Most of the time no one has to think about how the cellular networks we all rely on work. But it won’t be surprising to many people to learn that, as is often the case with infrastructure, some parts are the latest technology and others have remained unchanged for decades. Eridan is a well-funded startup that aims to replace one of the following with a fundamentally different hardware approach that can make mobile networks more efficient.
Everywhere you look – or more likely, right above where you look – there are cell towers that connect your phone to the wider internet. You can think of them as made up of three large parts: the modem, which exchanges data with the rest of the network; antenna, which broadcasts radio signals in sync with dozens or hundreds of nearby devices; and the transceiver, which sits in the middle and converts the modem’s digital data into the actual RF signal the antenna emits.
Obviously modems had to change over time and increase capacity, and it did. Likewise, the antenna must change to reflect the new and changing spectrum slices being used for mobile data. But the transceiver is like a digital-to-analog converter in that its job doesn’t really change much – data in, signal out.
Recently, however, we’ve begun to explore the limits of the bit in the middle, which is a technological dinosaur.
“The way that transceiver powers the antenna hasn’t changed in 70 years,” said Doug Kirkpatrick, co-founder of Eridan. “It’s called a linear amplifier and that device, that circuit architecture approach is really at the limit with today’s 5G.”
These amplifiers are essentially analog devices, and because of their basic nature, the more power you put into them, the less efficient they become. And over the years, the power used has only increased as the number and complexity of signals has increased; 5G transceivers are almost half as efficient as 4G transceivers, half as efficient as 3G – however due to spectrum constraints and growing demand we’ll need more more 5G cells. Even if the difference is only a few watts, it adds up very quickly if you plan to cover all populated areas plus highways.
“If you want to go everywhere, it’s like 20 times as many radios, and with efficiency also decreasing, you’re talking about taking 50% of the electricity in the United States,” says Kirkpatrick. “This dog doesn’t hunt.”
“What we’ve done is what this industry has been for 30 years, and what every wireless company is after, costing billions of dollars. If you want everyone to have 5G without melting the planet, we are the only way for this to happen,” he continued.
And what exactly did Eridan do? I hinted at my skepticism that a startup with limited means could surpass decades of research by some of the richest companies on the planet.
“Well, we cheated,” admitted Kirkpatrick. One of the company’s founders, Earl McCune, sad person passed away two years ago in development, was one of a number of people doing research at those big telecommunications companies, where his approach never worked. That doesn’t mean he’s carrying trade secrets, of course – they just figure out how to make the theory a reality outside of the corporate structure.
After meeting during a failed hire to do related work for a large company, the founders decided that they liked each other enough to pursue the concept independently.
“We did what you would do, sat down with a napkin at the bar and a pint of beer,” he recalls, and after filling the bucket to think for a bit, “It was one of the most obvious things. course. People’s eyebrows went up and they said, ‘that’s a crazy idea…but it can work.’ Then, literally, the first time we turned it on, it was the most perfect signal you’ll ever see. We said, ‘how the hell are we going to explain this to anyone?’ ”
Advancement is in some ways a simple step forward, like going from a vacuum tube to a transistor.
“A linear amplifier is an analog amplifier; Since they were worried about cost and efficiency, they took advantage of the worst bargain: [i.e. how noisy] Can you generate the signal and still be effective? This is fundamental to what linear amplifiers do,” he explains. “We are a digital switch – we send out a super clean signal. It’s a hundred times smaller and a hundred times cheaper.”
“This kind of architecture is directly polarizing, Earl has written books about it — he’s the man who knows this kind of architecture,” says Kirkpatrick. His other co-founder, Dubravko Babiç, is a materials specialist focused on gallium nitride, which, without going into detail, is used in conjunction with silicon to create highly efficient chip architectures. The GaN-silicon combination here allows them to make a huge leap from around 10% efficiency in installed devices to 50%.
They first secured a $5 million contract from DARPA, thinking it could be used to miniaturize military radios, but they soon realized that the technology was beyond defense. and bring it into the field of telecommunications.
The result is a “Miracle” device (only looks like a normal PCBby the way) so different from the existing infrastructure that they have a hard time making the potential companies understand its qualities. “But over a year ago, the skeptics had everything they needed – we demonstrated it atop the mast. You can be tech-savvy all you want, but when you light it up on the mast, it’s game over,” said Kirkpatrick.
However, he admits, the infrastructure market remains cautious. These are companies that pay huge sums to build millions of installations serving hundreds of millions of people – they tend to use what they know works even if a newcomer comes along better and cheaper price. But a pilot test at Fort Hood will show Eridan’s 5G mini-mobility capabilities, assuming that everything goes well, it will lead to commercial rollout sometime in late 2023.
What is likely to convince them, beyond the existential threat of linear amps reaching the theoretical limit of the amount of data they can handle, is the further scalability of the Eridan technology. It will be enough to make 5G deployment cheaper and better – but what about the next upgrade?
The latest signaling protocols come from mobile data regulators and standard setters regarding 8-bit signals pushing 256 QAM – we won’t go into technical details here but you You can think of it as the equivalent of home internet bandwidth. Essentially, the more bits you have that fit into a given signal range, the more data you can deliver, although as always happens with wireless, the more risk that the signal is This increasingly complex and fragile does not come intact.
As you can imagine, the production of that signal from analog to digital has a huge effect on how efficient the transmission is. “Which one do you think travels farther, a clean signal or a dirty signal?” Kirkpatrick asked, clearly flamboyant. The use of GaN allows the system to operate at high voltage, eliminating the need for an amplifier, further improving the signal as the amplifiers amplify the “noise plus the good” in a dirty signal. .
Eridan had a 10-bit, 1024-QAM test released by 3GPP that worked within hours of release and has moved to show in lab settings that they can transmit 16-bit signals , 64K (ie 65,536) -QAM. (Believe that people who like wireless protocols will find that number impressive.)
The promise of becoming a staple of the infrastructure that needs to be built over a decade and beyond has clearly activated the brains of investors writing checks. Following $5 million from DARPA, Eridan raised a total of $46 million between today’s announcement of a B round and a secret 8MA round. Latest round do Capricorn Investment Group, Monta Vista Capital and Social Capital (led round A).
“We’re set to make tens or hundreds of millions of these,” says Kirkpatrick. While he couldn’t name potential customers, it’s not hard to imagine who would benefit from this hardware. Basically, if you’ve heard them rave about 5G at some point in the last 5 years, they’re probably on the list.
Commercial deployments will begin to appear after official demonstrations at Ft Hood and elsewhere next year. You probably won’t notice anything – but then again, that’s the point.