FAQ on the basics of FMCW LiDAR: part 1

This alternative to time-of-flight LiDAR has unique complexities and distinct attributes.

When engineers hear the phrase LiDAR, an acronym for Light Detection and Ranging, they generally think of time-of-flight (ToF) systems. ToF LiDAR is widely used in automotive and robotic applications to “see” surroundings and obstacles.

But there is another, radically different LiDAR approach: frequency modulated continuous wave or FMCW LiDAR. This approach is derived from many of the principles of conventional RF-based radar, but modified extensively to be used with light instead of RF.

“Which is better?” is a contentious question. Practical FMCW systems are relatively newer than ToF LiDAR, but their proponents maintain that their time has come, while adherents of ToF maintain that their approach is superior. As usual, it is not a clear-cut argument, as each has strong and weak points compared to the other.

The right answer depends to some extent on which characteristics are most important to the user in the intended application. In addition, the underlying technologies are rapidly evolving, so the relative merits at any given time are in flux.

This FAQ will not attempt to resolve the “which approach is better where and when?” question. Instead, it will look at what FMCW LiDAR is, the architecture and components it uses, and its performance capabilities.

Q: What is the operating principle of FMCW LiDAR?
A: In FMCW, a laser diode is used to send out a continuous frequency-modulated beam towards the target area, seen in Figure 1. A receiving sensor co-located with the source captures the light-return signal (it’s often just a few photons) and uses the well-known Doppler shift to near-instantly determine where the objects are and how fast they are moving relative to the system.

Figure 1. This conceptual diagram of FMCW LiDAR shows (a) the basic principle; (b) beat frequency of the reference signal (green) and sample signal (blue); (c) beat frequency over time in the different moving cases, where the blue and purple circles represent the measured peak beat frequencies. (Image: ResearchGate)

A small fraction of the carrier is diverted to the receiver channel to enable coherent synchronous demodulation. Again, this is an optical expansion of well-known radar and demodulation principles.

Q: Does FMCW LiDAR create a 3-D image?
A: No, it creates what is called a 4-D image encompassing three-dimensional space along with velocity imaging of the scene.

Q: Why is it considered to be frequency modulated, as there doesn’t seem to be any conventional modulation here?
A: In traditional FM broadcast radio, a continuous signal, such as voice or music, is imposed on the carrier and modulates its frequency. In many radar systems, the frequency of the carrier is instead modulated by a “chirp” (compressed high-intensity radiated pulse), a pulse-compression signal where the frequency increases (up-chirp) or decreases (down-chirp) over the duration of the pulse, shown in Figure 2.

Figure 2. There are many possible “chirp” waveforms used to modulate the carrier, such as the downchirp (left) and triangular chirp (right); the image below each shows the respective impact on the carrier frequency versus chirp timing. (Image: Wireless Pi)

Instead of a single frequency, the chirp sweeps a range of frequencies to improve range resolution and target detection while keeping power usage low. Due to the relatively fast transitions of the modulating signal, it creates a wide range of frequencies as it modulates the carrier.

Q: What about the “continuous wave” aspect? The chirping does not seem continuous.
A: Again, in a parallel to radar and even broadcast FM, the carrier is on all the time and thus continuous. It’s the modulating waveform that is not continuous.

The next part looks at the electronic and optical components needed to build an FMCW LiDAR system.