International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXIX-B7, 2012 
XXII ISPRS Congress, 25 August — 01 September 2012, Melbourne, Australia 
LIDAR WAVEFORM SIMULATION OVER COMPLEX TARGETS 
Seongjoon Kim”, Impyeong Lee * *, Mijin Lee * 
* Dept. of Geoinfomatics, The University of Seoul, 90 Jeonnong-dong Dongdaemun-gu Seoul, Korea - 
(sinus7953, iplee, mj-lee)@uos.ac.kr 
Commission III, WG III/2 
KEY WORDS: Simulation, LIDAR, Waveform, sub-beam, geometry, radiometry, Model 
ABSTRACT: 
Unlike discrete-return LIDAR system that measures the return times of some echoes, waveform LIDAR provides the full-waveform 
data by digitising multi-echoes. From the waveform data we can extract more information about the geometry and reflectance of the 
target surfaces by applying various algorithms to interpret waveform. There have been many researches about the laser beam's 
interaction with the illuminated surfaces and the diverse algorithms for waveform processing. The purpose of this paper is to suggest 
the method to simulate waveform coming from complex targets. First, we analysed the previous relevant works. And based on these 
we attempted to generate the simulated waveform over the complex surfaces. For the waveform simulation, we defined the sub- 
beams spread with a consistent interval within the beam's divergence coverage. Each sub-beam has its geometry (origin and 
direction) and the transmitted energy considering the laser beam's profile. Then, we searched the surfaces that intersect with sub- 
beams using ray-tracing algorithm, and computed the intersection points and the received energies. Using on the computed distance, 
the received energy and predefined pulse model, we generated the signals of echoes and put them together into a waveform. Finally, 
we completed the waveform simulation adding the signal noise. As a result of performing waveform simulation, we confirmed that 
the waveform data was successfully simulated by the proposed method. We believe that our method of the waveform simulation will 
be helpful to understand the waveform data and develop the algorithms for the waveform processing. 
1. INTRODUCTION But it takes long time and a tremendous amount cost to acquire 
by real systems. 
A laser scanning system is capable of acquire the 3D point In this paper, we suggest the method to simulate realistic and 
cloud by sampling the surface of objects. In civilian precise waveforms by sensor modelling of full-waveform lidar 
applications, airborne laser scanning systems are widely used systems in geometric and radiometric aspects. 
for various applications such as bare ground modelling, target In order to simulate waveforms of complex surfaces, we 
detection, object reconstruction, forest biomass estimation, adopted sub-beam processing which are defined around the 
corridor mapping and change detection. Besides, because a laser laser beam ray within beam divergence angle. Assuming an 
beam is robust against the obscuration (e.g. smoke and flare) ^ echo is originated from the individual reflected sub-beam, each 
rather than electro optic systems, these systems are also ^ sub-beam is separately processed. First, we computed the 
employed for surveillance and reconnaissance to detect targets intersection points between sub-beams and target surfaces in 
in military applications, owing to the robustness. geometry. In the geometric process, we can obtain the ranges of 
In topographic mapping, lidar systems used typically have a sub-beam ray from the origin of the sensor to the target surfaces. 
single detector and scanning mirror to enlarge the coverage. In radiometric simulation, the return energy of each sub-beam 
They shot a laser pulse with Gaussian shape and measure its with the transmitted energy which is assigned by beam profile is 
TOF (time of flight) by pulse detection. When transmitted laser calculated using laser equation. And then, for each sub-beam, 
pulses encounter with target, there may be lots of scatterings by the return echo is generated using pulse model with the return 
many surfaces within the footprint and they contribute the time and energy. Finally, waveform is simulated by summing 
waveform, even if the beam divergence (0.2~2 m) is very small the individual echoes of sub- beams and noise. 
(Mallet et al, 2009). The traditional discrete return systems (or 
multiple pulse laser scanning systems) output the travel times of 
individual echoes up to six returns (Mallet ez al, 2009). On the 2. SIMULATION METHOD 
other hands, recently advanced systems, called full-waveform 
lidar system, have the capability to record the waveform signals, 2.1 Overview 
as it is, by digitizing with high frequency. These waveforms can 
provide additional information about the structures and the In general, the pulsed lidar system measures ranges by the flight 
physical properties of the surfaces. For example, there are time of the laser pulse. Then 3D points are computed by 
increasingly many early works to analyse vertical structure for georeferencing with the data from GPS, IMU and scanning 
forest biomass or detect concealed targets by interpreting system. Figure 1 illustrates how the system obtains full- 
waveforms. waveform data from a single laser shot. First, a transmitted laser 
To develop and validate the algorithms of processing pulse may encounter with many target surfaces on the footprint 
waveforms, there needs many waveform data of various targets. not on a point due to beam divergence. And then, the 
  
* Corresponding author. 
517