Skip to Content

A Cheaper Route to Speeding Up the Web

A new silicon-based optical device has the potential to improve the speed, cost, and reach of fiber-optic networks.
March 26, 2007

The fiber-optic networks that zip video clips and other data around the Web rely on devices made of expensive semiconductor materials for many operations, making it difficult to expand bandwidth inexpensively. Now, researchers at Alcatel-Lucent Bell Labs, in Murray Hill, NJ, have designed a crucial optical component–a filter that cleans up signals as they travel through a network–that is made completely of silicon. It performs better than many other filters, and because it’s made of silicon, it can be mass-produced cheaply and integrated easily with electronic systems.

Silicon superfilter: Shown are three optical equalizing filters fabricated entirely in a silicon electronics foundry. These filters, which cover a total area of 3 mm by 10 mm, “clean up” signals as they pass through a fiber optic network.

In addition to improving fiber-optic networks, the silicon filter has the potential to finally bring the speed of photonic data transfer, which is much faster than copper wiring, to computer circuit boards and microprocessors, says Sanjay Patel, technical manager of integrated photonic research at Bell Labs. Indeed, he says, chips that contain both photonic devices, such as the filter, and electronics, could be key components in all optical and cellular communication systems, as well as inside computers.

Over the past couple of years, there have been several key advances in using silicon–the mainstay of electronics–for photonics. Historically, photonic devices–such as lasers and detectors that produce and collect light, modulators that encode bits of data on that light, and equalizers and filters that clean up signals–have been made out of expensive semiconductors such as indium phosphide and gallium arsenide. But in 2005, researchers at Intel announced the first silicon laser, and soon thereafter, a handful of other organizations, including MIT, Cornell, and the University of California, announced other silicon-photonics firsts. (See “Intel’s Breakthrough.”)

Filters are an important addition to the growing family of silicon photonics. When a signal travels through a fiber-optic cable over great distances, it inevitably becomes distorted. Along the way, filters and other devices make the signal cleaner by collecting the light and modifying the waves’ phase (the relative position) and amplitude (its brightness).

Patel explains that the team’s device combines, into one structure, two classical techniques used for making filters. In the researchers’ design, light enters one end of the filter, where it is split into beams. These beams travel through a series of loops–called ring-resonators–where the light’s phase and amplitude are adjusted. Then the beams recombine and the “cleaned-up” signal is sent on its way. The trick to making the device in silicon, Patel says, was to use the type of resonator architecutre that works well even if the rings aren’t perfect. In other words, the filter design is able to withstand the natural variations that occur during silicon device fabracation.

The Bell Labs work is “a great step forward,” says Alan Willner, professor of electrical engineering at the University of Southern California. “It’s not just a filter,” he says. “It’s a superfilter.” The device is one in a set of tools that could increase performance of optical systems over longer distances and allow them to carry more data, Willner adds.

The research will be presented today at the Optical Fiber Conference in Anaheim, CA. Patel says that it may still be three to five years before the filter makes its way into an actual product.

Keep Reading

Most Popular

computation concept
computation concept

How AI is reinventing what computers are

Three key ways artificial intelligence is changing what it means to compute.

still from Embodied Intelligence video
still from Embodied Intelligence video

These weird virtual creatures evolve their bodies to solve problems

They show how intelligence and body plans are closely linked—and could unlock AI for robots.

conceptual illustration showing various women's faces being scanned
conceptual illustration showing various women's faces being scanned

A horrifying new AI app swaps women into porn videos with a click

Deepfake researchers have long feared the day this would arrive.

pig kidney transplant surgery
pig kidney transplant surgery

Surgeons have successfully tested a pig’s kidney in a human patient

The test, in a brain-dead patient, was very short but represents a milestone in the long quest to use animal organs in human transplants.

Stay connected

Illustration by Rose WongIllustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at with a list of newsletters you’d like to receive.