China has the most extensive network building and deploying the fifth generation of cellular networks. 5G runs up to 100 times faster than 4G. It has built 1.6 million 5G base stations across the country, according to the state-run People’s Daily. In comparison, the U.S. has only about 100,000. Yet, instead of trailing China, the Americans could have been the leader in 5G development had an immigrant been allowed to stay in the country. Erdal Arikan may not be a household name. Yet he’s so extraordinary that Chinese telecom giant Huawei rolled out the red carpet for him at its headquarters in Shenzhen in 2018. They treated him like a rock star because he developed the fundamental technology for 5G. Arikan was born and raised in Turkey, the son of a doctor and a stay-at-home mom. He loved science and studied electrical engineering at Middle East Technical University in Ankara. While in his first year, in 1977, Arikan’s high test scores allowed him to transfer to the California Institute of Technology, known as CalTech, one of the world’s top universities with a strong focus on science and engineering.
While there, he took a particular interest in information theory…a study of mathematical laws that govern communications, including 5G networks. Information theory was developed a few decades earlier by the brilliant Claude Shannon at AT&T’s research arm Bell Labs. His ideas laid the groundwork for the digital revolution. He made the internet possible. In 1948, Shannon published a mathematical theory that showed how to transmit messages accurately at the fastest speed possible. Say, Scott wants to send a message to Sarah. A transmitter encodes information into a signal, which is corrupted by noise and then decoded by the receiver. Shannon’s theory showed how to calculate the maximum rate at which messages can be sent with zero errors to ensure that the message Scott sent is the same one Sarah received. According to Shannon, every communications channel has a “speed limit” for transmitting information reliably, referred to as the “Shannon limit”. However, Shannon did NOT provide a technique for constructing the code to achieve the fastest rate possible or make it clear how to compile such code. That’s where Arikan came in. After graduating from CalTech, he took his interest in Shannon’s theory to MIT for his grad studies. His mentor there, Professor Robert Gallager was a renowned pioneer in information theory who came close to the Shannon limit but never reached it. Arikan would throw himself into this work for the next twenty years, trying to attain what no one had yet been able to do. After finishing his doctoral thesis in 1986, he briefly served as an assistant professor at the University of Illinois at Urbana-Champaign. However, he was reportedly unable to obtain a permanent academic post in the U.S. or funding to continue his work on information theory. And as a foreign citizen, he had to find an American employer interested in supporting him in order to remain in the country. But he failed to do so…and had no choice but to return to Turkey. He once told Wired the reason scientists and engineers in the U.S. don’t spend decades working on the same problem is “Because they wouldn’t be able to get tenure; they wouldn’t be able to get research funding.” He returned to Turkey to join the country’s first private, non-profit research institution, Bilkent University – now a leading institution in the country. He established its engineering school and gave lectures and published papers.
But the Shannon limit kept nagging at him. How do you transmit messages accurately at the fastest speed possible? Messages that you send – letters and emojis – are converted into a string of ones and zeroes called bits. The problem is noise can distort the message, so redundancies are required to make sure the correct message is received. Redundancies ensure that if an error does creep in, you can still be sure of the correct string and what the original message intended to say. However, there is a trade-off: the more redundancies you add the slower the transmission rate. So, how did you get around that problem? After working on it for over two decades, Arikan had a EUREKA moment in late 2005. His solution was to make a copy of the channel and transfer the noise from one channel to the other, so you’re left with one noisy channel, and the other that’s much cleaner. The noise-free channel could in theory reach the Shannon limit. He dubbed his discovery: “polar codes” In reference to banishing noise to the North Pole, so that there is clear communication at the South Pole. He continued to tinker with his discovery, and in 2009, he published his findings in the field’s top scientific journal, to much fanfare. Arikan tried to sell the idea of polar codes to both Qualcomm and Seagate, but they declined. No American company was interested in his ideas! He blamed himself for not being a good enough salesperson, telling Wired: “I was an academic who did not know how to promote an idea. Perhaps I did not believe in the idea that strongly myself.” But one company did believe in the idea very strongly. The head honchos at Huawei saw the potential of polar codes. They believed it was the breakthrough needed to leap from 4G networks to 5G. Polar codes are considered a direct challenge to the American-developed coding method “LDPC” and are believed to be superior to it. The Chinese company bet big on polar codes – investing hundreds of millions of dollars to build on Arikan’s work. In fact, when Canadian tech giant Nortel went under, Huawei poached its top researcher who grew up in China. Wen Tong would head Huawei’s R&D division on 5G. Huawei’s engineers came up with the software to make polar codes work in practice and invented some of the first 5G technologies. Today, Huawei holds the majority of polar code patents for 5G and has positioned itself as a world leader in tech innovation. Western countries insist Huawei’s technology may intentionally contain security holes that could be used by the Chinese government for spying…which Huawei flatly denies. The U.S. has banned its equipment. Despite the tension between the east and the west over the development of 5G technology, Arikan is thankful that his tireless work and refusal to give up…has finally paid off. As engineers, there is no greater reward than seeing our ideas turn into reality. Thank you for making it happen.
Engineers like Arikan are shaping the world of tomorrow. If you’d like to hone your STEM skills so you can build a product that has the potential to impact a tremendous number of people, Brilliant is a great way to brush up on your science, technology, engineering, and math skills. Their courses are super interactive. I know my viewers love the computer science fundamentals course which helps strengthen your knowledge of core computer science concepts and no coding is required. You can solve puzzles with science through their scientific thinking course, start your algebra journey with an introduction to variables and equations, or see fractions, percentages, and ratios in everyday contexts with their Everyday Math course. There’s something for everyone no matter your experience level – whether you’re a student or you’re working and want to brush up on what you already know. There are no tests and if you ever get stuck, you can view the explanation for more. Brilliant is FREE for you to try out if you head to the custom link in my description or pinned comment: brilliant.org/newsthink. The first 200 people to click my link will get 20% off the annual Premium subscription which gives you unlimited access to all of their offerings. Thanks for watching. For Newsthink, I’m Cindy Pom