The world’s first self-calibrated photonic chip: an interchange for optical data highways

Research led by Monash and RMIT Universities in Melbourne has found a way to create an advanced photonic integrated circuit that builds bridges between data highways, revolutionizes the connectivity of current optical chips and replaces bulky 3D optics with a wafer-thin slice silicon.

This development, published in the magazine Nature photonicshas the ability to accelerate the global advancement of artificial intelligence and offers key real-world applications such as:

• Safer driverless cars that can immediately interpret their surroundings
• Enabling AI to diagnose medical conditions faster
• Make natural language processing even faster for apps like Google Homes, Alexa and Siri
• Smaller switches to reconfigure optical networks that carry our internet to get data where it’s needed faster

Whether it’s turning on a TV or keeping a satellite on course, photonics (the science of light) is changing the way we live. The photonic chips can convert the processing power of bulky utilities the size of a bank into chips the size of a fingernail.

dr. Mike Xu of the Electrical and Computer Systems Engineering Department of Monash University and now Beijing University of Posts and Telecommunications, Professor Arthur Lowery of the Electrical and Computer Systems Engineering Department of Monash University, and Dr. Andy Boes, who conducted this research during his RMIT.

Professor Arnan Mitchell and Dr. Guanghui Ren developed the chip so that it was ready for the experimental demonstration.

The project’s lead researcher, Monash University ARC Laureate Fellow Professor Arthur Lowery, says this breakthrough complements Dr. Bill Corcoran of Monash University, who, in collaboration with RMIT, developed a new optical microcomb chip in 2020 that can squeeze three times the traffic of the entire NBN through a single optical fiberconsidered the world’s fastest internet speed from a single chip the size of a fingernail.

The optical microcomb chip built multiple lanes of the highway; now the self-calibrating chip has created the ramps and bridges that connect them all and allow greater movement of data.

“We demonstrated a self-calibrating programmable photonic filter chip, with a signal processing core and an integrated reference path for self calibration‘ explains Professor Lowery.

“Self-calibration is important because it makes tunable photonic integrated circuits useful in the real world; applications include optical communication systems that switch signals to destinations based on their color, very fast similarity calculations (correlators), scientific instruments for chemical or biological analysis, and even astronomy.

“Electronics saw similar improvements in radio filter stability using digital techniques, allowing many cell phones to share the same stretch of spectrum; our optical chips have similar architectures but can operate on signals with terahertz bandwidths.”

It took three years to work on this breakthrough.

New internet-dependent technologies such as self-driving cars, remote-controlled mining and medical equipment will require even faster and greater bandwidth in the future. Bandwidth enhancement isn’t just about improving the optical fibers that our Internet travels through, it’s about providing compact switches of many colors, going in many directions, so that data can be sent along many channels at once.

“This research is a major breakthrough – our photonic technology is now sufficiently advanced that really complex systems can be integrated on a single chip. The idea that a device can have an on-chip reference system that allows all components to work as one is a technological one. breakthrough that will enable us to tackle internet bottlenecks by rapidly reconfiguring the optical networks that carry our internet to get data where it’s needed most,” said Professor Arnan Mitchell of InPAC.

Photonic circuits are able to manipulate and route optical information channels, but they can also provide some computational power, for example when searching for patterns. The search for patterns is fundamental to many applications: medical diagnosis, autonomous vehicles, internet security, threat identification and search algorithms.

Due to the fast and reliable reprogramming of the chips, new search tasks can be programmed quickly and accurately. However, this fabrication must be accurate to the degree of a small wavelength of light (nanometers), which is currently difficult and extremely expensive – self-calibration solves this problem.

A major challenge of the research was integrating all optical functions into a device that could be “connected” to existing infrastructure.

“Our solution is to calibrate the chips after production, to effectively tune them by using an on-chip reference, rather than using external equipment,” said Professor Lowery, an ARC Laureate Fellow. “We use the beauty of causality, effect after cause, which dictates that the optical delays of the paths through the chip can be uniquely inferred from the intensity versus wavelength, which is much easier to measure than precise time delays. We have a strong reference path to our chip and calibrated. This gives us all the settings needed to ‘dial in’ and the desired switching function or spectral response.”

The method is a crucial step to make photonic chips practically usable. Instead of looking for a setting, similar to to coordinate in an old radio, the researchers were able to tune the chip in one step, allowing data streams to be switched quickly and reliably from one destination to another.

Reliable tuning of photonic chips opens up many other applications, such as optical correlators, which can find data patterns in data streams, such as images, almost instantly — something the group has also been working on.

“As we integrate more and more bank-sized devices into fingernail-sized chips, it’s getting harder and harder to get them all to work together to achieve the speed and function they did when they were bigger. We’re up to this challenge It was committed to creating a chip smart enough to calibrate itself so that all components could run simultaneously at the speed they needed,” said Dr. Andy Boes of the University of Adelaide.