The Bridge - Issue 1, 2023 - 9

Ultrafast Lidar Based on Signal Time Stretch and Various Transduction Techniques
Feature
resolution ultrafast optical ranging to extract this decoded
information [5]. Each of these techniques comes with
certain drawbacks. For instance, in an AMI system, the
ranging information is extracted from the phase shift of the
interferometer's reference and object signals, so to
avoid ambiguity problems, it is required to know the
ranging information beforehand with sub-wavelength
accuracy, and SRI needs equalization of the
geometrical path lengths in dispersive elements to
avoid ghost steps in the measurement. Here we will
provide an overview of various techniques used for
" time-stretch Lidar " data acquisition and transduction.
II. Background
In conventional Lidar, narrow-spectrum laser pulses
are transmitted, and the return signal is monitored
for the time-of-flight and amplitude measurements.
The laser pulses are spatially scanned to create a
mapping of the target objects. The wavelength used
is tailored to the intended target. The conventional
approach has limited performance for targets that
are present on short time scales, or which require
fast measurements, especially those with transient
or weak return signals. If a broad-spectrum, ultrafast
laser signal is used, the wavelengths can be spread out
through dispersion to allow interferometric detection of the
wavelength-dependent signal modulation. For instance,
various wavelengths in a broad-spectrum pulse will
travel at different speeds in a dispersive optical fiber. For
spectroscopic applications, the target could be components
of a liquid or gas which will encode the pulse wavelengths.
Time-stretch Lidar requires a broad-spectrum pulse or
time-to-wavelength mapping of a narrow-line ultrafast
laser pulse. Time-to-wavelength mapping is a process that
consists of two steps, including (i) stretching the laser pulse
and (ii) modulating it via beating with a reference signal
through a modulation process, such as in a Mach-Zehnder
interferometer (MZI). Differences arise when the order of
these steps is switched in an experiment. Stretching of a
narrow-line ultrafast pulse train can happen either through
kilometers of fiber providing linear group velocity dispersion
(GVD) such as dispersion compensating fiber (DCF) or
linearly polarized fiber Bragg grating (LC-FBG).
In this section, three time-stretch Lidar systems are
investigated based on the positioning of the dispersive
element and their respective transduction techniques used
to retrieve the information of the device under test. All
these techniques use ultrafast laser pulses and dispersion
elements to generate a temporal and spectral response
that will eventually allow us to resolve tiny changes in the
parameters of interest, including distance, shape, etc. Figure
1 describes the three Lidar systems to detect the small
displacements of a mirror. The optical circulators direct the
incident signal to the targets and then pass the modulated
signal to the detection elements. The optical delay lines
(ODL) and the polarization control match the signal and
reference paths.
Figure 1: (a) Time-stretched Lidar with direct optical transduction, (b) timestretched
Lidar with microwave processing transduction, (c) ultrafast Lidar with
correlated spectro-temporal encoding and wavelength division multiplexing. PD:
photodetector, Pol.: in-line polarizer, IM: intensity modulator, PC: polarization
controller, EDFA: erbium-doped fiber amplifier, (DE)MUX: (de)multiplexer, ODL:
optical delay line, OSC: oscilloscope, OSA: optical spectrum analyzer, BPF:
bandpass filter, CWL: continuous-wave laser.
A. Time-stretched Lidar with direct optical
transduction
Figure 1a shows a real-time ranging Lidar based on direct
optical detection and Fourier transformation post-processing
[1,6]. The ultrafast laser pulse train is divided into the probe
and the reference arms using a fiber splitter to generate
two intrinsically coherent copies of the same laser. The two
signals then recombine through a circulator into a fiberbased
MZI, after which their beat signal (this pattern due
to interference is called an interferogram) is guided into a
DCF to provide multiple orders of magnitude broadening
in the temporal and spectral profile. A tunable optical delay
line is incorporated in the reference arm of the MZI to
allow for beat signal tunability. The beat signal frequency
of the MZI is proportional to the time delay between the
two arms of the MZI, which is the delay introduced by the
tunable optical delay line. A high-precision delayer with a
large overall delay is needed to achieve both high sensitivity
and a large dynamic range, both of which are limited by the
speed of the photodetector and the oscilloscope. Therefore,
for continuous measurement and data processing for high
sensitivity and large dynamic range, a fast oscilloscope or
analog-to-digital conversion is required, which will result in
very large amounts of data being generated and stored.
Using a 50:50 splitter, both the temporal and spectral
interferograms of the signal can be measured
simultaneously to which optical data processing such as
fast Fourier transform will be applied. Examples of the
generated interferograms are provided in the next section,
HKN.ORG
9
https://hkn.ieee.org/
The Bridge - Issue 1, 2023

Table of Contents for the Digital Edition of The Bridge - Issue 1, 2023
