The interview: Marcus Weldon, CTO, Alcatel-Lucent, and president, Bell Labs

2 Dec 2014

The world is on the edge of a new age where broadband speeds of 1Gbps and higher will transform commerce and culture. It is Marcus Weldon’s job to see that this happens.

When you walk around Bell Labs’ headquarters at Murray Hill in New Jersey you very quickly get a sense of the technological history of the world as we knew it for the best part of the last two centuries.

The labs owe their heritage of course to Alexander Graham Bell, the inventor of the telephone. And since that time the labs have gone on to spawn numerous inventions and businesses, including telecoms giant AT&T.

Bell Labs has since gone on to invent radio astronomy, the transistor, the laser, information theory, the UNIX operating system, the C and C++ programming languages and lots more. Scattered among Emmy Awards and Academy Awards are machines like the first trans-Atlantic radio antenna and the first cinema machine used to combine movies with sound and much more.

The labs in New Jersey are now home to at least eight Nobel Prize winners and at least 15 Laureates, some of whom are still hard at work trying to crack the future of communications.

Bell Labs, which became part of Alcatel-Lucent when its parent company Lucent merged with Alcatel, is at the spearhead of the company’s efforts to solve problems like finite capacity in copper, fibre and wireless.

Walking through rooms like those where Ken Thompson and Dennis Ritchie invented the Unix operating system, you realise just how intimate and down-to-earth the working environment had been for some of the world’s most extraordinary minds.

Weldon, a British citizen living for decades in the US but who hasn’t lost a shred of his accent, barks a laugh at my observation and says the challenge the scientists and engineers are attempting to overcome are anything but ordinary.

He has set a challenge for his R&D teams to boost network technologies by 10X to function in a world where people will want data everywhere and the number of devices will more than quadruple in just a few years.

“We put that number out there 10X because we actually need 100 times more radio capacity. We say that because it’s an easy round number but in fact the better statement is we need to build a network that has seemingly infinite capacity.”

Weldon says we are already at the limits of the Shannon-Hartley theorem in terms of the rate at which data can be transmitted over bandwidth.

“We have only one dimension left to do this in and that is down to space. The final frontier of radio is we need to go to shorter distances for fewer users, which means creating small, indoor cells otherwise in terms of wireless we are going to run out of dimensions again.”

Weldon has a point. If you look at the latest 4G networks they are using redundant TV and 2G frequencies. TV whitespaces are also being championed as a way of managing data wirelessly. And rather than 5G being a new frequency band, 5G technologies will be collections of tools for managing an array of wireless signal types.

“We are literally at the final frontier of wireless as we know it and so we have to create lots of tiny wireless networks if we want to boost speeds and capacity.”

Another problem – which we know well in this part of the world – is wired connectivity. How much use can we get out of existing copper technologies, how realistic is it to bring fibre into every home (FTTH)?

“The more wireless you build the more wired it is. Realistically people will require a wireline network within 30 metres of them.

“Wires matter. Fibre matters and we care about radio for those last 30 or 40 metres.”

Joining dots on the internet of things

The internet of things heralds a time when every conceivable device will have an IP address and connect wirelessly to the internet, no doubt complicating Bell Labs’ engineers task.

“This means people, it means objects and it means buildings – anything you want to communicate and control. Text to open a door, use the same interface for talking to people.

“Fundamentally underlining this is the desire to connect to everything digitally and have everything delivered digitally – video, audio or data.

“But now with 3D printing you can deliver lots of things digitally and turn them back into a physical things. Think of it like you need a widget or a car part. Rather than going to the Home Depot to buy one screw, wouldn’t it be great if you could just print it.”

Alcatel-Lucent has embraced this thinking in producing new products. “Our gold box antenna on our small cells is a 3D printed antenna with gold on it.

“Think about this for a second – 3D printing materials contain the same polymers as nylon rope. If you melt the resin in the rope you can print wood. If you think about that for a bit it means you can deliver physical objects.

“Today we are all about communicating digitally and exchanging video files but it’s not out of the question that soon we will digitise physical objects by recreating them at the edge.

“It’s a very interesting recognition that our world is about to change massively.

“For example, is the future of Wal-Mart going to be in 3D printing goods, like half the store will be a 3D printing factory?

“A new model is emerging at Amazon where they actually print and bind the books as they are ordered rather than storing them in a warehouse.

“We are about to move to a world where not only will I be able to connect with everything but I can recreate it where ever I am and that is an interesting new world.”

Scaling the broadband chasm

As Weldon points out the only thing dividing us from our imagined future and the real future is physics and the battle is now on to squeeze the most capacity out of finite resources in wired and wireless. Even fibre, he says, has its limits.

“It means leveraging wireline infrastructure to build wireless networks. If we realised we had to put radio at the end of wires deep into the network and don’t fall short by trying to do everything on a macro layer I think we can get there.

“I think we can solve all the problems and actually get reasonable economics. The problem is more people take interim steps that delay and defer because they don’t see with clarity what the future will look like.

“When you do that you end up with a chasm where the demand exceeds supply and you slow down demand because you teach people to live a different way than they actually wanted to.

“Are we going to stop short in a world where the demand exceeds the supply? We don’t know. But the key step is if the network is valued correctly investment will appear and service providers will build infrastructure because the value is perceived by the market.”

Across the world nations are trying to find ways of better connecting people and businesses because the economic upside is enormous. In Ireland the State is aiming to invest €512m to connect 600,000 homes and 100,000 businesses via fibre. In addition to this a Vodafone/ESB consortium is preparing to build a nationwide fibre network and incumbent operator Eircom is also planning to build a 1Gbps fibre network that will connect 66 towns.

Like all telecoms equipment manufacturers Alcatel-Lucent is watching these developments closely. Weldon’s preference is for steady, focused investment in future-proofed networks.

“I am arguing for rational investment and a rational recognition of the need for this infrastructure.

“People want to connect to it, they want to live their lives digitally and wirelessly.

“Once people realise they need massively scalable networks investors will start investing.

“There is a new existence coming that goes beyond the hype of the internet to a more evolved digital existence. It won’t be just connecting to web services, but attaching yourself to lots of different things such as a pacemaker that communicates via the internet with the hospital. You need that connection to work.”

Envisioning the future of innovation

Considering the legacy of Bell Labs and Alcatel-Lucent, it must be quite a strain maintaining such a prolific pipeline of innovation. 

Weldon smiles wryly at the question and shrugs. “One sextillion is a real number. That’s a one followed by 21 zeros. That’s 1,000,000,000,000,000,000,000.

“That’s how many transistors there are in the world today. And the transistor was invented at Bell Labs. One sextillion is also the number of stars in the Milky Way. We’ve come a long way and it is the legacy of innovation here at Bell Labs that is driving us forward.”

He says out of eight Nobel Prize winners and 15 Laureates, many are still working at Bell Labs today. “The one we won this year for microscopy was based on work that began back in 1995. It typically takes 20 years to win the prize because you have to demonstrate the significant impact of your work. The first decade it is a thing, the second or third decade it could be a huge phenomena. It is a winner if it is transformative.

“There are many things we are working on today that have that potential. Maybe it is vectoring, optical capacity or an algorithm that enables the digitization of everything,” he concludes.

“It is increasingly phenomenal for the people at Bell Labs to realise they are transforming the world, not just device physics.”

John Kennedy is a journalist who served as editor of Silicon Republic for 17 years

editorial@siliconrepublic.com