In our collective imagination, the digital world has been defined by technology titans such as Bill Gates, Steve Jobs and Mark Zuckerberg. But they all stand on the shoulders of a man who has become far less of a household name, even as technologists the world over compare his influence to that of Isaac Newton, Albert Einstein and his distant relative Thomas Edison.
It is no stretch to trace all forms of modern computing and communications to Bell Labs legend Claude Shannon. Known as “the father of information theory,” it was his pivotal paper, called “A Mathematical Theory of Communications,” published 75 years ago in July 1948 that essentially spawned the digital age.
The principles Shannon laid out in his landmark article established the foundation for modern communications, computing, digital media, compression, cryptography and the internet. Wherever we go, wherever we are, we all carry a bit of Shannon with us in our pocket.
His theory defined the parameters under which any transmission of information takes place: a mobile phone connecting to a tower, a wavelength of light traversing a fiber-optic cable, two people having a conversation in a noisy room – all these are governed by Shannon’s laws.
Even today, Nokia Bell Labs achieves its most groundbreaking advancements in networking by working within the same information theory framework Shannon created 75 years ago.
“It’s hard to overstate the impact that Shannon has had in the revolution that is communication, and all the societal changes that enabled,” said Tod Sizer, Head of Optical Systems and Device Research at Nokia Bell Labs. “If you step back and consider the role that his work has had in connecting the world, it’s been simply amazing.”
An insatiable mind
Shannon’s 1948 breakthrough marked the apex of a lifetime of scientific adventures. As a playful child growing up in Michigan, Shannon tinkered with logic games, puzzles and cryptograms. He even set up a wire telegraph in his backyard. It was as an undergraduate student at the University of Michigan, though, where he first began to explore how Boolean algebra could be applied to communications, adapting the Morse Code of dots and dashes into a binary system of zeros and ones.
His renowned master’s thesis at the Massachusetts Institute of Technology, titled “A Symbolic Analysis of Relay and Switching Circuits,” demonstrated just how this concept could construct any logical numerical relationship and it established the idea that computers could be made to think and mimic the human brain. In fact, that concept later led to Shannon’s creation of “Theseus,” a mechanical mouse controlled by an electromechanical relay circuit that enabled it to move around a labyrinth of 25 squares. The 1950 invention was one of the first applications of artificial intelligence. Six years later, Shannon co-authored the paper that formalized AI at the inaugural Dartmouth conference.
After earning his PhD in 1940, Shannon became a National Research Fellow at the Institute for Advanced Study at Princeton, where he continued to expand his ideas and enjoyed occasional encounters with Einstein and Kurt Godel. He then joined Bell Labs, where he was recruited to the war effort to work on fire-control systems and cryptography. It was there he first met Alan Turing, the renowned British breaker of the World War II German Enigma code. The two bonded over tea in the cafeteria discussing their shared interest in code breaking and the enciphering of speech.
“He was this intuitive genius in both mathematics and engineering,” said Leonard Kleinrock, an internet pioneer and renowned professor of computer science at UCLA who was also a PhD student under Shannon at MIT. “Engineers look at devices and try to figure out how they work. A mathematician tries to figure out why they work that way. Shannon could do both.”
As Shannon’s stature and influence grew, so did his eccentricities. At Bell Labs, he was known for riding around the Murray Hill, New Jersey campus on his unicycle while juggling specialized bowling pins.
Besides his historic innovations, Shannon also proudly pursued a variety of quirky inventions that had little practical purpose beyond satisfying his own curiosity.
“I am very seldom interested in applications,” Shannon said later in an oral history. “I am more interested in the elegance of a problem. Is it a good problem, an interesting problem?”
Such “problems” included designing a computer that ran on roman numerals, a device that could solve a Rubik’s Cube and the first wearable computer that helped predict a roulette wheel result. Indulging his peculiar hobbies, he also pioneered a variety of juggling machines, rocket-powered frisbees, flame-throwing trumpets and a remote lawnmower. He dabbled with formulas that tried to predict the stock market and developed a chess machine, predicting that one day a computer would beat a world champion. (As IBM’s Deep Blue eventually did in 1997, beating grandmaster Garry Kasparov).
Shannon, who returned to MIT in 1956 to hold an endowed chair until his retirement in 1978, was no arithmetic whiz. In fact, his wife Betty, who was a “human computer” at Bell Labs back when women dominated the field, was far more renowned for her mathematical calculation prowess and the couple often collaborated closely. But Shannon made his name because of his unique approach to solving problems.
“He had that innate ability to take something complex, reach into it and yank out the essence,” said Kleinrock. “He had no constraints. He was always thinking outside the box.”
Kleinrock said that was the crux of Shannon’s genius.
“The pure brilliance of the man was to address problems that were unfathomable to others,” he said. “That’s how he brought a uniform way to talk about communication systems.”
The Bit Player
This rich career of scientific inquiry and unbridled curiosity led to Shannon’s 1948 breakthrough that all information – text, pictures, music, video, everything – could be broken down into a binary choice: a bit.
That, coupled with his fellow Bell Labs scientists’ invention of the transistor less than a year earlier, set off the revolution that transformed our world.
Shannon first laid out the basic principles of digital communications in “A Mathematical Theory of Communications,” published in the Bell System Technical Journal in July 1948. His seemingly simple diagram of an information source that produces a message – which through a transmitter conveys a signal through a noisy channel, which is then received and transformed back into the original message – established the foundation of everything that followed.
“The fundamental problem of communication is that of reproducing at one point either exactly or approximately a message selected at another point,” Shannon explained in the pioneering paper. “A basic idea in communication theory is that information can be treated very much like a physical quantity such as mass or energy.”
The article developed the concepts of entropy and redundancy and introduced the term bit as a unit of information. A “bit” is a portmanteau of “binary digit” and represents a random binary variable that is either 0 or 1 with equal probability. Shannon’s influence on the concept was so profound that this basic unit of information in computing also came to be known as a “Shannon.”
“I enjoyed working on this kind of a problem as I have enjoyed working on many other problems without any notion of financial gain or being famous and so on,” Shannon said in a 1986 interview about information theory. “I think indeed that most scientists are oriented that way that they are working because they like the game.”
In predigital days, communication channels such as phone lines or radio bands were particularly susceptible to electrical or electromagnetic disruptions known as “noise.” Shannon proved the counterintuitive result that no matter how noisy a channel, information could be sent over it error free. All that was needed was a way to add enough redundancy to the information so that errors could be corrected.
The Shannon Limit
Shannon showed that any communications channel could be characterized by two factors: bandwidth and noise. And he showed how to calculate the maximum rate at which data can be sent over any channel with zero error. He called that rate the channel capacity, but today, it’s just as often called The Shannon Limit.
Decades later, we have now bumped against this boundary in both wireless and optical communication. But Sizer, the current Bell Labs optical research leader, said Shannon’s work remained as relevant as ever since any innovation still must abide by his principles.
“It’s just the way we address growth and capacity that is different now. For the past decade, prior to 2018, we were trying to squeeze more information into the same sliver of spectrum – that’s the basis of Shannon’s Limit,” he explained. “Now we seek to increase the bandwidth by looking at new types of fibers as well as new types of amplifiers.”
The biggest trends in optical research today are focused on multiple-core, multiple-fiber and multiple-mode systems, technologies that would increase the number of transmission channels rather than increase the efficiency of an individual channel. Meanwhile, in wireless research, the biggest capacity advances are achieved through using wider spectral bands and massive antenna arrays to create simultaneous transmission beams with high spectral efficiency to multiple devices.
These trends are crucial since the global volume of communication is increasing at a historic pace. Sizer said consumer-driven demand for capacity typically grows at a rate of 60% a year, but in the age of artificial intelligence the increase in machine-driven demand is growing at a rate closer to 100% annually.
“Just about all the physical layer communication mediums have been challenged by the information limits that Shannon told us about,” Sizer said. “The broad application of Shannon’s work across so many domains illustrates how fundamental his global impact has been to the communications revolution.”
Shannon passed away in 2001, at the age of 84. But his influence only continues to expand as Shannon’s bit is rapidly penetrating every corner of human existence. The metaverse is fusing the digital and human worlds, the Internet of Things is growing to billions of devices and AI is omnipresent.
None of it would be possible without Claude Shannon.