Optical research: Why we break world records

1.52 Terabits per second – that’s the equivalent of 1.5 million YouTube video streams simultaneously transmitted. A Nokia Bell Labs research team led by Fred Buchali recently achieved that world-record-shattering milestone over one wavelength of color on a single strand of fiber that can carry many wavelengths at once, covering a distance of 80 km. That bitrate is nearly four times larger than the capabilities of current state-of-the-art for optical systems, and it is just one of several record-breaking innovations in fiber capacity, reach and efficiency that Nokia Bell Labs presented virtually at the Optical Fiber Communications Conference (OFC) this month.

There’s no question that 1.52 Tb/s is incredibly fast, and for some it might raise the question of why such enormous capacities are necessary. From our point of view, it’s not a question of “why” but “when” – or more to the point, “how soon.” The world’s thirst for information has been growing by 60% per year for 30 years and there is no reason to believe that thirst will be quenched at any point in the future.

Every time we’ve established a new high-water mark in optical capacity, the demand has risen to fill it. Our job at Nokia Bell Labs has always been to invent the future before it arrives. That’s why we’re constantly striving to achieve that next higher bitrate, that further transmission distance, that additional mode or that new grouping of cores.

What’s remarkable is that with each milestone we reach, the next milestone becomes that much harder to achieve. There is a fundamental limit to the quantity of error-free information that can be transmitted over any communication channel. It’s known as the Shannon Limit – named after famed Bell Labs researcher and father of information theory Claude Shannon – and in optical fibers we’re  approaching that theoretical boundary. In the past, orders-of-magnitude improvements in transmission rates on a fiber were viable goals in telecommunications research. Today there isn’t an order-of-magnitude improvement to be had.

That doesn’t mean innovation has slowed down. We’ve merely become more deliberate in how we direct our research. Just as wireless research has moved beyond increasing bitrates over a single path to focusing on multiple paths with multiple antennas, optical research is now directed at increasing speeds over multiple fibers, multiple modes and multiple cores.

Simultaneously the applications and demands for fiber are only growing. We’re far from the time when optical capacity was only needed in long-haul data transport across continents and oceans. Fiber is now permeating our neighborhoods, our homes and businesses, and it’s finding its way directly to end devices. We’re close to a future where we’ll see direct optical links between servers in a data center or between robots on a factory floor. And each of these applications will have different demands on communications – higher speeds, lower latency, higher reliability, higher security, lower costs and smaller physical connections – illustrating the need for flexible and integrated solutions to meet these demands.

Consequently, our research is geared at multiple dimensions, all of which we addressed at OFC through dozens of paper presentations. Many of these were accepted as post-deadline papers, a highly competitive category of papers that generally detail late-breaking work in rapidly advancing areas. Here are a few of the examples of our recent research accomplishments, several of which were presented at OFC:

• Nokia Bell Labs researcher Di Che led a team that established a data-rate world record for directly modulated lasers of 400 Gb/s over a span of 15 km. These types of links are crucially important for low-cost, high-speed applications such as datacenter connections. Di was also being honored as a young emerging researcher at OFC with the Tingye Li Innovation Prize, which itself was named after a renowned Bell Labs researcher.
• Researcher Roland Ryf and the spatial-division multiplexing (SDM) team performed the first field trials of SDM cable over a 2,000-km span of 4-core coupled-core fiber. The experiments clearly showed that coupled-core fibers are technically viable, offer high transmission performance, while maintaining an industry standard 125-um cladding diameter.
• A research team spearheaded by Roland, Rene-Jean Essiambre and Murali Kodialam introduced a new set of modulation formats that provide improved linear and nonlinear transmission performance at submarine distances of 10,000 km. The proposed transmission formats are generated by a neuronal network and can significantly outperform traditional formats (QPSK) used in today’s submarine systems. 
• Junho Cho led a team that experimentally demonstrated capacity gains of 23% for submarine cable systems that operate under electrical supply power constraints. The capacity gains were achieved by optimizing the gain-shaping filters using neural networks.

With OFC 2020 behind us, we still have plenty of work ahead of us. First, many of these innovations will form an essential part of Nokia’s already leading optical product portfolio in the months and years to come. The optical business is already on a tear to become a market leader with some of the most disruptive products in the industry, and these innovations will help to truly set it apart. Second, we’re already researching the next big optical innovations. We’ll continue to pursue new world records with the same commitment we always have. But we’ll do so not because they’re world records. We’ll do so because we must constantly redefine the applications and limits of optical communications to meet the world’s insatiable demand for data.