architectures, simple training procedures, and moderate compute budgets. Specifically, we train a minimally modified Swin Transformer V2 (SwinV2) on ERA5 data and find that it attains superior skill in terms of mean-square errors of deterministic forecasts when compared against the European Centre for Medium-Range Weather Forecasts’ Integrated Forecasting System (IFS). Almost all DL–NWP systems share a core set of hyperparameters and design decisions. To aid and expedite future DL–NWP research, we present an in-depth, systematic exploration of different loss functions, model sizes and depths, patch sizes, and multistep training objectives. We also examine the model performance with metrics beyond the typical accuracy (ACC) and RMSE and investigate how the performance scales with model size. Through our open-source code, scoring pipelines, and models, we share our findings on key aspects of the training pipeline. These ablations reduce the necessity for expensive hyperparameter tuning and lower the barrier to entry for future DL–NWP research. Significance Statement This study investigates the potential of using large-scale transformer-based models for weather prediction, showing that it is possible to achieve high forecast accuracy with simpler, off-the-shelf architectures. By training a minimally modified SwinV2 transformer on ERA5 data, we show that the model achieves competitive forecast skill in terms of mean-square error for key variables, outperforming the European Centre for Medium-Range Weather Forecasts’ Integrated Forecasting System (IFS) at all lead times. Our findings suggest that effective training strategies, such as multistep fine-tuning and channel-weighted losses, significantly enhance the model’s performance. However, we also highlight that these improvements come with trade-offs in other areas, such as ensemble spread and high-frequency spatial detail. This work highlights the promise of deep learning in improving weather forecasts, which could lead to better preparedness and response to weather events, ultimately benefiting society by providing more reliable weather predictions.

Willard, Jared D. [Lawrence Berkeley National Labo↗
NASA NTRS2015

Real-Time Kennedy Space Center and Cape Canaveral Air Force Station High-Resolution Model Implementation and Verification

Customer: NASA's Launch Services Program (LSP), Ground Systems Development and Operations (GSDO), and Space Launch System (SLS) programs. NASA's LSP, GSDO, SLS and other programs at Kennedy Space Center (KSC) and Cape Canaveral Air Force Station (CCAFS) use the daily and weekly weather forecasts issued by the 45th Weather Squadron (45 WS) as decision tools for their day-to-day and launch operations on the Eastern Range (ER). For example, to determine if they need to limit activities such as vehicle transport to the launch pad, protect people, structures or exposed launch vehicles given a threat of severe weather, or reschedule other critical operations. The 45 WS uses numerical weather prediction models as a guide for these weather forecasts, particularly the Air Force Weather Agency (AFWA) 1.67 kilometer Weather Research and Forecasting (WRF) model. Considering the 45 WS forecasters' and Launch Weather Officers' (LWO) extensive use of the AFWA model, the 45 WS proposed a task at the September 2013 Applied Meteorology Unit (AMU) Tasking Meeting requesting the AMU verify this model. Due to the lack of archived model data available from AFWA, verification is not yet possible. Instead, the AMU proposed to implement and verify the performance of an ER version of the AMU high-resolution WRF Environmental Modeling System (EMS) model (Watson 2013) in real-time. The tasking group agreed to this proposal; therefore the AMU implemented the WRF-EMS model on the second of two NASA AMU modeling clusters. The model was set up with a triple-nested grid configuration over KSC/CCAFS based on previous AMU work (Watson 2013). The outer domain (D01) has 12-kilometer grid spacing, the middle domain (D02) has 4-kilometer grid spacing, and the inner domain (D03) has 1.33-kilometer grid spacing. The model runs a 12-hour forecast every hour, D01 and D02 domain outputs are available once an hour and D03 is every 15 minutes during the forecast period. The AMU assessed the WRF-EMS 1.33-kilometer domain model performance for the 2014 warm season (May-September). Verification statistics were computed using the Model Evaluation Tools, which compared the model forecasts to observations. The mean error values were close to 0 and the root mean square error values were less than 1.8 for mean sea-level pressure (millibars), temperature (degrees Kelvin), dewpoint temperature (degrees Kelvin), and wind speed (per millisecond), all very small differences between the forecast and observations considering the normal magnitudes of the parameters. The precipitation forecast verification results showed consistent under-forecasting of the precipitation object size. This could be an artifact of calculating the statistics for each hour rather than for the entire 12-hour period. The AMU will continue to generate verification statistics for the 1.33-kilometer WRF-EMS domain as data become available in future cool and warm seasons. More data will produce more robust statistics and reveal a more accurate assessment of model performance. Once the formal task was complete, the AMU conducted additional work to better understand the wind direction results. The results were stratified diurnally and by wind speed to determine what effects the stratifications would have on the model wind direction verification statistics. The results are summarized in the addendum at the end of this report. In addition to verifying the model's performance, the AMU also made the output available in the Advanced Weather Interactive Processing System II (AWIPS II). This allows the 45 WS and AMU staff to customize the model output display on the AMU and Range Weather Operations AWIPS II client computers and conduct real-time subjective analyses. In the future, the AMU will implement an updated version of the WRF-EMS model that incorporates local data assimilation. This model will also run in real-time and be made available in AWIPS II.

Numerical Weather Prediction↗
DOE OSTI2025

Estimating evapotranspiration from deciduous broadleaf forests by considering the effect of foliar litter

Forest litter cover significantly alters the ground surface resistance, which influences the water and energy transfer processes between the land surface and atmosphere. Penman-Monteith (PM) model is widely used to estimate land surface evapotranspiration (ET). However, it was shown to overestimate ET during the period when foliar litter covers the ground surface. Therefore, incorporating the effect of foliar litter on ground surface resistance can potentially improve ET estimates. Here, in this study, we proposed a foliar litter surface resistance model to describe the effect of litter cover on water vapor transport, then incorporated the litter surface resistance into the PM model (noted as PM-EL model). The performance of the PM-EL model was evaluated using observations from 18 deciduous broadleaf forest (DBF) flux sites across global FLUXNET2015 datasets. Results showed that: (1) both the PM model and the PM-EL model are most sensitive to ground surface resistance among resistance parameters; (2) the proposed litter resistance model is capable of describing the seasonal dynamic of litter in DBF ; (3) the PM-EL model, has been shown to improve ET estimates with the coefficient of determination increased by 8.5 %, the mean absolute error and relative root mean square error decreased by 22.1 % and 20.9 %, respectively. The study contributes to our understanding of the effect of litter cover on ET estimation and provides insights into forest ecohydrology and land surface mass and energy transfer processes.

Deciduous Broadleaf Forest↗