TD_​POWERSPEC

The Power Spectrum function is another function which converts a series from the time or spatial domain over to the frequency domain in order to facilitate Frequency Domain Analysis. Once in the frequency domain, it makes it easier for the data scientist to identify periodicities that are inherent in the series, thereby enabling the data scientist to then eliminate them from the series.

Options

ALGORITHM
ALGORITHM('AUTOCOV') - Use the Fourier Cosine of the Auto covariance approach to calculate power spectrumALGORITHM('AUTOCORR') - Use the Fourier Cosine of the Auto correlation approach to calculate power spectrumALGORITHM('FOURIER') - Use the Fourier Transform approach to calculate power spectrumALGORITHM('INCRFOURIER') - Use the Incremental Fourier Transform approach to calculate power spectrum
FREQ_STYLE
FREQ_STYLE (FK) - The ROW_AXIS scale if frequency: fk = k/N = wk/2π = 1/P FREQ_STYLE (WK) - The ROW_AXIS scale if angular frequency: wk = 2πk/N = 2πfk = 2π/P FREQ_STYLE (PERIODICITY) - Periodicity = P = 1/fk = 2π/wk = N/k
HERTZ_SAMPLE_RATE
Floating point constant representing the sample rate, in hertz. A value of 10000.0 means that the sample points were obtained by sampling at a rate of 10,000 hertz.
INCRFOURIER_PARAM.INTERVAL_LENGTH
INCRFOURIER_PARAM.SPACING_INTERVAL
Output Schema
Output Schema, if Volatile is true then use user login as the schema.
Output Table
Output Table
VAL Location
VAL Location
Volatile
Specifies whether the table should be a VOLATILE table. If true, then the table is automatically deleted, otherwise it is users responsibility to remove or clean it up for space.
WINDOW.WINDOW_NAME
WINDOW('NONE') - Do not apply a smoothing window. Theoretically this translates into the application of a square wave window, which has a magnitude of '1.0' for the whole duration of the window.WINDOW_NAME('TUKEY') - Apply a Tukey smoothing window with the supplied alpha valueWINDOW_NAME ('BARTLETT') - Apply a Bartlett smoothing windowWINDOW_NAME ('PARZEN') - Apply a Parzen smoothing windowWINDOW_NAME ('WELCH') - Apply a Welch smoothing window
WINDOW.WINDOW_PARAM
Optional Parameter. Only valid with a FREQ_STYLE(K_HERTZ). A Floating point constant representing the sample rate, in hertz. A value of 10000.0 means that the sample points were obtained by sampling at a rate of 10,000 hertz.
ZERO_PADDING_OK
Determines whether it is OK to add zeros to the end of a given Series to achieve a more efficient computation of the FFT coefficients. A value of '1' means true; a value of '0' means false. Default is true.

Input Ports

Icon
Connection to a Teradata Database Instance
Icon
At its core, this function takes a logical-runtime series as an input and transforms the series from the time or spatial domain to the temporal frequency or spatial frequency domain, respectively. Like TD_LINESPEC, the input series can contain either real or complex numbers as array elements, aka, PAYLOAD(CONTENT(REAL/MULTIVAR_REAL/COMPLEX/MULTIVAR_COMPLEX)).

Output Ports

Icon
output of TD_POWERSPEC

Nodes

Extensions

Links