New Technique Side-steps Shot Noise for Precision Optics in a Photonic IC
A new technique eliminates the limitations of shot noise on laser interferometry.
When it comes to applications requiring very precise measurements, such as detecting gravitational waves or environmental sensing, we need specialized measuring techniques. One of the most accurate methods of measurement available to us is known as optical interferometry.
While highly precise, optical interferometry comes with limitations. Specifically, the maximum achievable sensitivity of this technique is limited by a naturally occurring phenomenon called shot noise.
Seeking to solve this issue, researchers at the University of Rochester recently published a paper that introduces a new way to achieve increased sensitivity of optical interferometry on an integrated circuit without being limited by shot noise.
An image of the photonic IC created by the University of Rochester. Image used courtesy of the University of Rochester and J. Adam Fenster
This article will look at the underlying concepts, like optical interferometry and shot noise, before diving into the techniques discussed in the new paper.
Brief Overview of Optical Interferometry
Optical interferometry is a precision measurement technique widely used in scientific applications to achieve very granular measurements.
The basic setup of an interferometer. Image used courtesy of Renishaw
These tools work by merging at least two light sources to create an interference pattern.
Since the wavelength of the light is so small, users can detect very minute differences in the distance traveled by each light source in the interference pattern.
In this way, interferometers can make very small measurements, detecting distances as small as 1/10,000 the width of a proton.
The Challenge of Shot Noise
As mentioned earlier, shot noise, in general, is a form of noise that occurs due to the discrete nature of charge carriers and photons.
Light and electrical current is defined by the discrete movement of particles, photons, and electrons. Due to their discrete and random nature, if you measured the number of photons or electrons in a given area, that measurement would never be the same for multiple trials.
This variation of charge carriers, and its impact on the signal, is known as shot noise and is modeled by a Poisson distribution.
Shot noise is modeled as a Poisson distribution, as shown in the graph above. Image used courtesy of Hamamatsu
Generally, the effects of shot noise are negligible because the amplitude of most signals is so large that these small variations in charge carriers are of no consequence.
However, when working at ultra-small amplitudes, like in the case of interferometry, shot noise can become the limiting factor on sensitivity.
Side-stepping Shot Noise With Weak Value Amplification
In their paper published in Nature, researchers from the University of Rochester have developed a new technique to improve precision without being limited by shot noise.
The new technique is based on a concept called "weak value amplification," which exploits the quantum mechanics of light to direct only certain, information-containing photons to a detector instead of sending all of the photons.
In this way, weak value amplification can amplify a signal without adding noise (i.e., increase its signal-to-noise ratio), sending only the important photons while discarding the rest.
A schematic of the weak value amplification device. Image used courtesy of the Song et al
By leveraging this concept, the researchers can achieve the same interferometric signal while utilizing less light. This ability effectively creates more headway for traditional signal amplification on top of weak value amplification, ultimately creating more precise measurements.
All in all, the researchers successfully implemented the technique on a 2 mm x 2 mm integrated photonic chip, achieving a 7 dB improvement in device sensitivity compared to traditional methods.
In 2015, the Laser Interferometer Gravitational-Wave Observatory (LIGO) used optical interferometry to detect the presence of gravitational waves for the very first time, thus showing the value that optical interferometry can present in the scientific world.
With this new technique from the University of Rochester, scientists may be poised to make even more incredible, previously impossible, discoveries in the future.
Interested in other photonics news? Find out more in the articles down below.
Integrated Laser-on-Silicon Photonics Gets a Boost from DARPA
Silicon Photonics May Remedy the Interconnect Bottlenecks of Moore’s Law
Purdue’s Magnetic-free Optical Isolator Aims to Push Photonic ICs Forward
I think the PR description invoking quantum physics makes the intuitive understanding difficult for most EEs.
Basically, the approach is to frequency (wavelength) filter the combined signal plus “other” wavelengths, eliminating “other” wavelengths before detection (light detector to electrical, whatever). Without wavelength filtering, the signal/shot noise ratio is: S/squareroot (S+B) where S is desired signal and B is background. With perfect filtering, then after filtering, the signal/shot noise is : S/squareroot (S), so better signal/shot. While filtering is not perfect, the background cold be quite large, so practically could be quite helpful. The MachZender optics is common in other applications.. (Not sure, but on it may require some acquisition time to tune the MachZender interferometer to the signal wavelength.)