Mind Peak WaveRider Bedienerhandbuch PDF herunterladen (Seite 52)

WaveRider Operating Manual

Transform controls the characteristics of the global FFT parameters used in WaveWare.

Select TRANSFORM... from the OPTIONS menu in WaveWare. Windowing refers to a set

of co-efficients by which the data coming into the FFT are multiplied. The Windows have

been designed to optimize time-frequency resolution trade-offs. The logarithmicity factor

changes the scaling of the output of the FFT. A higher number will increase the apparent

power of lower powered frequency bands in relation to the higher powered frequencies.

Windowing

Each time the frequency transform is computed, it is calculated over a set of 128 consecutive

data points; this set of points is referred to as a Window. It is often desirable to attenuate the

data points near the window ends.

The Rectangular window uses the data points as they are, without attenuation; this option

requires the least computation, thus reducing system loading and speeding response slightly.

The Hamming, Hanning, Triangular, and Blackman windowing options represent various

mathematical approaches to the attenuation of these end points. They offer different trade-

offs between sensitivity of the transform, and reduction of harmonics.

Log Power

After the frequency transform is computed, the power is calculated for each frequency band

from 1 to 40 Hz. When this power spectrum is plotted, for example in the fast bars window,

the power may be mapped linearly into the length of bars (Log Power = 0), or the power may

be mapped logarithmically into the length of bars. Selecting greater Log Power numbers

serves to accentuate the low-power frequency bands, making them more visible relative to the

higher-power bands. Looking at Brain data, this markedly improves the visibility of the Beta

region.

1 2 ... 47 48 49 50 51 52 53 54 55 56 57 ... 156 157

Keine Kommentare

Mind Peak WaveRider Bedienerhandbuch Seite 52