Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Upper regularity”

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 55 records · Page 3
All sourcesNASA NTRSDOE OSTI
NASA NTRS2019

Constraining Stellar Parameters and Atmospheric Dynamics of the Carbon AGB star V Oph

Molecules and dust produced by the atmospheres of cool evolved stars contribute to a significant amount of the total material found in the interstellar medium. To understand the mechanism behind the mass loss of these stars, it is of pivotal importance to investigate the structure and dynamics of their atmospheres. Our goal is to verify if the extended molecular and dust layers of the carbon-rich asymptotic giant branch (AGB) star V Oph, and their time variations, can be explained by dust-driven winds triggered by stellar pulsation alone, or if other mechanisms are in play. We model V Oph mid-infrared interferometric VLTI-MIDI data (8–13 μm), at phases 0.18, 0.49, and 0.65, together with literature photometric data, using the latest-generation self-consistent dynamic atmosphere models for carbon-rich stars: DARWIN. We determine the fundamental stellar parameters: T(eff) = 2600 K, L(bol) = 3585 L⊙, M = 1.5 M⊙, C/O = 1.35, Mass flow rate= 2.50 x 10(exp -6) M⊙/yr. We calculate the stellar photospheric radii at the three phases: 479, 494, 448 R⊙; and the dust radii: 780, 853, 787 R⊙. The dynamic models can fairly explain the observed N-band visibility and spectra, although there is some discrepancy between the data and the models, which is discussed in the text. We discuss the possible causes of the temporal variations of the outer atmosphere, deriving an estimate of the magnetic field strength, and computing upper limits for the Alfvén waves velocity. In addition, using period–luminosity sequences, and interferometric modeling, we suggest V Oph as a candidate to be reclassified as a semi-regular star.

Gioia Rau↗
DOE OSTI2022

Hydrokinetic energy conversion using flow induced oscillations of single-cylinder with large passive turbulence control

Various types of flow-induced oscillations (FIOs) have been implemented in development of marine hydrokinetic (MHK) energy converters. With passive turbulence control (PTC), energy harvesting starts at a flow speed of about 0.5 m/s. However, there is worldwide MHK energy available in even slower currents. In the present study, the effect of damping on FIO and power extraction is investigated for a converter with large turbulence stimulation (PTC) consisting of straight strips with a height of 15% of the cylinder diameter and placed symmetrically on the cylinder surface. The oscillating amplitude decreases, as the damping ratio increases, with unchanged sinusoidal pattern of the displacement time-history. The frequency ratio is also affected by damping especially in the VIV initial branch and transition region between VIV and galloping. An important flow characteristic of the large-PTC cylinder is that a recirculation region is formed behind the PTC, causing appreciable disturbance to the flow past the cylinder. Power can be harvested in the whole FIO range and the harnessed power maximum appears at the largest inflow velocity tested. However, the optimum of harnessing efficiency is located at the beginning of the VIV upper branch. The gap between VIV and galloping is bridged when large PTC is used, eliminating the drop in power and efficiency even at higher damping, which would be a weakness of regular-PTC cylinder for energy harvesting. Finally, the mechanism behind the variation of harnessing efficiency with inflow velocity and damping ratio is revealed, and the optimality criterion for the converter design is discussed.

16 TIDAL AND WAVE POWER↗
NASA NTRS1989

Considerations for an exercise prescription

A number of past and most recent research findings that describe some of the physiological responses to exercise in man and their relationship with exposure to various gravitational environments are discussed. Most of the data pertain to adaptations of the cardiovascular and body fluid systems. It should be kept in mind that the data from studies on microgravity simulation in man include exposures of relatively short duration (5 hours to 14 days). However, it is argued that the results may provide important guidelines for the consideration of many variables which are pertinent to the development of exercise prescription for long-duration space flight. The following considerations for exercise prescriptions during long-duration space flight are noted: (1) Relatively high aerobic fitness and strength, especially of the upper body musculature, should be a criterion for selection of astronauts who will be involved in EVA, since endurance and strength appear to be predominant characteristics for work performance. (2) Some degree of upper body strength will probably be required for effective performance of EVA. However, the endurance and strength required by the upper body for EVA can probably be obtained through preflight exercise prescription which involves swimming. (3) Although some degree of arm exercise may be required to maintain preflight endurance and strength, researchers propose that regular EVA will probably be sufficient to maintain the endurance and strength required to effectively perform work tasks during space flight. (4) A minimum of one maximal aerobic exercise every 7 to 10 days during space flight may be all that is necessary for maintenance of normal cardiovascular responsiveness and replacement of body fluids for reentry following prolonged space flight. (5) The possible reduction in the amount of exercise required for maintenance of cardiovascular system and body fluids in combination with the use of electromyostimulation (EMS) or methods other than conventional exercise for maintaining size and strength of muscles and bones needs great consideration for further research. These approaches represent a potential solution to the problem of compromising valuable time for exercise that is needed for daily operations.

Convertino, Victor A.↗
NASA NTRS2003

EAGLE can do Efficient LTL Monitoring

We briefly present a rule-based framework, called EAGLE, that has been shown to be capable of defining and implementing finite trace monitoring logics, including future and past time temporal logic, extended regular expressions, real-time logics, interval logics, forms of quantified temporal logics, and so on. In this paper we show how EAGLE can do linear temporal logic (LTL) monitoring in an efficient way. We give an upper bound on the space and time complexity of this monitoring.

Barringer, Howard↗
NASA NTRS2008

Delay-Throughput Performance of the Deep-Space Ka-band Link

In this paper, performance of a first-in, first-out (FIFO), selective retransmission scheme for the deep-space Ka-band link is presented and compared to the performance of a comparable X-band link. In this analysis, 16 months of water vapor radiometer (WVR) and advanced water vapor radiometer (AWVR) data from the three Deep Space Network (DSN) Communication Complexes (DSCC) were used to emulate weather effects on X-band and Ka-band links from Mars. Mars Reconnaissance Orbiter (MRO) X-band and Ka-band telecommunications parameters were used for spacecraft telecommunications capabilities. One pass per week per complex was selected from MRO's Deep Space Network (DSN) schedule from April 1, 2006 to August 31, 2007 for a total of 207 passes (69 passes per complex) for this analysis. For each pass both X-band and Ka-band links were designed using at most two data rates so that the expected pass capacity would be maximized subject to a minimum availability requirement (MAR). In conjunction with the WVR/AWVR data, elevation profiles of the selected passes and models for the performance of the antennas in the DSN were used to emulate the performance of both links. It was assumed that the retransmission of the data takes place not on the same pass as the original transmission but during subsequent passes. The data collected before a pass was assumed to be a fraction of the expected capacity of the pass as calculated through the link design process. Infinite spacecraft storage was assumed to obtain an upper bound on the spacecraft storage requirement. The independent parameters of this analysis were MAR and the ratio of data collected before a pass to the expected pass capacity. Since the selected passes did not occur at regular intervals, the delay in this analysis was measured in terms of number of passes. The throughput was measured in terms of number of bits received successfully on the ground. The results indicate that reasonable delay performance could be achieved with very high throughput for relatively low MAR values for data collection to expected pass capacity ratio of around 97% for Ka-band. The results indicate that, except for very low average delay requirements, the Ka-band link provides more than twice the throughput of the X-band link for the same amount of power consumed by the spacecraft. In addition, the results indicate that the required storage onboard the spacecraft is not prohibitive and good performance could be achieved by using a buffer size less than three times the maximum amount of data collected before a pass.

Shambayati, Shervin↗
NASA NTRS2008

Delay-Throughput Performance the Deep-Space Ka-Band Link

In this paper, performance of a first-in, first-out (FIFO), selective retransmission scheme for the deep-space Ka-band link is presented and compared to the performance of a comparable X-band link. In this analysis, 16 months of water vapor radiometer (WVR) and advanced water vapor radiometer (AWVR) data from the three Deep Space Network (DSN) Communication Complexes (DSCC) were used to emulate weather effects on X-band and Ka-band links from Mars. Mars Reconnaissance Orbiter (MRO) X-band and Ka-band telecommunications parameters were used for spacecraft telecommunications capabilities. One pass per week per complex was selected from MRO's Deep Space Network (DSN) schedule from April 1, 2006 to August 31, 2007 for a total of 207 passes (69 passes per complex) for this analysis. For each pass both X-band and Ka-band links were designed using at most two data rates so that the expected pass capacity would be maximized subject to a minimum availability requirement (MAR). In conjunction with the WVR/AWVR data, elevation profiles of the selected passes and models for the performance of the antennas in the DSN were used to emulate the performance of both links. It was assumed that the retransmission of the data takes place not on the same pass as the original transmission but during subsequent passes. The data collected before a pass was assumed to be a fraction of the expected capacity of the pass as calculated through the link design process. Infinite spacecraft storage was assumed to obtain an upper bound on the spacecraft storage requirement. The independent parameters of this analysis were MAR and the ratio of data collected before a pass to the expected pass capacity. Since the selected passes did not occur at regular intervals, the delay in this analysis was measured in terms of number of passes. The throughput was measured in terms of number of bits received successfully on the ground. The results indicate that reasonable delay performance could be achieved with very high throughput for relatively low MAR values for data collection to expected pass capacity ratio of around 97% for Ka-band. The results indicate that, except for very low average delay requirements, the Ka-band link provides more than twice the throughput of the X-band link for the same amount of power consumed by the spacecraft. In addition, the results indicate that the required storage onboard the spacecraft is not prohibitive and good performance could be achieved by using a buffer size less than three times the maximum amount of data collected before a pass.

Shambayati, Shervin↗
NASA NTRS1992

Moessbauer Spectroscopy for Lunar Resource Assessment: Measurement of Mineralogy and Soil Maturity

First-order assessment of lunar soil as a resource includes measurement of its mineralogy and maturity. Soils in which the mineral ilmenite is present in high concentrations are desirable feedstock for the production of oxygen at a lunar base. The maturity of lunar soils is a measure of their relative residence time in the upper 1 mm of the lunar surface. Increasing maturity implies increasing load of solar wind species (e.g., N, H, and He-3), decreasing mean grain size, and increasing glass content. All these physicochemical properties that vary in a regular way with maturity are important parameters for assessing lunar soil as a resource. For example, He-3 can be extracted and potentially used for nuclear fusion. A commonly used index for lunar soil maturity is I(sub s)/FeO, which is the concentration of fine-grained metal determined by ferromagnetic resonance (I(sub s)) normalized to the total iron content (as FeO). I(sub s)/FeO has been measured for virtually every soil returned by the Apollo and Luna missions to the Moon. Because the technique is sensitive to both oxidation state and mineralogy, iron Moessbauer spectroscopy (FeMS) is a viable technique for in situ lunar resource assessment. Its utility for mineralogy is apparent from examination of published FeMS data for lunar samples. From the data published, it can be inferred that FeMS data can also be used to determine soil maturity. The use of FeMS to determine mineralogy and maturity and progress on development of a FeMS instrument for lunar surface use are discussed.

Morris, R. V.↗
NASA NTRS2013

The Hayabusa Curation Facility at Johnson Space Center

The Japan Aerospace Exploration Agency (JAXA) Hayabusa spacecraft made contact with the asteroid 25143 Itokawa and collected regolith dust from Muses Sea region of smooth terrain [1]. The spacecraft returned to Earth with more than 10,000 grains ranging in size from just over 300 μm to less than 10 μm [2, 3]. These grains represent the only collection of material returned from an asteroid by a spacecraft. As part of the joint agreement between JAXA and NASA for the mission, 10% of the Hayabusa grains are being transferred to NASA for parallel curation and allocation. In order to properly receive process and curate these samples, a new curation facility was established at Johnson Space Center (JSC). Since the Hayabusa samples within the JAXA curation facility have been stored free from exposure to terrestrial atmosphere and contamination [4], one of the goals of the new NASA curation facility was to continue this treatment. An existing lab space at JSC was transformed into a 120 sq.ft. ISO class 4 (equivalent to the original class 10 standard) clean room. Hayabusa samples are stored, observed, processed, and packaged for allocation inside a stainless steel glove box under dry N2. Construction of the clean laboratory was completed in 2012. Currently, 25 Itokawa particles are lodged in NASA's Hayabusa Lab. Special care has been taken during lab construction to remove or contain materials that may contribute contaminant particles in the same size range as the Hayabusa grains. Several witness plates of various materials are installed around the clean lab and within the glove box to permit characterization of local contaminants at regular intervals by SEM and mass spectrometry, and particle counts of the lab environment are frequently acquired. Of particular interest is anodized aluminum, which contains copious sub-mm grains of a multitude of different materials embedded in its upper surface. Unfortunately the use of anodized aluminum was necessary in the construction of the clean room frame to strengthen it and eliminate corrosion and wear over time. All anodized aluminum interior to the lab was thus covered or replaced by minimally contaminating materials.

Zolensky, M.↗
NASA NTRS1992

Small-scale fracture patterns on the volcanic plains of Venus

Linear features with a regular spacing of about 1 km in the gridded plains of Guinevere Planitia observed in Magellan radar images of Venus are examined. Many sets of parallel, regularly spaced lineations similar to the closely spaced features of the gridded plains were found. These sets of parallel lineations, which are interpreted to be fractures, typically cover areas with dimensions of hundreds of kilometers. Several examples of these regular lineations are characterized in terms of their average fracture spacing, which is between 1 and 2.5 km; the scatter in individual spacings is about +/- 1/3 the average. It is hypothesized that these features are extension fractures in the brittle upper layers of the volcanic plains material, which formed well after emplacement and solidification of the flows.

Banerdt, W. B.↗
DOE OSTI2026

CHESS 2025: Orthorectified airborne RGB imagery from NEON AOP surveys

This dataset provides Level 1 (L1) and Level 3 (L3) orthorectified Red-Green-Blue (RGB) imagery collected for the 2025 Colorado Headwaters Ecological Spectroscopy Study (CHESS). This high-resolution imagery is a photographic record of red, green, and blue visible light from sunlight reflected off of the Earth’s surface. The data comprise full-color images of the ground surface and are primarily intended to provide context to imaging spectroscopy and light detection and ranging (LiDAR) data. The data were acquired over three study domains in the Upper Gunnison river basin: the upper East River watershed (CRBU); Almont Triangle and Taylor Canyon (ALMO); and Upper Taylor River watershed (UPTA) between 2025-06-13 and 2025-07-15. RGB images were acquired using the PhaseOne IXM-RS150F high-resolution digital camera onboard the National Ecological Observatory Network (NEON) Airborne Observation Platform (AOP). The package data include both an L1 product comprising one camera frame per file and an L3 mosaic aligned to the Universal Transverse Mercator (UTM) Zone 13N grid and the World Geodetic System (WGS) 84 projection. Both products are provided in geotif (.tif) format at 0.1 m ground resolution. The bulk of the imagery was collected during the main CHESS field campaign from June 13 to July 15, 2025. Additional images of a portion of the Upper Taylor (UPTA) domain were collected on September 18, 2025, to fill gaps in imagery identified after the main campaign was complete. RGB camera imagery is not radiometrically calibrated, and therefore pixel values should not be exploited for scientific analysis. Pixel values have undergone a manual adjustment to enhance feature identification. The imagery is rigorously geolocated which does allow for reliable geometric information to be retrieved. To generate the orthorectified imagery, the NEON AOP camera captured visible spectrum in red, green, and blue bands. The raw images were then processed using NEON’s camera orthorectification workflow. A boresight calibration flight was made to build a complete camera, distortion, and alignment model. Color balance/white balance and exposure correction were applied to the raw RGB images. The corrected images were orthorectified by ray-tracing image pixels to a lidar-derived digital surface model (DSM) mesh using the refined camera model, outputting orthorectified raster pixels on a regular grid. Flightline-level data were mosaicked by selecting per-pixel contributions from overlapping orthorectified images using line-of-sight (LOS) zenith angle minimization to reduce edge distortions. CHESS Project Description: The Colorado Headwaters Ecological Spectroscopy Study (CHESS) comprised a multi-week airborne remote sensing and field observation campaign in the Upper Gunnison Basin, Colorado, conducted in June and July of 2025. Airborne remote sensing was conducted by the National Ecological Observatory Network Airborne Observation Platform (NEON AOP), concurrent with a field campaign run by the Rocky Mountain Biological Laboratory (RMBL), the Lawrence Berkeley National Laboratory (LBNL) and SLAC National Accelerator Laboratory Watershed Function Science Focus Area (SFA), and NASA-JPL (Jet Propulsion Laboratory) Earth Surface Mineral Dust Source Investigation (EMIT) program. Between June 10 and July 18, 2025, the NEON AOP flight team collected high-resolution aerial imaging spectroscopy and Light Detection and Ranging (LiDAR) data over three domains: the Upper East River (CRBU), Almont Triangle (ALMO), and the Upper Taylor Basin (UPTA). In coordination with the flights, a field campaign acquired ground-truth observations, including observations of vegetation composition, foliar traits, forest demography, and subsurface properties in 18 core sampling areas within the domains. Additional surface water observations were taken at over 380 point locations. All CHESS campaign datasets can be found within the CHESS ESS-DIVE data portal: https://data.ess-dive.lbl.gov/portals/chess. Funding Acknowledgement: Field and remote-sensing data acquisition was performed under a grant from the National Aeronautics and Space Administration (80NSSC24K1005). This work was also supported by the Watershed Function Science Focus Area at Lawrence Berkeley National Laboratory funded by the US Department of Energy, Office of Science, Biological and Environmental Research under Contract No. DE-AC02-05CH11231.

2018 NEON and 2025 CHESS Campaigns↗
NASA NTRS2002

Serendipitous Detections of XTE J1906+09 with the Rossi X-Ray Timing Explorer

The 89 s X-ray pulsar XTE J1906+09 was discovered during Rossi X-Ray Timing Explorer (RXTE) observations of SGR 1900+14 in 1996. Because of monitoring campaigns of SGR 1900+14, XTE J1906+09 was also monitored regularly in 1996 September, 1998 May-June, 1998 August-1999 July, and 2000 March-2001 January. A search for pulsations resulted in detections of only the two previously reported outbursts in 1996 September and 1998 August-September. Pulsed flux upper limits for the rest of the observations show that XTE J1906+09 is a transient X-ray pulsar and likely has a Be star companion. The RXTE all-sky monitor did not reveal XTE J1906+09. Pulse-timing analysis of the second outburst discovered a sinusoidal signature in the pulse frequencies that is likely produced by an orbital periastron passage. Fits to pulse phases using an orbital model and quadratic phase model have chi(exp 2) minima at orbital periods of 26-30 days for fixed mass functions of 5, 10, 15, and 20 solar masses. The pulse shape showed energy- and intensity-dependent variations. Pulse-phase spectroscopy quantified the energy-dependent variations. The phase-averaged spectrum used the pulse minimum spectrum as the background spectrum to eliminate effects from SGR 1900+14 and the Galactic ridge and was well fitted by an absorbed power law with a high-energy cutoff with column density N(sub H) = 6 +/- 1 x 10(exp 22)/sq cm, a photon index of 1.01 +/- 0.08, cutoff energy E(sub cut) = 11 +/- 1 keV, and e-folding energy E(sub fold) = 19 +/- 4 keV. Estimated 2-10 keV peak fluxes, corrected for contributions from the Galactic ridge and SGR 1900+14, are 6 x l0(exp -12) and 1.1 x 10(exp -10) ergs/sq cm/s for the 1996 and 1998 outbursts, respectively. XTE J1906+09 may be part of an unusual class of Be/X-ray binaries that do not lie on the general spin period versus orbital period correlation with the majority of Be/X-ray binaries.

Wilson, Colleen A.↗
NASA NTRS1973

Evidence for the binary nature of 2U 1700-37.

Analysis of Uhuru data on the X-ray source 2U 1700-37 has revealed the presence of regular occultations. The period is 3.4 days with an occultation duration of 1.1 days. The source is highly variable on a time scale of minutes. Significant intensity variations have also been observed on a time scale of 0.1 second with a 3 sigma upper limit of 17 percent on the percentage periodically pulsed. The energy spectrum is flat and shows a large amount of low-energy absorption. The data indicate that 2U 1700-37 is an X-ray source in a binary system. The 6.6-mag O7f star HD 153919 is suggested as the optical candidate.-

Jones, C.↗
NASA NTRS2004

Program Monitoring with LTL in EAGLE

We briefly present a rule-based framework called EAGLE, shown to be capable of defining and implementing finite trace monitoring logics, including future and past time temporal logic, extended regular expressions, real-time and metric temporal logics (MTL), interval logics, forms of quantified temporal logics, and so on. In this paper we focus on a linear temporal logic (LTL) specialization of EAGLE. For an initial formula of size m, we establish upper bounds of O(m(sup 2)2(sup m)log m) and O(m(sup 4)2(sup 2m)log(sup 2) m) for the space and time complexity, respectively, of single step evaluation over an input trace. This bound is close to the lower bound O(2(sup square root m) for future-time LTL presented. EAGLE has been successfully used, in both LTL and metric LTL forms, to test a real-time controller of an experimental NASA planetary rover.

Barringer, Howard↗
NASA NTRS1991

The interpretation of crustal dynamics data in terms of plate motions and regional deformation near plate boundaries

During our participation in the NASA Crustal Dynamics Project under NASA contract NAS-27339 and grant NAG5-814 for the period 1982-1991, we published or submitted for publication 30 research papers and 52 abstracts of presentations at scientific meetings. In addition, five M.I.T. Ph.D. students (Eric Bergman, Steven Bratt, Dan Davis, Jeanne Sauber, Anne Sheehan) were supported wholly or in part by this project during their thesis research. Highlights of our research progress during this period include the following: application of geodetic data to determine rates of strain in the Mojave block and in central California and to clarify the relation of such strain to the San Andreas fault and Pacific-North American plate motions; application of geodetic data to infer post seismic deformation associated with large earthquakes in the Imperial Valley, Hebgen Lake, Argentina, and Chile; determination of the state of stress in oceanic lithosphere from a systematic study of the centroid depths and source mechanisms of oceanic intraplate earthquakes; development of models for the state of stress in young oceanic regions arising from the differential cooling of the lithosphere; determination of the depth extent and rupture characteristics of oceanic transform earthquakes; improved determination of earthquake slip vectors in the Gulf of California, an important data set for the estimation of Pacific-North American plate motions; development of models for the state of stress and mechanics of fold-and-thrust belts and accretionary wedges; development of procedures to invert geoid height, residual bathymetry, and differential body wave travel time residuals for lateral variations in the characteristic temperature and bulk composition of the oceanic upper mantle; and initial GPS measurements of crustal deformation associated with the Imperial-Cerro Prieto fault system in southern California and northern Mexico. Full descriptions of the research conducted on these topics may be found in the Semi-Annual status Reports submitted regularly to NASA over the course of this project and in the publications listed.

Solomon, Sean C.↗
NASA NTRS1992

Ions with low charges in the solar wind as measured by SWICS on board Ulysses

We present new data on rare ions in the solar wind. Using the Ulysses-SWICS instrument with its very low background we have searched for low-charge ions during a 6-d period of low-speed solar wind and established sensitive upper limits for many species. In the solar wind, we found He(1+)/He(2+) of less than 5 x 10 exp -4. This result and the charge state distributions of heavier elements indicate that all components of the investigated ion population went through a regular coronal expansion and experienced the typical electron temperatures of 1 to 2 million Kelvin. We argue that the virtual absence of low-charge ions demonstrates a very low level of nonsolar contamination in the source region of the solar wind sample we studied. Since this sample showed the FlP effect typical for low-speed solar wind, i.e., an enhancement in the abundances of elements with low first ionization potential, we conclude that this enhancement was caused by an ion-atom separation mechanism operating near the solar surface and not by foreign material in the corona.

Geiss, J.↗
NASA NTRS1985

Short meridional scale anomalies in the lower stratosphere and upper troposphere

Consequences of the discovery of short-meridional scale anomalies (SMSAs) in the lower stratospheric temperature field in the analyses of Microwave Sounding Unit channel 4 data are examined. The time scales of these features, their presence in the u-wind, v-wind, and geopotential, and the heat and momentum fluxes resulting from the SMSAs are investigated. The SMSAs are not regularly propagating features. Latitudinal movement occurs in both northern and southern directions. Both eastward movement and stationarity are observed in the zonal direction. In episodes, SMSAs require a few days to grow, persist from a few days to two weeks, and then decay over a few days. Heat fluxes are small. It is suggested that the SMSAs are possibly caused by secondary effects of baroclinic disturbances.

Newman, P. A.↗
DOE OSTI2023

FPGA Acceleration of GCN in Light of the Symmetry of Graph Adjacency Matrix

Graph Convolutional Neural Networks (GCNs) are widely used to process large-scale graph data. Different from deep neural networks (DNNs), GCNs are sparse, irregular, and unstructured, posing unique challenges to hardware acceleration with regular processing elements (PEs). In particular, the adjacency matrix of a GCN is extremely sparse, leading to frequent but irregular memory access, low spatial/temporal data locality and poor data reuse. Furthermore, a realistic graph usually consists of unstructured data (e.g., unbalanced distributions), creating significantly different processing times and imbalanced workload for each node in GCN acceleration. To overcome these challenges, we propose an end-to-end hardware-software co-design to accelerate GCNs on resource-constrained FPGAs with the features including: (1) A custom dataflow that leverages symmetry along the diagonal of the adjacency matrix to accelerate feature aggregation for undirected graphs. We utilize either the upper or the lower triangular matrix of the adjacency matrix to perform aggregation in GCN to improve data reuse. (2) Unified compute cores for both aggregation and transform phases, with full support to the symmetry-based dataflow. These cores can be dynamically reconfigured to the systolic mode for transformation or as individual accumulators for aggregation in GCN processing. (3) Preprocessing of the graph in software to rearrange the edges and features to match the custom dataflow. This step improves the regularity in memory access and data reuse in the aggregation phase. Moreover, we quantize the GCN precision from FP32 to INT8 to reduce the memory footprint without losing the inference accuracy. We implement our accelerator design in Intel Stratix10 MX FPGA board with HBM2, and demonstrate 1.3x-110.5x improvement in end-to-end GCN latency as compared to the state-of the-art FPGA implementations, on the graph datasets of Cora, Pubmed, Citeseer and Reddit.

Nair, Gopikrishnan R.↗
NASA NTRS2017

Wavenumber-frequency Spectra of Pressure Fluctuations on a Generic Space Vehicle Measured via Unsteady Pressure-Sensitive Paint

Time histories of pressure fluctuations on a generic, hammerhead space vehicle model were measured using unsteady Pressure-Sensitive Paint (uPSP). The test was conducted in the 11-foot transonic wind tunnel of NASA Ames Research Center over a Mach number range of 0.6 M 1.2, and angles of attack of -4 4. The model was coated with a porous binder and PtTFPP-based porous polymer paint. An elaborate system of four high-speed cameras, and forty LED lamps was used for image acquisition. Various steps for image registration, reduction of shot noise, photogrammetry procedure to map images from the four cameras on a grid for the model, and finally a calibration procedure to convert the measured fluctuations in light intensity to fluctuating pressure, are discussed in the paper. The calibration process using a set of unsteady pressure sensors mounted on the model, was found to overcome some of the inherent problems of the fast response paint, such as rapid photo-degradation, non-linearity in pressure response, and significant temperature sensitivity. Comparison of spectra of pressure fluctuations between UPSP and pressure sensors demonstrated the ability of the paint to faithfully follow fluctuations up to 10 kHz, the maximum attempted. It was also found that the camera bit-depth and the illumination level limited the lowest measurable levels of pressure fluctuations to around 140dB. The large data set exposed various critical transonic flow physics not seen before, such as a coupling of the shock motion on the Payload Fairing (PF) with the separated flow region on the upper stage of the launch vehicle, and upstream convection of pressure fluctuation on PF at certain Mach numbers. The data also confirmed the expectation of a general lowering of the coefficient of pressure fluctuation with Mach number. The availability of the data set on a dense, regularly-spaced, surface grid allowed for the calculation of wavenumber-frequency (k-) spectra via straightforward applications of Fourier transform. The k- spectra were compared for the separated flow regions on the Second Stage, and the shock-boundary layer interactions on PF. The former showed self-similarity with Mach number while the latter was distinctly different, and confirmed the upstream propagation of pressure fluctuations. The k- spectra were dominated by the convected fluctuations; the acoustic domain was not discernable. These data, valuable for the vibro-acoustics analysis of aerospace vehicles, are believed to be the first obtained for the transonic flight regime, and pave the path for application on production models of aerospace vehicles.

buffet↗