Method of analyzing dynamic data characterized by a time-varying frequency spectrum
Computer tape recording of dynamic data characterized by time varying frequency spectrum
SEARCH · Engineering Papers
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.
Computer tape recording of dynamic data characterized by time varying frequency spectrum
The hybrid optimization techique for attitude control design is described in detail. The technque is capable of optimizing an n-dimensional adjustable parameter vector, but a 1-dimensional vector is used as an example to explain the procedure. This permits an easier explanation as opposed to a multidimensional case. The procedure used to derive the perturbation equations of motion describing the 6-DOF shuttle ascent phase is presented. These equations were programmed on the EAI 8800 analog computer to describe the perturbations of the shuttle vehicle from a nominal zero lift trajectory due to wind disturbances. Included are the control system equations, trim equations, and wind angle of attack equations. Analog wiring diagrams, raw data and time-varying coefficients, and state variable responses during shuttle ascent are also considered.
This paper presents an adaptive estimation algorithm for bit synchronization, assuming that the parameters of the incoming data process are time-varying. Experiment results have proved that this synchronizer is workable either judged by the amount of data required or the speed of convergence.
Analytic techniques aid in the design of nearly optimal linear time-varying sampled-data stochastic controllers. The techniques also aid in the simplification and automation of program designing for control computers.
Noncoherent detection of telemetry data transmitted through randomly time varying channel and subject to additive Gaussian noise, deriving error probability
Data compression for telemetry transmission with sample choice based on time-varying stream slope thresholds illustrated for Ranger photograph
Statistical filter theory is employed to develop a method for determining the best possible estimate of the position and velocity of a space vehicle in the midcourse phase of flight. Results of a computer simulation are given to illustrate the performance attainable. An on-board system is visualized in which the source of information is an arbitrary sequence of observations of space angles, corrupted by measurement errors. The scheme is in effect a dynamical time-varying filter, implemented by a digital computer, which processes the incoming data to compute an up-to-date optimal estimate of position and velocity.
The development of the test/analysis technique known as the impulse transfer function (ITF) method is discussed. This technique, when implemented with proper data processing systems, should become a valuable supplement to conventional dynamic testing and analysis procedures that will be used in the space shuttle development program. The method can relieve many of the problems associated with extensive and costly testing of the shuttle for transient loading conditions. In addition, the time history information derived from impulse testing has the potential for being used to determine modal data for the structure under investigation. The technique could be very useful in determining the time-varying modal characteristics of structures subjected to thermal transients, where conventional mode surveys are difficult to perform.
The large quantity of Mariner 9 television pictures taken at phase angles ranging up to twice those accessible from earth has been used to describe integrated photometric properties of Mars. Although frame-to-frame variations emphasize the shortcomings of vidicon cameras when used in a photometric mode, statistical trends have yielded data comparable with earth-based observations. Analysis of atmospheric parameters over a period of changing opacity - November 1971 through March 1972 - has provided time-varying optical depths and light-scattering information. Linear function characterization of the Mars reflectance is clearly discrepant at large incidence or emission angles and at phase angles greater than 40 deg.
The vacuum transient response of the moon to a time-varying spatially uniform magnetic field is determined for a lunar electrical conductivity model that was based on the harmonic analysis of Apollo 12 and Explorer 35 dayside magnetometer data. The transient response of the model is found to provide a plausible explanation of the behavior of the local vertical-surface magnetic field for an Apollo 12 magnetometer darkside transient event. A model containing a conducting core and a highly conducting thin subsurface layer is presented, and its transient behavior is discussed.
The problem of real-time estimation of a lifting reentry vehicle trajectory of the shuttle orbiter type is considered. Simulations feature large position and velocity uncertainties at radar acquisition and realistic model errors in lift, drag and other model parameters. Radar tracking and accelerometer data are simulated. Significant nonlinearities are found to exist on spacecraft acquisition. An iterated nonlinear filter is shown to perform optimally during the radar acquisition phase. An adaptive filter is shown to track time-varying model errors, such as errors in the lift and drag coefficients, down to the noise level. Such real-time model tracking (identification) is frequently required for guidance and control implementation.
A real-time, computer-based, phase vector display system has been developed which will output a vector whose phase is equal to the delay between a trigger and the peak of a function which is quasi-coherent with respect to the trigger. The system also contains a sliding averager which enables the operator to average successive trials before calculating the phase vector. Data collection, averaging and display generation are performed on a LINC-8 computer. Output displays appear on several X-Y CRT display units and on a kymograph camera/oscilloscope unit which is used to generate photographs of time-varying phase vectors or contourograms of time-varying averages of input functions.
Discussed are the mean-square response exceedance characteristics of a single-tuned system to amplitude modulated noise. The results bear on the accuracy of spectral estimates of nonstationary data, and subsequently, relate directly to the design, analysis, and testing of structural systems in environments as gusts, earthquakes, and ignition transients. For noise correlated as an exponentially damped cosine, the nonstationary response may exceed its stationary value by a factor in excess of two. A time-varying shaping filter explanation is offered for this behavior. For white noise, such exceedances do not occur.
The use of a Monte Carlo model for generating sample directional reflectance data for two simplified target canopies at two different solar positions is reported. Successive iterations through the model permit the calculation of a mean vector and covariance matrix for canopy reflectance for varied sensor view angles. These data may then be used to calculate the divergence between the target distributions for various wavelength combinations and for these view angles. Results of a feature selection analysis indicate that different sets of wavelengths are optimum for target discrimination depending on sensor view angle and that the targets may be more easily discriminated for some scan angles than others. The time-varying behavior of these results is also pointed out.
A number of conductivity models were investigated for compatibility with Apollo 12 magnetometer data. Except at the highest frequencies, a simple core-crust model is compatible with the observed dayside transfer function, which is expressed as the ratio of the lunar surface field spectrum to the interplanetary magnetic field spectrum. All conductivity profiles exhibit a peak near 1500 km, when the models are constrained to conform to the observed flat response at the higher frequencies. However, at frequencies above .01 Hz the long wavelength limitation of the theoretical model is no longer valid. A plausible explanation for the difference between the north-south and east-west transfer functions is that it is due to a time-varying compression of the remanent (dc) field at the Apollo 12 site by fluctuations in the solar wind plasma.
Randomly time variant linear communication channels, discussing signal measurement and data processing problems
Mathematical model for predicting lunar response to time varying interplanetary magnetic fields and its applicability to interpretation of ALSEP magnetometer data
The capabilities of the Marshall system for aerospace system simulation (MARSYAS) and how to use it are described. MARSYAS is a software system that allows easy setup and control of the simulation of the dynamics of large physical systems on a digital computer. The physical systems are modeled in the form of block diagrams or equations. The blocks can have multiple inputs and multiple outputs, and they can be nested to form hierarchies. The block diagrams can contain transfer functions, nonlinear and logical functions, equations, analog computer elements and FORTRAN programs. The input format of the equations can be combinations of nonlinear, time-varying differential equations and algebraic equations in their original format. MARSYAS could also serve as a storage and retrieval system for models as a basis for a model configuration control system on a central time-shared computer. The outputs of the simulation system can be not only time-responses but also other analysis data such as frequency response, power spectrum and stability parameters. The MARSYAS translator is written in FORTRAN running on the Univac 1108 computer under the EXEC 8 operating system.