Fill signal loss
Measurement data can contain unusually large time gaps, for example when a recorder temporarily loses connection to a signal source. Use the Fill signal loss operation to detect such gaps and mark them with a single inserted sample.
If the last sample before a detected gap is at time t, the operation inserts one new sample at t + time to detect signal loss, using the configured target value. The existing samples remain unchanged.
|
Note |
|---|
|
This operation does not resample or fill the entire gap. It only inserts one marker sample that indicates when the signal is considered lost. |
-
After adding the Fill signal loss operation, select the signals to check for signal loss in the Signals field.
If no signals are selected, all signals are checked.
-
(Optional) In the Time to detect signal loss input field, enter the duration after which a time gap between consecutive samples is considered signal loss.
If the field is left empty, the threshold is calculated separately for each signal as the mean time between samples plus two times the standard deviation of the time between samples.
-
To select the unit, use the drop-down list.
-
(Optional) In the Target signal loss value input field, enter the numeric value to write to the inserted marker sample.
If the field is left empty, NaN is used.
A signal contains the following samples:
|
time |
value |
|---|---|
|
1 |
0 |
|
2 |
0 |
|
3 |
0 |
|
4 |
0 |
|
5 |
0 |
|
20 |
1 |
|
21 |
1 |
|
22 |
1 |
Set Time to detect signal loss to 1 and leave Target signal loss value empty. Therefore, NaN is used as the target value.
The time gap between the samples at t = 5 and t = 20 exceeds the configured threshold. The operation detects the signal loss and inserts a single sample at t = 6 (5 + 1) with the value NaN.
The resulting signal is:
|
time |
value |
|---|---|
|
1 |
0 |
|
2 |
0 |
|
3 |
0 |
|
4 |
0 |
|
5 |
0 |
|
6 |
NaN |
|
20 |
1 |
|
21 |
1 |
|
22 |
1 |
The existing samples remain unchanged.