Technology Focus Computers/Electronics - NASA · 2013-04-10 · Technology Focus Computers/Electronics Software Materials Mechanics Machinery/Automation Manufacturing ... High-Rate

Technology Focus

Computers/Electronics

Software

Materials

Mechanics

Machinery/Automation

Manufacturing

Bio-Medical

Physical Sciences

Information Sciences

Books and Reports

12-04 December 2004

https://ntrs.nasa.gov/search.jsp?R=20110020440 2020-03-18T00:25:04+00:00Z

INTRODUCTIONTech Briefs are short announcements of innovations originating from research and develop-

ment activities of the National Aeronautics and Space Administration. They emphasizeinformation considered likely to be transferable across industrial, regional, or disciplinary linesand are issued to encourage commercial application.

Availability of NASA Tech Briefs and TSPsRequests for individual Tech Briefs or for Technical Support Packages (TSPs) announced herein shouldbe addressed to

National Technology Transfer CenterTelephone No. (800) 678-6882 or via World Wide Web at www2.nttc.edu/leads/

Please reference the control numbers appearing at the end of each Tech Brief. Information on NASA’s Commercial Technology Team, its documents, and services is also available at the same facility or on the World Wide Web at www.nctn.hq.nasa.gov.

Commercial Technology Offices and Patent Counsels are located at NASA field centers to providetechnology-transfer access to industrial users. Inquiries can be made by contacting NASA field centersand program offices listed below.

Ames Research CenterLisa L. Lockyer(650) [email protected]

Dryden Flight Research CenterGregory Poteat(661) [email protected]

Goddard Space Flight CenterNona Cheeks(301) [email protected]

Jet Propulsion LaboratoryKen Wolfenbarger(818) [email protected]

Johnson Space CenterCharlene E. Gilbert(281) [email protected]

Kennedy Space CenterJim Aliberti(321) [email protected]

Langley Research CenterJesse Midgett(757) [email protected]

John H. Glenn Research Center atLewis FieldLarry Viterna(216) [email protected]

Marshall Space Flight CenterVernotto McMillan(256) [email protected]

Stennis Space CenterRobert Bruce(228) [email protected]

Carl RaySmall Business InnovationResearch Program (SBIR) &Small Business TechnologyTransfer Program (STTR)(202) 358-4652 [email protected]

Benjamin NeumannInnovativeTechnology TransferPartnerships (Code TD)(202) [email protected]

John MankinsOffice of Space Flight (Code TD)(202) 358-4659 [email protected]

Terry HertzOffice of Aero-SpaceTechnology (Code RS)(202) 358-4636 [email protected]

Glen MucklowOffice of Space Sciences(Code SM)(202) 358-2235 [email protected]

Roger CrouchOffice of Microgravity ScienceApplications (Code U)(202) 358-0689 [email protected]

Granville PaulesOffice of Mission to Planet Earth(Code Y) (202) 358-0706 [email protected]

NASA Field Centers and Program Offices

NASA Program Offices

At NASA Headquarters there are seven major program offices that develop and oversee technology projects of potential interest to industry:

NASA Tech Briefs, December 2004 1

5 Technology Focus: DataAcquisition

5 High-Rate Digital Receiver Board

5 Signal Design for Improved Ranging AmongMultiple Transceivers

6 Automated Analysis, Classification, and Display ofWaveforms

6 Fast-Acquisition/Weak-Signal-Tracking GPSReceiver for HEO

7 Format for Interchange and Display of 3D Terrain Data

7 Program Analyzes Radar Altimeter Data

9 Electronics/Computers

9 Indoor Navigation Using Direction Sensor and Beacons

11 Software

11 Software Assists in Responding to AnomalousConditions

11 Software for Autonomous Spacecraft Maneuvers

11 WinPlot

12 Software for Automated Testing of Mission-Control Displays

13 Materials

13 Nanocarpets for Trapping Microscopic Particles

13 Precious-Metal Salt Coatings for DetectingHydrazines

14 Amplifying Electrochemical Indicators

15 Better End-Cap Processing for Oxidation-ResistantPolyimides

17 Machinery/Automation

17 Carbon-Fiber Brush Heat Exchangers

19 Physical Sciences

19 Solar-Powered Airplane With Cameras and WLAN

20 A Resonator for Low-Threshold FrequencyConversion

23 Information Sciences

23 Masked Proportional Routing

24 Algorithm Determines Wind Speed and DirectionFrom Venturi-Sensor Data

25 Feature-Identification and Data-CompressionSoftware

27 Books & Reports

27 Alternative Attitude Commanding and Controlfor Precise Spacecraft Landing

27 Inspecting Friction Stir Welding UsingElectromagnetic Probes

27 Helicity in Supercritical O2/H2 and C7H16/N2Mixing Layers

12-04 December 2004

This document was prepared under the sponsorship of the National Aeronautics and Space Administration. Neither the United States Governmentnor any person acting on behalf of the United States Government assumes any liability resulting from the use of the information contained in thisdocument, or warrants that such use will be free from privately owned rights.



Technology Focus: Data Acquisition

High-Rate Digital Receiver BoardThis board converts a personal computer into a versatile telemetry-data-acquisition system.Goddard Space Flight Center, Greenbelt, Maryland

A high-rate digital receiver (HRDR)implemented as a peripheral compo-nent interface (PCI) board has been de-veloped as a prototype of compact, gen-eral-purpose, inexpensive, potentiallymass-producible data-acquisition inter-faces between telemetry systems and per-sonal computers. The installation of thisboard in a personal computer togetherwith an analog preprocessor enables thecomputer to function as a versatile, high-rate telemetry-data-acquisition and de-modulator system. The prototype HRDRPCI board can handle data at rates ashigh as 600 megabits per second, in a va-riety of telemetry formats, transmittedby diverse phase-modulation schemesthat include binary phase-shift keyingand various forms of quadrature phase-shift keying. Costing less than $25,000(as of year 2003), the prototype HRDRPCI board supplants multiple racks ofolder equipment that, when new, costover $500,000. Just as the developmentof standard network-interface chips hascontributed to the proliferation of net-worked computers, it is anticipated thatthe development of standard chips

based on the HRDR could contribute toreductions in size and cost and increasesin performance of telemetry systems.

The circuitry on the HRDR board in-cludes an analog-to-digital converter(ADC) and two high-rate digital demod-ulator (HRDD) application-specific inte-grated circuits (ASICs). The HRDRboard accepts a baseband radio fre-quency telemetry modulation signal asinput. The ADC ASIC samples the input,and the sampled data is demultiplexedand sent to the two HRDD ASICs, whichdemodulate the signal, recover the clockand data components of the modulation,bit-synchronize the data, and serializeand forward the data to the next stage. Inaddition, the HRDD ASICs removeDoppler shifts from the carrier and datasignals. Within each HRDD ASIC, thedata are further demultiplexed by a fac-tor of two so that the HRDD processingtakes place in a total of four streams —each stream at a quarter of the incoming-data rate. Processing in multiple streamsat a rate lower than the incoming-datarate makes it possible to use complemen-tary metal oxide/semiconductor process-

ing circuitry that is relatively inexpensiveand could not perform adequately at theincoming-data rate.

The HRDR board outputs, depend-ing on output interface setup, one ortwo synchronous differential emittercoupled logic (ECL) clock and dataoutput streams. The output interfacecan be programmed to process and out-put demodulated telemetry data inmultiple ways — for example, to per-form CCSDS standard Viterbi decodingof convolutionally encoded data usingeither 3 bit soft symbols or hard sym-bols as inputs, interleave data I and Qchannels into a single output stream, orto output each channel independently.The user can easily choose the outputformat by means of a simple graphicaluser interface.

This work was done by Parminder Ghu-man, Thomas Bialas, and Clifford Bramboraof Goddard Space Flight Center and DavidFisher of QSS Group, Inc. For further infor-mation, access the Technical Support Package(TSP) free on-line at www.techbriefs.com/tspunder the Electronics/Computers category.GSC-14780-1

Signal Design for Improved Ranging Among Multiple TransceiversAcquisition, ranging, and telemetry signals are always present.NASA’s Jet Propulsion Laboratory, Pasadena, California

“Ultra-BOC” (where “BOC” signifies“binary offset carrier”) is the name of animproved generic design of microwavesignals to be used by a group of space-craft flying in formation to measureranges and bearings among themselvesand to exchange telemetry needed forthese measurements. Ultra-BOC couldalso be applied on Earth for diverse pur-poses — for example, measuring rela-tive positions of vehicles on highwaysfor traffic-control purposes and deter-mining the relative alignments of ma-chines operating in mines and of con-struction machines and structures atconstruction sites. Ultra-BOC providesfor rapid and robust acquisition of sig-

nals, even when signal-to-noise ratiosare low. The design further providesthat each spacecraft or other platformconstantly strives to acquire and trackthe signals from the other platformswhile simultaneously transmitting sig-nals that provide full range, bearing,and telemetry service to the other plat-forms. In Ultra-BOC, unlike in othersignal designs that have been consid-ered for the same purposes, it is not nec-essary to maneuver the spacecraft orother platforms to obtain the dataneeded for resolving integer-carrier-cycle phase ambiguities.

A prior design provided for thebroadcasting of acquisition signals, fol-

lowed by rough-clock-synchronizationsignals, followed by ranging and teleme-try signals. In contrast, in Ultra-BOC,the acquisition, ranging, and telemetrysignals are always present: Ultra-BOCcombines the BOC structure with con-stant transmission of unmodulatedtones (that is, subcarrier signals) as ac-quisition signals, plus low-rate clock syn-chronization data, a pseudorandom-noise (PRN) precise-ranging code, andtelemetry. A unique combination ofcode-division multiple access and fre-quency-division multiple access are em-ployed to support simultaneous trans-mission and reception of these signalsby many radio transceivers in the same

6 NASA Tech Briefs, December 2004

allocated frequency band while en-abling the use of the signals for precisemetrology.

The acquisition signals (unmodulatedtones) do extra duty by making it possi-ble to increase the precision of rangeand bearing measurements: The rang-ing code used in Ultra-BOC is adequateto resolve the ambiguity of a synthesizeddelay formed by a pair of closely-spacedunmodulated BOC tones. This delay isused to resolve the ambiguity on a more

widely spaced pair of tones. This processis continued with increasingly widelyspaced tones until either the range andbearing precision requirements are satis-fied by use of such pairs of tones or theinteger-cycle ambiguities in the phasesof the carrier signals are resolved. Therange measurements made in this man-ner can be more precise than are thosethat can be made by use of the PRNcodes alone, because (1) the delays syn-thesized from pairs of tones have smaller

errors attributable to system noise and(2) multipath-induced errors are theleading errors in ranging by use of PRNand the delays synthesized from pairs oftones are less susceptible to multipath-induced errors.

This work was done by Lawrence Young,Jeffrey Tien, and Jeffrey Srinivasan of Caltechfor NASA’s Jet Propulsion Laboratory.Further information is contained in a TSP(see page 1).NPO-40569

A report discusses the technical back-ground and design of the NavigatorGlobal Positioning System (GPS) re-ceiver a radiation-hardened receiverintended for use aboard spacecraft. Nav-igator is capable of weak signal acquisi-tion and tracking as well as much fasteracquisition of strong or weak signalswith no a priori knowledge or external

aiding. Weak-signal acquisition andtracking enables GPS use in high Earthorbits (HEO), and fast acquisition al-lows for the receiver to remain withoutpower until needed in any orbit. Signalacquisition and signal tracking are, re-spectively, the processes of finding anddemodulating a signal. Acquisition isthe more computationally difficult

process. Previous GPS receivers employthe method of sequentially searchingthe two-dimensional signal parameterspace (code phase and Doppler). Navi-gator exploits properties of the Fouriertransform in a massively parallel searchfor the GPS signal. This method resultsin far faster acquisition times [in thelab, 12 GPS satellites have been ac-

Fast-Acquisition/Weak-Signal-Tracking GPS Receiver for HEOGoddard Space Flight Center, Greenbelt, Maryland

Automated Analysis, Classification, and Display of WaveformsTrends in operation of systems that generate waveforms can be spotted in real time.John F. Kennedy Space Center, Florida

A computer program partly automatesthe analysis, classification, and display ofwaveforms represented by digital sam-ples. In the original application forwhich the program was developed, theraw waveform data to be analyzed by theprogram are acquired from space-shut-tle auxiliary power units (APUs) at asampling rate of 100 Hz. The programcould also be modified for application toother waveforms — for example, elec-trocardiograms.

Before this program became available,the raw APU waveforms were recordedon paper strip charts — a practice thatimposed a substantial workload onhuman operators and was not conduciveto consistently accurate, real-time analy-sis and classification. The program re-duces the operator workload, increasesthe accuracy of classifications, and pre-sents results in real time.

The program begins by performingprincipal-component analysis (PCA) of50 normal-mode APU waveforms. Eachwaveform is segmented. A covariancematrix is formed by use of the seg-

mented waveforms. Three eigenvectorscorresponding to three principal com-ponents are calculated. To generate fea-tures, each waveform is then projectedonto the eigenvectors. These featuresare displayed on a three-dimensional di-agram, facilitating the visualization ofthe trend of APU operations.

It is necessary to classify each of thenormal-mode waveforms as being char-acteristic of one of three mode typesknown among APU specialists as “nomi-nal,” “engine,” or “aero.” For this pur-pose, each waveform is segmented andits average energy is computed. For en-gine and aero modes, time informationis also used, and information aboutpeaks in the waveforms is used to deter-mine which mode is present.

It is also necessary, when there is amalfunction, to classify waveforms asbeing characteristic of one or moreerror mode(s). To enable such classifica-tion of a waveform in real time, it is nec-essary to prepare the software and asso-ciated data base in a prior process thatincludes a careful analysis of the wave-

form known to be associated with eachof at least five known error modes towhich the APUs are subject. For eacherror mode, some distinct features ofthe waveform are extracted. Thereafter,in operation, a waveform is automati-cally classified as belonging to an errormode according to a few rules based onthese features.

This program was written by ChimanKwan, Roger Xu, David Mayhew, and FrankZhang of Intelligent Automation, Inc., andAlan Zide and Jeff Bonggren of the BoeingCo. for Kennedy Space Center.

In accordance with Public Law 96-517,the contractor has elected to retain title to thisinvention. Inquiries concerning rights for itscommercial use should be addressed to:

Intelligent Automation, Inc.Attn. Chiman Kwan7519 Standish Place, Suite 200Rockville, MD 20855Phone: (301) 294-5238E-mail: [email protected] Refer to KSC-12568, volume and number

of this NASA Tech Briefs issue, and thepage number.


quired with no a priori knowledge in aLow-Earth-Orbit (LEO) scenario in lessthan one second]. Modeling has shownthat Navigator will be capable of acquir-ing signals down to 25 dB-Hz, appropri-ate for HEO missions. Navigator is builtusing the radiation-hardened ColdFire

microprocessor and housing the mostcomputationally intense functions indedicated field-programmable gate ar-rays. The high performance of the algo-rithm and of the receiver as a whole aremade possible by optimizing computa-tional efficiency and carefully weighing

tradeoffs among the sampling rate, dataformat, and data-path bit width.

This work was done by Luke Winternitz,Greg Boegner, and Steve Sirotzky of GoddardSpace Flight Center. Further information iscontained in a TSP (see page 1).GSC-14793-1

A computer program has been writtento perform several analyses of radar al-timeter data. The program was designed toimprove on previous methods of analysis ofaltimeter engineering data by (1) facilitat-ing and accelerating the analysis of largeamounts of data in a more direct mannerand (2) improving the ability to estimateperformance of radar-altimeter instrumen-tation and provide data corrections. Thedata in question are openly available to theinternational scientific community and can

be downloaded from anonymous file-trans-fer-protocol (FTP) locations that are acces-sible via links from altimetry Web sites. Thesoftware estimates noise in range measure-ments, estimates corrections for electro-magnetic bias, and performs statisticalanalyses on various parameters for com-parison of different altimeters. Whereasprior techniques used to perform similaranalyses of altimeter range noise requirecomparison of data from repetitions ofsatellite ground tracks, the present soft-

ware uses a high-pass filtering technique toobtain similar results from single satellitepasses. Elimination of the requirement forrepeat-track analysis facilitates the analysisof large amounts of satellite data to assesssubtle variations in range noise.

This program was written by Doug Vande-mark and David Hancock of Goddard SpaceFlight Center and Ngan Tran of RaytheonCo. For further information, contact NonaCheeks at [email protected]

Program Analyzes Radar Altimeter DataGoddard Space Flight Center, Greenbelt, Maryland

Visible Scalable Terrain (ViSTa) is asoftware format for production, inter-change, and display of three-dimen-sional (3D) terrain data acquired bystereoscopic cameras of robotic visionsystems. ViSTa is designed to supportscalability of data, accuracy of displayedterrain images, and optimal utilizationof computational resources. In a ViSTafile, an area of terrain is represented, atone or more levels of detail, by coordi-nates of isolated points and/or verticesof triangles derived from a texture mapthat, in turn, is derived from original ter-

rain images. Unlike prior terrain-imagesoftware formats, ViSTa includes provi-sions to ensure accuracy of texture coor-dinates. Whereas many such formats arebased on 2.5-dimensional terrain mod-els and impose additional regularity con-straints on data, ViSTa is based on a 3Dmodel without regularity constraints.Whereas many prior formats require ex-ternal data for specifying image-data co-ordinate systems, ViSTa provides for theinclusion of coordinate-system datawithin data files. ViSTa admits high-speed loading and display within a Java

program. ViSTa is designed to minimizefile sizes and maximize compressibilityand to support straightforward reduc-tion of resolution to reduce file size forInternet-based distribution.

This program was written by Paul Backes,Mark Powell, Marsette Vona, Jeffrey Norris,and Jack Morrison of Caltech for NASA’s JetPropulsion Laboratory. Further informa-tion is contained in a TSP (see page 1).

This software is available for commerciallicensing. Please contact Don Hart of the Cal-ifornia Institute of Technology at (818) 393-3425. Refer to NPO-30600.

Format for Interchange and Display of 3D Terrain DataNASA’s Jet Propulsion Laboratory, Pasadena, California


Electronics/Computers

Indoor Navigation Using Direction Sensor and BeaconsPosition and orientation are determined from directions to at least three beacons. NASA’s Jet Propulsion Laboratory, Pasadena, California

A system for indoor navigation of a mo-bile robot includes (1) modulated in-frared beacons at known positions on thewalls and ceiling of a room and (2) a cam-eralike sensor, comprising a wide-anglelens with a position-sensitive photodetec-tor at the focal plane, mounted in a knownposition and orientation on the robot.The system also includes a computer run-ning special-purpose software thatprocesses the sensor readings to obtainthe position and orientation of the robotin all six degrees of freedom in a coordi-nate system embedded in the room.

For a given beacon imaged on thefocal plane, the output of the sensorcomprises two parameters that dependin a known way on the characteristics ofthe lens and the direction to the beaconin a coordinate system attached to thesensor and robot. If at least three bea-cons are within the field of view of thesensor, then the sensor outputs from ob-servations of all three beacons can becombined to obtain six parameters in-dicative of the directions to all threebeacons. These directions, in combina-

tion with the known positions of the bea-cons, uniquely determine the positionand orientation of the robot in theroom. Equivalently, the six parametersconstitute, in principle, sufficient data tolocate the robot in all six degrees of free-dom by solving the equations that ex-press the applicable geometric relation-ships summarized above.

The nature of a position-sensitivephotodetector is such that it is not pos-sible to measure the centroids of twobeacon images simultaneously. There-fore, it is necessary to provide for illu-mination of the beacons in rapid suc-cession and to provide means for theimage-data-processing software to rec-ognize which beacon is under observa-tion at a given instant. To satisfy thisneed, the beacons are turned on andoff in a sequence that coincides with apredetermined code. The sensor sub-system accumulates beacon readingsand their times until it begins to recog-nize the code sequence. Thereafter, thecomputer processes the readings fromthe recognized beacons within the field

of view of the sensor.The equations for the geometric rela-

tionships are nonlinear. The software includes a module that solves these equa-tions by means of an iterative optimiza-tion procedure, in which it strives to finda position and orientation that, when in-serted in the equations, minimizes a mea-sure of the difference between the actualsensor readings and the sensor readingspredicted by the equations.

Another software module provides aninitial guess of position and orientationto start the optimization procedure.Knowing which beacons are in view, thismodule applies to the equations for anumber of postulated robot poses anddetermines which pose, when insertedin the equations yields the closest matchto the sensor readings. The closestmatch becomes the initial guess for theoptimization procedure.

This work was done by Joel Shields andMuthu Jeganathan of Caltech for NASA’s JetPropulsion Laboratory. Further informa-tion is contained in a TSP (see page 1).NPO-40730


Software

Software Assists in Respond-ing to Anomalous Conditions

Fault Induced Document Retrieval Of-ficer (FIDO) is a computer program thatreduces the need for a large and costlyteam of engineers and/or technicians tomonitor the state of a spacecraft and asso-ciated ground systems and respond toanomalies. FIDO includes artificial-intelli-gence components that imitate the rea-soning of human experts with referenceto a knowledge base of rules that repre-sent failure modes and to a database ofengineering documentation. These com-ponents act together to give an unskilledoperator instantaneous expert assistanceand access to information that can enableresolution of most anomalies, without theneed for highly paid experts. FIDO pro-vides a system state summary (a config-urable engineering summary) and docu-mentation for diagnosis of a potentiallyfailing component that might havecaused a given error message or anomaly.FIDO also enables high-level browsingof documentation by use of an interfaceindexed to the particular error message.The collection of available documents in-cludes information on operations and as-sociated procedures, engineering problemreports, documentation of components,and engineering drawings. FIDO also af-fords a capability for combining informa-tion on the state of ground systems withdetailed, hierarchically-organized, hyper-text-enabled documentation.

This program was written by Mark James, F.Kronbert, A. Weiner, T. Morgan, B. Stroozas, F.Girouard, A. Hopkins, L. Wong, J. Kneubuhl,and R. Malina of Caltech for NASA’s JetPropulsion Laboratory. Further informa-tion is contained in a TSP (see page 1).

This software is available for commercial li-censing. Please contact Don Hart of the Cali-fornia Institute of Technology at (818) 393-3425. Refer to NPO-40361.

Software for AutonomousSpacecraft Maneuvers

The AutoCon computer programs fa-cilitate and accelerate the planning andexecution of orbital control maneuversof spacecraft while analyzing and resolv-ing mission constraints. AutoCon-F is ex-ecuted aboard spacecraft, enabling thespacecraft to plan and execute maneu-vers autonomously; AutoCon-G is de-

signed for use on the ground. The AutoCon programs utilize advancedtechniques of artificial intelligence, in-cluding those of fuzzy logic and natural-language scripting, to resolve multipleconflicting constraints and automaticallyplan maneuvers. These programs can beused to satisfy requirements for missionsthat involve orbits around the Earth, theMoon, or any planet, and are especiallyuseful for missions in which there are re-quirements for frequent maneuvers andfor resolution of complex conflictingconstraints. During operations, the soft-ware targets new trajectories, places andsizes maneuvers, and controls spacecraftburns. AutoCon-G provides a user-friendly graphical interface, and can beused effectively by an analyst with mini-mal training. AutoCon-F reduces latencyand supports multiple-spacecraft andformation-flying missions. The AutoConarchitecture supports distributive pro-cessing, which can be critical for forma-tion-control missions. AutoCon is com-pletely object-oriented and can easily beenhanced by adding new objects andevents. AutoCon-F was flight demon-strated onboard GSFC’s EO-1 spacecraftflying in formation with Landsat-7.

These programs were written by John Bris-tow, Dave Folta, Al Hawkins, and GregDell of Goddard Space Flight Center.Further information is contained in a TSP(see page 1).GSC-14629-1

WinPlotWinPlot is a powerful desktop graphical

analysis tool that allows the user to gener-ate displays of unrestrictive amounts ofdata. It is designed to operate on a Win-dows 98/NT/2000 based desktop plat-form. WinPlot was developed to fulfill theneed for fast and easily managed graphi-cal displays of NASA test articles and fa-cilities with extreme amount of test datain a desktop-computer environment.WinPlot features include seamless dis-plays of real-time and post-test-time datawith time and event synchronization ofdata from multiple sources. WinPlot alsoprocesses full scripting capability for au-tomation of processes. Entire analysisprocedures may be recorded and re-played with a single command. Users mayrecord their actions within WinPlot ormay write scripts using text editor. Scripts

may also call and execute other scripts,providing even greater automation oftasks. WinPlot is also unique in its abilityto plot large volumes of data on a desktopworkstation. Up to 1,000 test data filesmay be opened simultaneously with plotsgenerated containing up to 1,000 curvesper plot. WinPlot also has extensive abili-ties in generation of “on-the-fly” calcula-tions, reducing or eliminating the needfor external programs to generate thedata. Calculations may include a series ofrecorded parameters, constants, andmath functions. WinPlot’s ability to ex-port plots on a single mouse click canmake easy work of preparing presenta-tion material with office applications.One simply produces the plot with de-sired style and click of a button on thetool bar. Plots will be saved in a prede-fined folder with a sortable naming con-vention. One then just pastes the filesinto one’s presentation. The ease of get-ting data on the screen is just the begin-ning with the user having many ways tomanipulate data once on screen. Theuser can use the mouse to zoom in on anyarea of interest, use the arrow keys to panaround the view, or page up/down forgeneral zooming. One may also use themouse to select a slice of data and gener-ate an instant report of min, max, aver-age, range, sigma, or other values forplotted parameter within a slice. A singlemouse click can export data into aspreadsheet and execute a spreadsheetapplication. A user may plot a parameterfrom a number to tests and instantlygather statistical data from the display.Importing of data from spreadsheets is assimple as copying the data to the clip-board and, within WinPlot, importing theclipboard and selecting the parameters toplot. The software package runs on astandard Windows desktop system. Mem-ory and storage requirements are drivenby the amount of data desired to beviewed and/or stored locally. Under mostcircumstances, the recommended systemrequirements for the operating system issufficient for WinPlot. The source codemodules and dynamic libraries are in-cluded in the software, which allows userversatility in importing, defining, viewing,and printing data.

This program was written by John R.Moody of Computer Sciences Corp. for Mar-shall Space Flight Center. Further infor-mation is contained in a TSP (see page 1).MFS-31664


Software for AutomatedTesting of Mission-ControlDisplays

MCC Display Cert Tool is a set of softwaretools for automated testing of computer-terminal displays in spacecraft mission-con-trol centers, including those of the spaceshuttle and the International Space Sta-tion. This software makes it possible to per-

form tests that are more thorough, take lesstime, and are less likely to lead to erroneousresults, relative to tests performed manu-ally. This software enables comparison oftwo sets of displays to report command andtelemetry differences, generates test scriptsfor verifying telemetry and commands, andgenerates a documentary record contain-ing display information, including versionand corrective-maintenance data. At the

time of reporting the information for thisarticle, work was continuing to add a ca-pability for validation of display parametersagainst a reconfiguration file.

This program was written by Brian O’Haganof Johnson Space Center. For further infor-mation, contact the Johnson Commercial Tech-nology Office at (281) 483-3809. MSC-23573

Nanocarpets — that is, carpets of car-bon nanotubes — are undergoing devel-opment as means of trapping micro-scopic particles for scientific analysis.Examples of such particles include inor-ganic particles, pollen, bacteria, andspores. Nanocarpets can be character-ized as scaled-down versions of ordinary

macroscopic floor carpets, which trapdust and other particulate matter, albeitnot purposefully. Nanocarpets can alsobe characterized as mimicking both thestructure and the particle-trapping be-havior of ciliated lung epithelia, the car-bon nanotubes being analogous to cilia(see figure).

Carbon nanotubes can easily be chem-ically functionalized for selective trap-ping of specific particles of interest. Onecould, alternatively, use such otherthree-dimensionally-structured materi-als as aerogels and activated carbon forthe purposeful trapping of microscopicparticles. However, nanocarpets offerimportant advantages over these alterna-tive materials:• Nanocarpets are amenable to nonin-

trusive probing by optical means; and• Nanocarpets offer greater surface-to-

volume ratios.This work was done by Flavio Noca, Fei

Chen, Brian Hunt, Michael Bronikowski,Michael Hoenk, Robert Kowalczyk, andDaniel Choi of Caltech for NASA’s JetPropulsion Laboratory. Further informa-tion is contained in a TSP (see page 1).

In accordance with Public Law 96-517,the contractor has elected to retain title to thisinvention. Inquiries concerning rights for itscommercial use should be addressed to:

Innovative Technology Assets ManagementJPLMail Stop 202-2334800 Oak Grove DrivePasadena, CA 91109-8099(818) 354-2240E-mail: [email protected] to NPO-30659, volume and number



Materials

Nanocarpets for Trapping Microscopic ParticlesProperties of nanocarpets can be tailored for selective trapping.NASA’s Jet Propulsion Laboratory, Pasadena, California

Microscopic Particles (Bacillus pumilis spores) were trapped on a nanocarpet by immersing thenanocarpet in a spore-containing solution, then drying the nanocarpet.

Substrates coated with a precious-metal salt KAuCl4 have been found tobe useful for detecting hydrazine va-pors in air at and above a concentra-tion of the order of 0.01 parts per mil-lion (ppm). Upon exposure to aircontaining a sufficient amount of hy-drazine for a sufficient time, the coat-ing material undergoes a visiblechange in color. Although the color

change is only a qualitative indication,it can serve as an alarm of a hazardousconcentration of hydrazine or as adviceof the need for a quantitative measure-ment of concentration. Detection ofhydrazine vapors by this techniquecosts much less and takes less time thandoes laboratory analysis of sorbenttubes using high-performance liquidchromatography, which is the tech-

nique used heretofore to detect hy-drazines at concentrations down to0.01 ppm.

A substrate for use in this techniqueshould be made of a chemically inertmaterial (e.g., fiberglass filter paper).The substrate is uniformly coated with1 to 10 weight percent of the precious-metal salt in a solvent (e.g., dilute HCl)that does alter the physical characteris-

Precious-Metal Salt Coatings for Detecting HydrazinesColors change upon exposure to hydrazines and perhaps other hazardous gases.Lyndon B. Johnson Space Center, Houston, Texas


tics of the substrate. After driving offthe solvent by gentle heating and/or byuse of a vacuum, the coated substrate ispacked into an inert tube with open-ings at each end. (The dried precious-metal coating is somewhat sensitive tolight; the dried coated substrate shouldbe handled accordingly and stored inthe dark.)

The coated substrate is held in placewith small quantities of inert wadding(i.e., borosilicate glass wool). A gas suc-tion pump is attached to one end of thetube, and the air or other gas suspectedto contain hydrazine vapor is drawnthrough the tube at a specified pump-ing rate for an amount of time sufficientto obtain a sufficient chemical change

(and thus an observable color change)in the coating material. A semiquantita-tive relationship between the degree ofchemical change and the quantity ofvapor sampled can be established fromobservations of intensities of colorchanges and/or areas of color change intests on similarly prepared substrates andtubes using known concentrations of hy-drazine vapors.

In experiments, tubes containingKAuCl4-coated substrates prepared asdescribed above were exposed to 40-liter flows of air containing, variously,hydrazine, monomethylhydrazine, orunsymmetrical dimethylhydrazine atconcentrations of the order of 0.01 ppm.These exposures caused the colors of

the substrates to change from yellow tovarious purplish colors and, in one case,to black.

No such color changes were observedupon exposure of the KAuCl4-coatedsubstrates to flows of air that containedother gases (ammonia, isopropyl alco-hol, NO2, and H2). Whether or not otherprecious-metal coating materials couldbe used as color-change indicators ofthese or other nonhydrazine gases re-mains to be determined.

This work was done by Louis A. Dee andBenjamin Greene of Allied-Signal Aerospace Co.for Johnson Space Center. For further infor-mation, contact the Johnson Commercial Tech-nology Office at (281) 483-3809.MSC-22870

Amplifying Electrochemical IndicatorsReporter compounds can be formulated for high sensitivity and miniaturization of sensor units.Ames Research Center, Moffett Field, California

Dendrimeric reporter compounds havebeen invented for use in sensing and am-plifying electrochemical signals from mol-ecular recognition events that involvemany chemical and biological entities.These reporter compounds can be for-mulated to target specific molecules ormolecular recognition events. They canalso be formulated to be, variously, hy-drophilic or amphiphilic so that they aresuitable for use at interfaces between (1)aqueous solutions and (2) electrodes con-nected to external signal-processing elec-tronic circuits. The invention of these re-porter compounds is expected to enablethe development of highly miniaturized,low-power-consumption, relatively inex-pensive, mass-producible sensor units fordiverse applications, including diagnoses

of infectious and genetic diseases, testingfor environmental bacterial contamina-tion, forensic investigations, and detec-tion of biological warfare agents.

The multiple functionality of a reportercompound of this type is achievedthrough integration of a variety of chemi-cal moieties into each molecule. Thestructure and composition of such a mol-ecule is depicted schematically in the fig-ure and represented by the general for-mula ALBn. As used here, A signifies atargeting group, L signifies a linkinggroup, and B signifies an active group.

The targeting group (A) can includenucleic-acid intercalators or other or-ganic functional subgroups. It is designedto interact directly with a targeted mole-cule or molecular recognition event; that

is to say, it is designed to bring itself andthe rest of the reporter molecule into thevicinity of the target. Hence, the collec-tive effect of the targeting groups of mul-tiple reporter molecules is to concentratethe reporter compound in the region ofthe target compound or molecular recog-nition events that one seeks to detect.

An active group (B) is, more specifi-cally, either (1) electroactive in a mannerthat enables detection of an electro-chemical signal or (2) hydrophilic to en-hance solubility. It is preferable that thenumber (n) of B groups exceed 1. Thelinking group (L) comprises two moi-eties: (1) a linker between the targeting(A) group and the B groups and (2) anamplifying moiety, through which the Bgroups are connected in series, parallel,or a combination of series and parallelconnections in a dendritic structure.

The active (B) groups can also be char-acterized as indicator groups because theseare the ones that generate the desired elec-tronic sensory signals. Because they arelinked to the targeting group, the activegroups are concentrated in the vicinity ofthe target, and the probability that eachwill generate a signal is correspondingly in-creased. The multiple active groups, con-nected together in the dendritic molecularstructure, contribute to an aggregate signalmuch greater than that generated by a sin-gle-indicator reporter molecule. Depend-ing upon the specific formulation of a re-porter molecule according the invention,the primary signal could be as little as two

B

A

L

A Reporter Compound according to the invention can have any of a wide variety of dendritic struc-tures. The A, B, and L groups contribute synergistically to the overall effect of generating a highly am-plified primary electrochemical sensory signal.


times or more than a thousand times asgreat as that generated by a single-indica-tor reporter molecule. By increasing sig-nal-to-noise ratios relative to those avail-able from prior reporter compounds, theinvention of these reporter compounds

can be expected to facilitate the detectionof very small amounts of target com-pounds — for example, particular genes inblood samples.

This work was done by Wenhong Fan,Jun Li, and Jie Han of Ames Research

Center. Further information is containedin a TSP (see page 1).

Inquiries concerning rights for the commer-cial use of this invention should be addressedto the Patent Counsel, Ames Research Center,(650) 604-5104. Refer to ARC-14908-1.

Better End-Cap Processing for Oxidation-Resistant PolyimidesCross-linking in an inert atmosphere (as opposed to air) yields better results.John H. Glenn Research Center, Cleveland, Ohio

A class of end-cap compoundsthat increase the thermo-oxida-tive stability of polyimides of thepolymerization of monomeric re-actants (PMR) type has been ex-tended. In addition, an improvedprocessing protocol for this classof end-cap compounds has beeninvented.

The class of end-cap com-pounds was described in “EndCaps for More Thermo-Oxida-tive Stability in Polyimides”(LEW-17012), NASA Tech Briefs,Vol. 25, No. 10 (October 2001),page 32. To recapitulate: PMRpolyimides are often used as ma-trix resins of high-temperature-resistant composite materials.These end-cap compounds are intendedto supplant the norbornene end cap(NE) compound that, heretofore, hasserved to limit molecular weights duringoligomerization and, at high tempera-tures, to form cross-links that becomeparts of stable network molecular struc-tures. NE has been important to process-ability of high-temperature resins be-cause (1) in limiting molecular weights,it enables resins to flow more readily forprocessing and (2) it does not give offvolatile byproducts during final cureand, therefore, enables the productionof void-free composite parts. However,with respect to ability of addition poly-mers to resist oxidation at high tempera-ture, NE has been a “weak link.” Conse-quently, for example, in order to enable

norbornene-end-capped polyimide ma-trices to last for lifetimes up to 1,000hours, it is necessary to limit their usetemperatures to ≤315 °C.

Like NE, these end caps are also sub-ject to oxidation at high temperatures,but they oxidize in a different manner,such that the long-term stability of a poly-mer made with one of these end caps ex-ceeds the long-term stability of the corre-sponding polymer made with NE.Hence, use temperatures and/or life-times can be increased. The approachtaken in formulating these end caps is toseek derivatives that preserve the desir-able processing properties of NE whileexploiting one of the modes of thethermo-oxidative degradation of thenadic end cap in such a way as to retard

the overall thermo-oxidative degra-dation of the affected polymers.The figure depicts the genericmolecular structures of the priorversion and the present extendedversion of this class of end caps.Each end cap is a 1,2,3,6-tetrahy-drophthalic anhydride, substi-tuted in such a way as to lower thecross-linking temperature. Theend cap maintains its stability dur-ing imidization (at 200 °C) andcross-linking.If the imidization is carried out inair, then the end cap subsequentlyaromatizes in competition withcross-linking. This aromatizationis undesirable. Therefore, the im-proved processing protocol speci-

fies that the process be carried out in aninert atmosphere, wherein cross-linkingis the predominant, if not the exclusive,reaction path. Following cross-linking,the end cap is spontaneously converted,upon aging in air, to a thermally stablecapping group.

This work was done by Mary Ann B.Meador of Glenn Research Center andAryeh A. Frimer of Bar Ilan University, Is-rael. Further information is contained in aTSP (see page 1).

Inquiries concerning rights for the commer-cial use of this invention should be addressedto NASA Glenn Research Center, CommercialTechnology Office, Attn: Steve Fedor, MailStop 4-8, 21000 Brookpark Road, Cleveland,Ohio 44135. Refer to LEW-17429.

R1

R2

O

O

O

R1

R4

O

O

O

R2

R3

PRESENT (EXTENDED) VERSION

PREVIOUS VERSION

End Caps of These Molecular Structures are alternatives to previ-ously reported end caps for increasing the thermo-oxidative stabil-ity of polyimides. R1 through R4 can be any of a variety of sub-stituents (for example, alkyl, alkoxy, aryl, halogen, or nitro), theinclusion of which reduces the cross-linking temperature.


Machinery/Automation

Carbon-Fiber Brush Heat ExchangersHigh thermal conductance between uneven surfaces could be achieved with low clamping force.Lyndon B. Johnson Space Center, Houston, Texas

Velvetlike and brushlike pads of carbonfibers have been proposed for use as me-chanically compliant, highly thermally con-ductive interfaces for transferring heat. Apad of this type would be formed by at-taching short carbon fibers to either orboth of two objects that one desires toplace in thermal contact with each other.

The purpose of using a thermal-contactpad of this or any other type is to reducethe thermal resistance of an interface be-tween a heat source (e.g., a module thatcontains electronic circuitry) and a heatsink (e.g., a common finned heat sink).Conventionally, to obtain high thermalconductance, a thermal interface is assem-bled by use of high contact pressure be-tween faying surfaces that match eachother precisely (e.g., both are preciselyflat). Unfortunately, high contact pressurenecessitates rigid components and strongfasteners and does not allow relative mo-tion between the clamped parts. Compli-ant rubber pads or thermally conductivegreases or adhesives are often used alter-

natively or in addition to precisely match-ing surfaces and high contact pressure.

The proposed carbon-fiber brush heatexchangers would offer high thermal con-ductance with mechanical compliance andlow contact pressure, even in the case ofsurfaces that are uneven, do not matcheach other precisely, are separated by rela-tively wide gaps, and/or move relative toeach other. In a given interface, the effec-tive surface area of the carbon fibers couldbe orders of magnitude larger than thenominal footprint area of an interface.

A given thermal interface could be ei-ther single-sided (consisting of a brush oneither the heat source or the heat sink) ordouble-sided (consisting of brushes onboth the source and the sink). If the carbonfibers had high thermal conductivity andwere well connected to a substrate, theywould tend to isothermalize with the sub-strate and become thermally efficient fins.High-thermal-conductivity fibers would bewell suited for brush heat exchangers be-cause they are straight, are available in

small diameters, and are compatible withmany materials, even at high temperatures.

Double-sided carbon-fiber brush heatexchangers would be related to interleav-ing metal-fin heat exchangers, but for agiven footprint area, the carbon-fiberbrush heat exchangers would have largerradiating surface areas and would weighless. The high thermal conductances occa-sioned by the use of carbon-fiber brushheat exchangers could be utilized to de-crease the sizes and weights of heat sinks(including radiators) for a given heat-dissi-pation rate, increase heat-dissipation ratesfor heat sinks of a given size and weight,and/or enable heat-generating equipmentto operate at lower temperatures. Theelimination of the need for structures to re-sist large thermal-interface clamping forceswould enable further weight reductions.

This work was done by Timothy R. Knowlesof Energy Science Laboratories, Inc., for John-son Space Center. For further information,contact the Johnson Commercial TechnologyOffice at (281) 483-3809. MSC-23018


Physical Sciences

An experimental airborne remote-sensing system includes a remotely con-trolled, lightweight, solar-powered air-plane (see figure) that carries twodigital-output electronic cameras andcommunicates with a nearby groundcontrol and monitoring station via awireless local-area network (WLAN).The speed of the airplane — typically<50 km/h — is low enough to enable loi-tering over farm fields, disaster scenes,or other areas of interest to collect high-resolution digital imagery that could bedelivered to end users (e.g., farm man-agers or disaster-relief coordinators) innearly real time.

In addition to achieving the desiredflight, remote-sensing, remote-control,and remote-monitoring capabilities, one

of the goals in the development of thissystem has been minimizing its costthrough the use of commercial off-the-shelf hardware and software. Accord-ingly, the mention of brand names inthe following description of the systemdoes not constitute an endorsement andis not intended to exclude hardware andsoftware of different brand names thatafford equivalent capabilities.

One of the camera systems — for ac-quiring high-resolution red/green/blueimages — includes a Hasselblad 555ELDcamera body assembled with a KodakProfessional DCS Pro Back 4,000-by-4,000-pixel charge-coupled-device (CCD)array and a color filter array. The othersystem — for imaging in three narrowwavelength bands — comprises a Dun-

canTech MS3100 camera with a singleNikon 35-mm lens, which, in combina-tion with a dichroic prism, focuses in-coming light through three separatenarrow-band filters onto three 1,280-by-1,024-pixel CCD arrays. The three wave-length bands are 760±20, 660±10, and580±10 nm.

The WLAN is implemented by use ofCisco Aironet 340-series Ethernetbridges, which operate at frequenciesbetween 2.4 and 2.5 GHz. These bridgesare capable of functioning as bidirec-tional, line-of-sight, high-speed datalinks between two or more networks (inthis case, an airborne and a ground-based network). These bridges wereoriginally designed as building-to-build-ing links but are advertised as being ca-

Solar-Powered Airplane With Cameras and WLANHigh-resolution images are sent to a ground station in nearly real time.Ames Research Center, Moffett Field, California

A Solar-Powered, Unpiloted Airplane includes two payload pods, on the underside of the middle section, that carry electronic cameras and telemetry subsystems.


pable of data rates of 11 Mb/s over dis-tances up to 40 km. The WLAN is con-figured for remote control of the cam-era and transmission of acquiredimagery to the ground station. A bridgein the airborne network serves as thelink between an airborne system payloadcomputer and an omnidirectional stubantenna on the underside of the air-plane. A bridge in the ground stationserves as a link between the ground an-tenna and a laptop computer. The re-mote-control software is installed inboth the system payload computer andthe portable laptop computer. Theground-based payload operator controlseach camera remotely by use of the lap-top computer.

Testing and development of the systemwere continuing at the time of reportingthe information for this article. Particu-larly notable is a flight test, performed inSeptember 2002, to demonstrate safeand effective operation of the system inan agricultural setting in FAA controlledairspace. The airplane was flown for fourhours over a 15-km2 coffee plantation inHawaii, under supervision by Honoluluair-traffic controllers as though it were aconventionally piloted aircraft. The air-plane was shown to be capable of flyingplanned routes and to perform sponta-neous maneuvers to collect imagery incloud-free areas. The WLAN was capableof downloading image data at rates ex-ceeding 5 Mb/s, making all image data

available for viewing, enhancing, andprinting within a few minutes of collec-tion. During the latter part of the flight,the payload was operated over an estab-lished wide-area network by an operatorlocated on the United States mainland ata distance of 4,000 km.

This work was done by Robert G. Higgins,Steve E. Dunagan, Don Sullivan, Robert Slye,and James Brass of Ames Research Center;Joe G. Leung, Bruce Gallmeyer, Michio Aoy-agi, and Mei Y. Wei of Dryden Flight ResearchCenter; Stanley R. Herwitz of Clark Univer-sity; Lee Johnson and Jian Zheng of Califor-nia State University; and John C. Arvesen ofKauai Airborne Sciences. Further informationis contained in a TSP (see page 1).ARC-15061

A proposed toroidal or disklike dielec-tric optical resonator (dielectric opticalcavity) would be made of an opticallynonlinear material and would be opti-mized for use in parametric frequencyconversion by imposition of a spatiallyperiodic permanent electric polariza-tion. The poling (see figure) would sup-press dispersions caused by both the ma-terial and the geometry of the opticalcavity, thereby effecting quasi-matchingof the phases of high-resonance-quality(high-Q) whispering-gallery electromag-netic modes. The quasi-phase-matchingof the modes would serve to maximizethe interactions among them. Such aresonator might be a prototype of a fam-ily of compact, efficient nonlinear de-

vices for operation over a broad range ofoptical wavelengths.

A little background information isprerequisite to a meaningful descriptionof this proposal:• Described in several prior NASA Tech

Briefs articles, the whispering-gallerymodes in a component of spheroidal,disklike, or toroidal shape are wave-guide modes that propagate circum-ferentially and are concentrated in anarrow toroidal region centered onthe equatorial plane and located nearthe outermost edge.

• For the sake of completeness, it mustbe stated that even though optical res-onators of the type considered hereare solid dielectric objects and light is

confined within them by total internalreflection at dielectric interfaces with-out need for mirrors, such compo-nents are sometimes traditionallycalled cavities because their effectsupon the light propagating withinthem are similar to those of true cavi-ties bounded by mirrors.

• For a given set of electromagneticmodes interacting with each other inan optically nonlinear material (e.g.,modes associated with the frequenciesinvolved in a frequency-conversionscheme), the threshold power for os-cillation depends on the mode vol-umes and the mode-overlap integral.

• Whispering-gallery modes are attrac-tive in nonlinear optics because theymaximize the effects of nonlinearitiesby occupying small volumes and af-fording high Q values.In designing a cavity according to the

proposal, one could reduce the modevolume and increase the mode-overlapintegral, and thereby reduce the thresh-old power needed for oscillation, rela-tive to those of a the nonlinear materialin bulk form. The amplitude, configura-tion, and periodicity of the poling wouldbe chosen so that the whispering-gallerymodes to be quasi-phased-matched werethe modes associated with the pump, sig-nal, and idler frequencies involved inthe parametric frequency conversion. Itwould be necessary to perform somecomplex computations, including calcu-lation of quantum-mechanical mode

A Resonator for Low-Threshold Frequency ConversionA nonlinear dielectric whispering-gallery resonator would be poled for quasi-phase-matching.NASA’s Jet Propulsion Laboratory, Pasadena, California

A Disk of LiNbO3 or perhaps another suitable optically nonlinear material would be poled periodically,possibly in one of these two patterns. The labels E and H denote the electric and magnetic field axes,respectively, of a whispering-gallery electromagnetic field. The labels Z denote the vectors of perma-nent electric polarization.

Z Z H

E

Z Z H

E


wave functions and evaluation of mode-overlap integrals, in order to analyze theperformance of the cavity and design itfor quasi-phase-matching.

The nonlinear cavity material wouldlikely be commercially available flat, Z-cut LiNbO3. The optimum poling geom-etry would be the one symmetrical aboutthe center, shown on the left side of thefigure. However, the imposition of cen-trally symmetric poling would be diffi-cult. It would be much easier to use aslice of LiNbO3 as supplied commer-

cially with poling stripes; this would en-tail an increase in the threshold powerfor oscillation, relative to the optimumsymmetrical poling pattern. On theother hand, the striped poling would en-able the parametric generation of oscil-lations at multiple frequencies.

This work was done by Vladimir Iltchenko,Andrey Matsko, Anatoliy Savchenkov, andLute Maleki of Caltech for NASA’s JetPropulsion Laboratory. Further informa-tion is contained in a TSP (see page 1).

In accordance with Public Law 96-517, the

contractor has elected to retain title to this in-vention. Inquiries concerning rights for itscommercial use should be addressed to:

Innovative Technology Assets ManagementJPLMail Stop 202-2334800 Oak Grove DrivePasadena, CA 91109-8099(818) 354-2240E-mail: [email protected] to NPO-30638, volume and number

of this NASA Tech Briefs issue, and the pagenumber.


Information Sciences

Masked Proportional RoutingThis procedure enables adaptation to changing network conditions.Ames Research Center, Moffett Field, California

Masked proportional routing is an im-proved procedure for choosing links be-tween adjacent nodes of a network forthe purpose of transporting an entityfrom a source node (“A”) to a destina-tion node (“B”). The entity could be, forexample, a physical object to be shipped,in which case the nodes would representwaypoints and the links would representroads or other paths between waypoints.For another example, the entity could bea message or packet of data to be trans-mitted from A to B, in which case thenodes could be computer-controlledswitching stations and the links could becommunication channels between thestations. In yet another example, an en-tity could represent a workpiece whilelinks and nodes could represent, respec-tively, manufacturing processes andstages in the progress of the workpiecetowards a finished product. More gener-ally, the nodes could represent states ofan entity and the links could representallowed transitions of the entity.

The purpose of masked proportionalrouting and of related prior routing proce-dures is to schedule transitions of entitiesfrom their initial states (“A”) to their finalstates (“B”) in such a manner as to mini-mize a cost or to attain some other mea-sure of optimality or efficiency. Maskedproportional routing follows a distributed(in the sense of decentralized) approachto probabilistically or deterministicallychoosing the links. It was developed to sat-isfy a need for a routing procedure that1. Does not always choose the same

link(s), even for two instances charac-terized by identical estimated valuesof associated cost functions;

2. Enables a graceful transition fromone set of links to another set of linksas the circumstances of operation ofthe network change over time;

3. Is preferably amenable to separate op-timization of different portions of thenetwork;

4. Is preferably usable in a network in whichsome of the routing decisions are madeby one or more other procedure(s);

5. Preferably does not cause an entity tovisit the same node twice; and

6. Preferably can be modified so that

separate entities moving from A to Bdo not arrive out of order.Definitions of several terms are prereq-

uisite to even a brief summary of themathematical nature of masked propor-tional routing. Consider a network of Nnodes (N ≥2) including a source node Aand destination node B (see figure).Node i is directly connected to an arbi-trary number J(µ) of nodes, which are la-beled j = j1, j2, ..., j J(µ). The term µ repre-sents a characteristic or a set ofcharacteristics of an entity that one seeksto transport from node i to one of theconnected nodes j along the route fromA to B. The characteristics represented byµ could include the source and/or desti-nation node(s), the routing priority,and/or the time elapsed since leaving thesource node. Associated with node i is aJ(µ)-component vector, denoted a base-line proportion vector, p(i;µ).

In a deterministic version of maskedproportional routing, p(i;µ) is used tocompute a J(µ)-component vector, de-noted an applied proportion vector,p*(i;µ), that prevents the entity from vis-iting the same node more than once. Inthis case, if k is a node that has alreadybeen visited, then the jth component ofp*(i;µ) is made zero; that is, p*(i;µ)k=0.

In another version of masked propor-tional routing, there are computed (as de-scribed below) two other J(µ)-component

vectors, denoted Target(i;n(µ);µ) and Ac-tual(i;n(µ);µ), where n(µ) is a sequencenumber or a count at node i that may de-pend on one or more component(s) of µ.Except as described in the last sentence ofthis paragraph, the link from node i tonode j'(µ) is selected as being the onethat yields the largest difference betweenTarget(i;n(µ);µ) and Actual(i;n(µ);µ).The entity is then transported along the i-to-j'(µ) link. The vectors Target(i;n(µ);µ)and Actual(i;n(µ);µ) are computed itera-tively as follows:

Target(i;n(µ);µ) = α(µ)Target(i;n(µ)–1;µ) + β(µ)p*(i;µ)

andActual(i;n(µ)+1;µ) =

α(µ)Actual(i;n(µ);µ) +β(µ)Sent(i;j'(µ);n(µ);µ),

where α(µ) and β(µ) are selected realnumbers and Sent(i;j'(µ);n(µ);µ) is a J(µ)-component vector, the j'(µ)th componentof which is 1 and all other components ofwhich are 0. The exception mentionedabove applies in special circumstances inwhich the same link is optionally used totransport consecutively arriving entities.

This work was done by David Wolpert ofAmes Research Center. Further informa-tion is contained in a TSP (see page 1).

Inquiries concerning rights for the commer-cial use of this invention should be addressedto the Patent Counsel, Ames Research Center,(650) 604-5104. Refer to ARC-14366-1.

A(Source)

B(Destination)

j3

jJ(µ)

j1

j2i

An Entity Is Transported from node A via network links to node B.


Algorithm Determines Wind Speed and Direction From Venturi-Sensor DataSpeed and direction are calculated from the spatial distribution of pressure readings.John F. Kennedy Space Center, Florida

An algorithm computes the velocity ofwind from the readings of an instrumentlike the one described in “Three-Dimen-sional Venturi Sensor for Measuring Ex-treme Winds” (KSC-12435), NASA TechBriefs, Vol. 27, No. 9 (September 2003),page 32. To recapitulate: The sensor hasno moving parts and is a compact, ruggedmeans of measuring wind vectors havingmagnitudes of as much as 300 mph(134 m/s). The sensor includes a Venturi

gap bounded by a curved upper and acurved lower surface that are axisymmet-ric with respect to a vertical axis and mir-ror-symmetric with respect to a horizontalmidplane. One of the curved surfaces isinstrumented with multiple ports for mea-suring dynamic pressures (see figure).The sensor also incorporates auxiliary sen-sors for measuring temperature, relativehumidity, and static atmospheric pressure.

The design and operation of the sensor

are based on the concepts of (1) usingBernoulli’s equation (which expresses therelationship among variations of speed,density, and pressure along a streamline)to calculate the speed of the wind fromdifferences among the pressure readingsat the various ports; and (2) calculatingthe direction of the wind from the angu-lar positions of ports selected according tocomparisons among their pressure read-ings. The present algorithm performsthese calculations. Although the algo-rithm is much too complex to describehere in detail, it is worthwhile to expandon the major underlying physical andmathematical concepts:• The auxiliary measurements of tem-

perature and relative humidity areused, along with the measurement ofstatic pressure, to calculate the densityof air for use in Bernoulli’s equation.

• The pressure at the central port on theVenturi surface is always the lowest andis directly related to static pressure andthe wind speed by Bernoulli’s equation.

• The pressure readings from all theVenturi ports except the central onedepend on both the speed and direc-tion of the wind. Some convey more in-formation about speed, while someconvey more information about direc-tion. The algorithm combines infor-mation from all the readings to resolveuncertainties in calculating the speedand direction.

• Pressures at upwind ports are greaterthan those at the central and down-wind ports. Pressures are lowest at portslocated at angular positions orthogonalto the wind. These directional charac-teristics are utilized to calculate thewind direction to within an angular in-terval of 45° for the pressure-portarrangement shown in the figure.

• The wind direction can be estimatedmore accurately by means of a poly-nomial interpolation from the realpressure readings to a fictitious set ofpressure readings at ports in a rotatedversion of the real pattern of ports.This work was done by Jan A. Zysko and Jose

M. Perotti of Kennedy Space Center andJohn Randazzo of Dynacs, Inc. Further infor-mation is contained in a TSP (see page 1).KSC-12516

North

East

VenturiSurfaces

West

South

TOP VIEW OF ONE VENTURI SURFACE SHOWINGLOCATIONS OF PRESSURE PORTS

CROSS SECTION IN VERTICAL EAST-WEST PLANE

Pressure Ports at Multiple Locations on a Venturi surface provide samples of the spatial distributionof pressure, which distribution is directly related to the speed and direction of the wind.


A report discusses the continuing devel-opment of Windows Interface for Nomi-nal Displacement Selection (WINDS), acomputer program for automated analysisof images of the Sun and planets acquiredby scientific instruments aboard space-craft. WINDS is intended to afford capa-bilities for identification of features, mea-surement of displacements and velocities,analysis of terrain and of atmospheres,and synthesis of animation sequences ofimages of terrains and atmospheres fromsmall sets of samples by use of velocity-

based interpolation. A major element ofWINDS will be a nonlinear correlator ca-pable of tracking small features in com-plex image sequences. For dynamic imagesequences, the correlator will enable com-pression of data by factors >100. In pro-cessing image data, WINDS will take ac-count of such factors as texture in imagedata, rotation of features during measure-ment intervals, effects of viewing and solarillumination angles, and vertical structuresof atmospheres. WINDS will also take ac-count of positions, aiming directions, and

fields of view of cameras to determinethree-dimensional feature structures byuse of triangulation and stereoscopicanalysis techniques.

This work was done by Eric De Jong andJean Lorre of Caltech for NASA’s Jet Propul-sion Laboratory. Further information iscontained in a TSP (see page 1).

This software is available for commerciallicensing. Please contact Don Hart of the Cal-ifornia Institute of Technology at (818) 393-3425. Refer to NPO-30360.

Feature-Identification and Data-Compression SoftwareNASA’s Jet Propulsion Laboratory, Pasadena, California


Books & Reports

Alternative Attitude Com-manding and Control forPrecise Spacecraft Landing

A report proposes an alternativemethod of control for precision landingon a remote planet. In the traditionalmethod, the attitude of a spacecraft is re-quired to track a commanded transla-tional acceleration vector, which is gen-erated at each time step by solving atwo-point boundary value problem. Norequirement of continuity is imposed onthe acceleration. The translational accel-eration does not necessarily varysmoothly. Tracking of a non-smooth ac-celeration causes the vehicle attitude toexhibit undesirable transients and poorpointing stability behavior. In the alter-native method, the two-point boundaryvalue problem is not solved at each timestep. A smooth reference position pro-file is computed. The profile is recom-puted only when the control errors getsufficiently large. The nominal attitudeis still required to track the smooth ref-erence acceleration command. A steer-ing logic is proposed that controls theposition and velocity errors about thereference profile by perturbing the atti-tude slightly about the nominal attitude.The overall pointing behavior is there-fore smooth, greatly reducing the de-gree of pointing instability.

This work was done by Gurkirpal Singh ofCaltech for NASA’s Jet Propulsion Labo-ratory. Further information is contained ina TSP (see page 1).NPO-40585

Inspecting Friction Stir Welding Using Electromagnetic Probes

A report describes the use of ad-vanced electromagnetic probes to mea-sure the dimensions, the spatial distribu-tion of electrical conductivity, andrelated other properties of friction stirwelds (FSWs) between parts made of thesame or different aluminum alloy(s).The probes are of the type described in“Advanced Electromagnetic Probes forCharacterizing Materials” (GSC-13878),NASA Tech Briefs, Vol. 21, No. 11 (No-vember 1997), page 4a. To recapitulate:A probe of this type is essentially aneddy-current probe that includes a pri-mary (driver) winding that meandersand multiple secondary (sensing) wind-ings that meander along the primarywinding. Electrical conductivity is com-monly used as a measure of heat treat-ment and tempering of aluminum al-loys, but prior to the development ofthese probes, the inadequate sensitivityand limited accuracy of electrical-con-ductivity probes precluded such use onFSWs between different aluminum al-loys, and the resolution of those probeswas inadequate for measurement of FSWdimensions with positions and metallur-gical properties. In contrast, the presentprobes afford adequate accuracy andspatial resolution for the purposes ofmeasuring the dimensions of FSW weldsand correlating spatially varying electri-cal conductivities with metallurgicalproperties, including surface defects.

This work was done by David G. Kinchenof Lockheed Martin Corp. for MarshallSpace Flight Center. For further informa-tion contact Gary Willett at (504) 257-4786.

Title to this invention has been waivedunder the provisions of the National Aero-nautics and Space Act [42 USC 2457 (f)] toLockheed Martin Space Systems Company —Michoud Operations. Inquiries concerning li-censes for its commercial development shouldbe addressed to:

Lockheed Martin Michoud Space SystemsP.O. Box 29304New Orleans, LA 70189.Refer to MFS-31979, volume and number


Helicity in SupercriticalO2/H2 and C7H16/N2Mixing Layers

This report describes a study of data-bases produced by direct numerical sim-ulation of mixing layers developing be-tween opposing flows of two fluids undersupercritical conditions, the purpose ofthe study being to elucidate chemical-species-specific aspects of turbulence,with emphasis on helicity. The simula-tions were performed for two differentfluid pairs —O2/H2 and C7H16/N2 — atsimilar values of reduced pressure.

This work was done by Nora Okong’o andJosette Bellan of Caltech for NASA’s JetPropulsion Laboratory. Further informa-tion is contained in a TSP (see page 1).NPO-30894

National Aeronautics andSpace Administration

Technology Focus Computers/Electronics - NASA · 2013-04-10 · Technology Focus Computers/Electronics Software Materials Mechanics Machinery/Automation Manufacturing ... High-Rate

Documents