By passing information at rates of a 200Gbit/s via a drone, US researchers have pushed the limits of “twisted light” communications. Alan Willner’s University of Southern California (USC) team achieved this bandwidth by encoding data into light using orbital angular momentum (OAM). This powerful technique lets the scientists combine beams with different OAM modes, each of which can carry data.

This work built on the team’s first 2Tbit/s demonstration using 16 OAM beams with the same wavelength in 2012. From this the USC team moved to 42 wavelengths, each combining 24 OAM beams. “We had around 1000 different beams that were all orthogonal,” Willner says. “We had 100Tbit/s in the lab.” More recently, they have communicated at 400Gbit/s between transmit and receive stations 120 meters apart outdoors. In 2019 they published their work sending data by reflecting the beams off a drone, and separately published an experiment demonstrating a 40Gbit/s link transmitting beams through water. These impressive results suggest great promise in communications and beyond, which scientists discussed in February at Photonics West 2020.

Willner explains that OAM relies on more complex electromagnetic wave properties of light than optical communication does conventionally. In conventional circular Gaussian beams, photons are all “oscillating up and down together,” he explains. “With OAM, the relative phases around your circle are changing, oscillating out of phase with each other.” OAM can be used in communications through the combination of beams so that they form orthogonal channels. “One twists fast, one twists slower, one clockwise, one counterclockwise,” Willner says. “You put them all together and you multiply your capacity.”

That contrasts from how existing optical communication techniques exploit the on-off signalling of time division multiplexing, and divide information into frequency channels. “We’re starting to max out both of those in terms of the amount of information we can include,” explains Martin Lavery from the University of Glasgow, UK. Normally, a laser produces a Gaussian beam – a beam whose magnetic and electric field amplitude profiles look like a bell curve, Lavery says. OAM exploits light in the forms known as Laguerre-Gaussian modes.

nterferograms produced from optical interference between OAM beams with orders +4, +8, -8, +16 and a Gaussian beam reference. The number and direction of petals in the interferogram indicates the absolute value and sign of OAM order, respectively. Credit: Alan Willner/USC.

“These have almost donut-shaped structures,” Lavery says. “They have dark centers and an annulus of bright light.” Lavery then switches metaphors to a spiral staircase, with an empty space running down the middle. In OAM beams, light twists around this empty ‘vortex point’ with orbital momentum, he explains, which changes in discrete amounts, like electronic energy levels. “These things you can use as orthogonal channels because they’re completely separate in terms of the information you want to encode,” Lavery stresses.

Giovanni Milione from NEC Labs in Princeton, US, puts this mathematically. “Predominantly, we exploit light’s OAM modes,” he explains, which comprise a set of solutions to the wave equation in cylindrical coordinates, with cylindrical symmetry. “What makes OAM modes so interesting are the striking physical analogies that naturally follow due to that symmetry,” Milione says.

The approach has clear potential in optical fiber communication, where OAM modes are typically used for mode division multiplexing. This approach is analogous to other well-known multiplexing techniques, such as wavelength division multiplexing. However, Milione stresses that any spatial mode can be used for mode division multiplexing. He explains that the approach uses each spatial mode of a few- or multi-mode optical fiber to transmit an independent data signal. “As compared to the use of a single mode optical fiber, mode division multiplexing can multiply the data speed of optical fiber communication by the number of spatial modes used,” he says.

Commercial breakthrough

“A major challenge of mode division multiplexing is that spatial modes can unavoidably experience mode coupling in optical fibers,” adds Milione. Mode coupling is the transfer of power from one spatial mode to another, which results in data errors when the spatial modes are detected, he explains. There are two major trends to resolve this challenge. “The first trend is the use of a digital signal processing (DSP) technique referred to as multiple-input-multiple-output (MIMO),” Milione says. “This is effectively a digital uncoupling of the spatial modes’ data signals after detection. The second trend is the attempt to avoid mode coupling as much as possible.”

To avoid mode coupling, researchers are also studying optical fiber design. “If using OAM modes, the design of an optical fiber with a ring-core can significantly reduce mode coupling over kilometer-scale distances,” Milione says. At Photonics West, he presented a talk entitled “Mode-division multiplexing using few-mode elliptical- core optical fibers” in Session Two of Optical Communications. “It is possible to reduce mode coupling in elliptical core optical fibers to such an extent that only a fraction of a fraction of the power from one spatial mode is transferred to another spatial mode even when the optical fiber is bent into small loops,” Milione says. NEC’s mode division multiplexing products may be useful in data centers, he adds, which require high speed links between nodes.

“The push towards communication seems to be getting bigger,” Lavery adds. He believes the area where this technology would initially get deployed would be “long-distance communication channels between continents.” In this context, undersea cables are replaced roughly every 20 years, so “there is a possibility in the next upgrade to have OAM fibers or some other form of spatial division multiplexing (SDM) integrated,” Lavery says. He notes that Nokia Bell Labs is “really pushing SDM in fiber as a way of solving overstretched capacity issues with subsea optical communications cables.”

Researchers are also exploring OAM outside of the fiber medium, for example in free-space communications. Lavery’s research group is therefore currently working with British Telecom on free-space communication technologies. However there remain challenges regarding atmospheric turbulence, he explains. Consequently, researchers are also trying to apply DSP and MIMO techniques to resolve this – although without complete success, according to Lavery. “A couple of hundred meters is not so bad,” he says. “But when you start getting to a couple of kilometers, the technology we have is still not quite mature enough.” Early spatial light modulators were also “quite big devices,” Lavery adds.

“If you want to make some of these schemes commercially viable you need chip-scale ways of controlling and generating beams,” he says. Lavery and his collaborators have therefore been developing both OAM generators and detectors at the chip scale.

Another way that avoids the use of spatial light modulators was presented by Milione in a paper titled, “Machine Learning Orbital Angular Momentum Spectra,” in Session Six of Complex Light and Optical Forces XIV. “Using only a conventional camera and deep learning via an artificial neural network, OAM modes can be detected with great accuracy,” Milione says.

Dot to dot: NEC has demonstrated real-time digital OAM mode multiplexing transmission in the 80GHz RF band to meet the increasing demand for 5G networks. Image credit: NEC.

NEC is also working on radio-frequency (RF) OAM for high-capacity wireless connections for 5G networks. “Many mode-division multiplexing techniques used for optical fiber communication originate from RF communication,” Milione says. “NEC’s use of OAM in the RF domain originates from optical communication.” In that regard, NEC designed an RF antenna array and digital signal processing circuit that transmitted, separated, and demodulated multiple OAM modes in real-time. “This may be applicable in connecting ultra-dense urban areas and data centers, where it is difficult to achieve sufficient transmission capacity via conventional methods,” Milione says.

Defeating distortion

Willner is pleased by the involvement of companies in OAM. “It’s less blue sky than it used to be,” he says. Yet Willner also recognizes that industry involvement is the exception rather than the rule. Prior to the recent excitement, issues like distortion of the wavefronts of the OAM signal by water or air turbulence meant that there had been “a lot of naysayers” about the technology, he concedes. “Turbulence can distort the phase of different parts of the beam,” Willner explains. “Some of the photons are now going to be in different modes. So you get crosstalk to other modes.” Going beyond DSP and MIMO, his team has exploited adaptive optics to apply an inverse phase function at OAM receivers that “undoes” the phase change, he says.

Those advances have resolved the turbulence issue sufficiently to enable Willner’s team to publish work on quantum communication with Robert Boyd’s team at the University of Rochester, NY. Willner notes that OAM is well suited to this application because quantum bits, called qubits, encode both a zero and a one state orthogonally. The many orthogonal beams OAM combines orthogonally enable researchers to transport many more such combined quantum states. Jiapeng Zhao from the University of Rochester presented this work in a poster at Free-Space Laser Communications XXXII. In Optical, Opto-Atomic, and Entanglement-Enhanced Precision Metrology II, Willner talked about the next steps that will lead to more practical OAM systems.

Andrew Forbes, at the University of Witswaterand, Johannesburg, South Africa, emphasizes that mixing quantum with classical encoding using approaches like OAM, means information is not only transmitted faster, but more securely. His team’s work includes quantum self-healing beams for through-space communication. “Using Bessel beams to carry OAM, we can pass through obstacles, and still keep the communication link intact,” Forbes explains. “That makes it robust to any perturbations. And then that would be a scheme to improve the speed by using different patterns to get the robustness up.”

By combining two structured light fields in the form of Bessel beams (a), the resulting beam of “petals” (b) spins as it travels. The spin can be made to speed up or slow down. Credit: Andrew Forbes.

In this approach, light that passes around small obstacles re-interferes and creates the beam that existed before, enabling the communication link to remain intact. While this approach has been around for a decade, Forbes‘ team seeks to “blend in the quantum, so that we can make the link fundamentally secure. We’re really pushing the boundary in terms of how many dimensions we can use with quantum and structured OAM light,” Forbes says.

Forbes gave the keynote presentation in Session Six of Complex Light and Optical Forces XIV. In the keynote, Forbes gave an overview of the concepts behind structured light and lasers, and his group’s recent work. Forbes and Lavery also led a workshop on methods of complex light during Complex Light and Optical Forces XIV on the same day. Meanwhile Forbes’ collaborators gave several other presentations throughout Photonics West, using quantum and classical methods to create and detect structured light, and self-healing beams – and also discussing OAM toolkits they’re creating.

Beyond communication

Such toolkits are needed because rather than offering everyday practical use, “OAM still sits very much in the labs” Forbes admits. “It’s because the kits to create, detect and manipulate it are still too sophisticated.” Forbes‘ team is therefore trying to develop toolkits to enable practical applications. “You can’t expect the user to be an expert in OAM,” Forbes says. “Let’s say they’re doing some microscopy or metrology experiment. They just want to press a button and have what they need.”

This will help enable one of the major trends that Forbes sees: using “structured light” in metrology and quantum control. “You can tailor the light with OAM, and then you can measure rotations more accurately,” he explains. “A quantum experiment traditionally would be done with polarisation, and you’d get two levels. But OAM has got an infinite number of levels. So you can do lots of cool quantum mechanics with structured light.” As such, Forbes notes that OAM is already being used in such non-communications applications, but it is hidden “inside the box.”

In Session Seven of Complex Light and Optical Forces XIV, Lavery presented one such measurement application of OAM, this time in environmental sensing. In this case, the approach considers photons travelling through a medium that scatters them, known as ballistic light. When light shines through muddy water, a lot of light bounces off the particles in the water, Lavery explains. A sensor detecting ballistic light travelling through such water could provide real-time monitoring.

“We wanted to see if you could have a laser beam going across the River Thames and detect if someone’s dumping something illicit further up the river,” Lavery says. Light beams in an OAM mode physically interact with the particles and become distorted, which Lavery’s team use to infer particle size. “If you think of an OAM mode as a spinning top, when you go through a channel full of particles it gets knocked around,” he says. “We are analyzing how that movement is changing, and using that as a way to detect particle size.”

This is a good example of how Forbes sees the promise of OAM evolving, beginning from non-communications applications. Microscopy and imaging are OAM’s starting points, he says, although they may not explicitly mention they use it. For OAM to proceed from here to communication, Forbes sees three key considerations. The first is efficiently creating twisted light beams at high enough power. The second is how to detect those beams. The third is how to deliver them from one place to another.

“In communications, the creation and detection steps have been completed,” Forbes says. But for communications “further work is needed on the channel,” he adds. When the channel is fiber, more work is needed to fix modal crosstalk. When the channel is free space, more work is needed to fix distortion caused by rain, fog, dust, and atmospheric turbulence. “How do you unravel this messy channel?” Forbes asks. “That’s the final hurdle.”


The study was published in SPIE