Full Waveform Upgrade

The discrete system is the same as before. The ASL50 has been upgraded to include a new signal splitter and two full waveform digitiser boards (A and B). The signal goes through the AGC and is then split to full waveform digitiser (FWD) and to the discrete system. AGC does not change through one wave - the same intensity gain is used. The signal for the FWD goes through an amplifier.

The FWD starts recording when triggered by first return being detected. There is a buffer zone prior to the first return measured. Default is set to 5 meters, but can be changed/specified in flight planning (in winter, first return may be from tree trunk but data from branches/top of tree may also be required).

Maximum acquisition of 120Hz. If it is run at a higher rate then only every second pulse is recorded (e.g running at 150Hz is effectively running at 75Hz). Need to optimise height so capturing at close to, but not above, 120Hz.

Sample rate is chosen at planning stage in FCMS, but can be changed during the flight.

Intensity is digitised at either 1 or 2 nano seconds with either 256, 128 or 64 samples recorded.
1ns represents 15cm/256 samples --> max. collection of 38.4 meters (256*15cm).

Sample depth /interval Equivalent tree height
256 @ 1 ns 38.4m
128 @ 1 ns 19.2m
256 @ 2 ns 76.8m
128 @ 2 ns 38.4m
64 @ 2 ns 19.2m

If it is known that the height is under 19 meters, then can change to a lower sampling rate (128@1ns or 256@2ns).
If flying over flat(ish) terrain, then can reduce to a lower sample depth without loss of useful information.

Two separate outputs:
Standard discrete .SCN files in the RawLaser folder
Full waveform .LWV files in the RawWfd folder

There are two sampling boards, A & B. In the raw_wfd folder the data should be in pairs with a file from each board (FWDxxx01A.LWV & FWDxxx01B.LWV) which are combined during processing. Max. file size of 62501KB. ALSPP gives an error message if it is missing any files.

There is a "constant" offset between the discrete and the FWD which needs to be matched during calibration. This offset is referred to as the trigger delay. This offset has been seen to vary temperature, but not enough info available yet (+/- 100 picoseconds variation within one flightline). Leica have suggested that the discrete may be "more correct" as it has long term stability.

Old System:
Weak signals not recorded.
Cannot distinguish two peaks at less then 4m separation.

With FW:
Cleaner peaks at nadir
Peak signal received is from the ground so can be used to measure the height of a tree. With the discrete do not know true ground position as returns could be from ground, branch, treetop etc.

To Process the FW Data:

Only the raw_laser data needs to be loaded into ALSPP. The raw FW data needs to be in the correct file structure.
Check the box for processing FWD and enter the trigger delay value (calculated during calibration).
Produces LAS1.3 (for full waveform).

Can still output LAS 1.0 for discrete data.
Under Output Options, select LAS Ver 1.0 and ensure that the check box "LAS File with time stamp" is selected and "process FWD" check box in unselected. If the check box "LAS file with time stamp" is not selected ALSPP will happily process the data but not output a .LAS file at the end!

Can identify whether a LAS file is full waveform or discrete by the appended number at the end of the .LAS file
xxx_1.LAS - discrete
xxx_4.LAS - full waveform

There is no information given in ALSPP to indicate the quality/goodness of the FW data - just the discrete return statistics (% for each return).

Post processor speed and volume of disk space significantly increased.

In order to view the FW LAS files in TerraScan, need to export the trajectory file from ALSPP. Utilities -> Generate trj files

Note:
There is a change to the GPS lever arm from FCMS 3.15 onwards. Now refers to a 'PAV80' (stabilied mount) rotation center . Post processor knows when to apply the new GPS lever arm offset and when to use the old one. It uses information from the .sup file to check which version of FCMS was used to create the file. No change to the processing.

Wave Viewer

New wave viewer incorporated into ALSPP. Utilities -> Load Wave Viewer.

Use the scroll bar at the bottom to scroll through each wave. If data was collected at over 120hz there will be a discrete point on every image but a FW wave only on every second image.

The right hand side of the screen displays information for the first discrete return.
Start/end time under ALSPP main window should be the same as the viewer (currently showing different values!).

Can view the data by sample number/time/range on x-axis, raw counts/volts on the y-axis. Change the axes by selecting Options and choosing the values to display.
The viewer itself does not specify which options have been selected for the axes, just gives the values.
Series 0 refers to FW Series 1 refers to discrete

Intensity values for FWD and discrete are on different scales => cannot directly compare/match intensity values (this may cause a problem when creating new discrete points from the full waveform data)

There is also a LasHistoViewer which displays the statistics for the LAS file. Can input the .csv log file to show the flight information along side the statistics.

Calibration

To calibrate the FW, need to calculate the trigger delay.

  • Can use LasHistoViewer to verify that the LAS file is suitable for calibration purposes.
    • check the AGC activity is not jumping around or saturated - value of ~150
  • 1 volt is good target for timing offset.
  • Intensity ~ 60-80
  • Waves where there is one discrete return only

Open the Wave Viewer and display by time and volts. The peak of the FW signal and the discrete point should coincide. Zoom in and manually calculate the difference in time between the peak of the FW and the first discrete return. Under Options, there is a box where the trigger delay can be entered. This can be used to assist in visualising the calibration but does not affect the processing. Repeat until an accurate result is obtained.

Check for several waves over varying parts of the flightline and calculate the average value.

Above and below TPR (transition pulse rate) are independent and calculated separately.
Below TPR -> 9 nanosecond pulse width
Above TPR -> 4 nanosecond pulse width

Requested Leica to include pulse width on the Wave Viewer to avoid confusion over which value (above or below TPR) has been calculated.

Viewing FW Data in TerraScan:

Load flightlines into TerraScan and import the relevant trajectory files. This is done using Manage Trajectories - the icon with three angled vertical lines on the General tool bar (the same tool bar as Settings). File -> Import
On the main TerraScan window, select Flightline -> Deduce using time. This loads the GPS time for each point.
Alternatively can use the .sol file instead of the .trj files, but this uses unnecessary memory as it contains the between flightline data.

Can view the waveform relevant to a single point by drawing in 3D. This adds a coloured ray onto the main view window, with red representing high intensity values and blue representing low intensity values.

Select a discrete point to view the waveform profile. Can change the scales of the axes by varying maximum value (x axis) or sample height (y axis). The horizontal lines on the Waveform window represent the discrete returns.

From the Waveform window, can save the waveform data as a text file. Gives information on position, scan angle and a list of intensities. Produces an individual file per wave if there is only one discrete return associated with the wave. Multiple files are produced if there is more that one discrete return - files are appended with _only, _first, _last.

Can create more information by extracting echoes and creating new discrete points from the FW data. Manipulate the extracted data by adjusting the threshold. These new points can be added to the discrete LAS file. Need to check what information is stored for these new points (scan angle, return number etc...). TerraScan has not quite figured it out yet as it places new points below the ground.

Trj. files stay in the TerraScan - could not find a way to remove them without deleting the trj. file. The 3D rays also stay in TerraScan unless they are manually deleted. May be necessary to have a per project design file.

Useful Bits of Information:

  • Can set microstation so that it automatically loads terrascan when it is opened.
  • When processing large data sets through ALSPP:
    • can cap the size of the output LAS files to a certain number of points.
    • when inputting the raw_laser files can set it to process a subset of .scn files rather than the entire flight line

Last modified 7 years ago Last modified on Feb 23, 2010 9:29:46 AM