Sound filter shaves microseconds from photonic processing

Chip noise a nuisance nano more, can increase optical processing speeds say Aussie researchers

At the business ends of the extensive web of fibre optic cables that thread around the world, the pulses of light they carry have to be converted into electronic signals.

This is done using laser-based ‘local oscillator’ systems and complex digital signal processing, which essentially unpack the information sent by photons (the elementary particles of light) and translate it into electronic information computers can use.

It all happens in far less time than a blink of an eye. But when it comes to financial trading systems, the Internet of Things, mobile phone networks and warehouse-scale data centres every (micro)second counts.

This week, researchers from the University of Sydney, Monash University and Australian National University, have published a paper on a new technique that does a quicker job than the laser oscillators – using sound.

“The fact that this system is lower in complexity and includes extraction speedup means it has huge potential benefit in a wide range of local and access systems,” said the paper’s co-author and director of the University of Sydney Nano Institute, Professor Ben Eggleton.

In the paper – Chip-based Brillouin processing for carrier recovery in self-coherent optical communications, published yesterday in Optica – the researchers describe their method of processing a photonic signal in a filter built into a chip.

The filter is made from a glass known as chalcogenide, a thin film of which is used on rewritable optical discs like CD-RWs and certain random-access memory devices such as PRAM.

The material has special, acoustic properties that allows it to ‘capture’ the incoming information – with a technique called large-gain stimulated Brillouin scattering (SBS) – and transport it on the chip to be processed into electronic information.

The result is quicker and has more capacity than the usual method, removing the need for complicated laser oscillators and digital signal processing.

“This will increase processing speed by microseconds, reducing latency or what is referred to as ‘lag’ in the gaming community,” said lead-author Dr Amol Choudhary from the University of Sydney.

“While this doesn’t sound a lot, it will make a huge difference in high-speed services, such as the financial sector and emerging e-health applications,” he said.

Dr Amol Choudhary and Professor Ben Eggleton
Dr Amol Choudhary and Professor Ben Eggleton

The researchers also hinted at their technique’s potential for quantum-state measurements.

“Our demonstration device using stimulated Brillouin scattering has produced a record-breaking narrowband of about 265 megahertz bandwidth for carrier signal extraction and regeneration. This narrow bandwidth increases the overall spectral efficiency and therefore overall capacity of the system,” Choudhary added.

The team is now preparing to prototype receiver chips for further testing.

The work follows a demonstration last year from researchers at the ARC Centre of Excellence for Ultrahigh bandwidth Devices for Optical Systems (CUDOS), based at the Sydney Nanoscience Hub, in which photonic data on a microchip was slowed down for processing by storing it as sound.

“It is like the difference between thunder and lightning,” researchers said at the time.

Join the newsletter!


Sign up to gain exclusive access to email subscriptions, event invitations, competitions, giveaways, and much more.

Membership is free, and your security and privacy remain protected. View our privacy policy before signing up.

Error: Please check your email address.

Tags nanotechnologyMonash Universityphotonicschipuniversity of sydneynanoAustralian National Universitysoundoptical computingfibre-opticSydney Nano Instituteacoustics

More about ARCAustralian National UniversityMonash UniversitySBSUniversity of Sydney

Show Comments