Averages; Minimum Loss In Pdl Mode - JDS Uniphase SWS15100 User Manual

5. Compute the bandwidth of the device, as the difference between the two wavelengths
- λ
(λ ).
R L
Finding Bandwidth using a Threshold Relative to the Peak
The algorithm used by the SWS to find the bandwidth is:
1. Find the specified peak signal level (the peak). Store the wavelength at which this occurs.
2. At a power level x dB from the peak found in Step 1, find the shorter wavelength (λ
to the signal. The wavelength is determined by finding two data points, one to the left of the
wavelength at which the signal crosses x dB, and one to the right of that wavelength. The
wavelength (λ ) corresponding to x dB is interpolated from these two data points. The search
L
is begun, starting outward (left) from the peak position.
3. At the same level (x dB from the peak), find the longer wavelength (λ
as in Step 2. The search is begun, moving outward (right) from the peak position.
4. Compute the bandwidth of the device, as the difference between the two wavelengths
- λ
(λ ).
R L

Averages

Two types of averaging are implemented. The first type is averaging over a number of traces,
which is done in single scan mode. In this form of averaging, n successive scans are added
together and the sum is divided by n. The addition and averaging are performed on the insertion
loss in linear scale, not in dB. At the end of each scan, the current average is converted to dB
and displayed.
The second method is weighted averaging used in continuous scan mode. Weighted averaging
is performed using the following algorithm, where w (0.0 < w < 1.0) is the chosen weight:
Average[pass_1] = Reading[pass1]
Average[pass_2] = ( w × Average[pass_1] + (1-w) × Reading[pass_2] )
Average[pass_3] = ( w × Average[pass_2] + (1-w) × Reading[pass_3] )
Average[pass_4] = ( w × Average[pass_3] + (1-w) × Reading[pass_4] )
Or, generally:
Average[pass_n] = [ w × Average[pass_(n-1)] + (1-w) × Reading[pass_n] ]
In this example, "Reading" are measured insertion losses in linear power scale. At the end of
each scan, the running average is converted to dB and displayed.

Minimum Loss in PDL Mode

In PDL mode, the minimum insertion loss is computed at each wavelength, without respect to
the polarization. To compute this, the Mueller matrix is computed from the insertion loss
measurement at the four linearly independent polarization states produced by the four-state
polarization controller. From the Mueller matrix, the polarization vector that produces the
Artisan Technology Group - Quality Instrumentation ... Guaranteed | (888) 88-SOURCE | www.artisantg.com
) nearest
L
) nearest to the signal,
R
Reference – 223