Having spent heavily on massive MIMO development, both Ericsson and Nokia have a vested interest in promoting it over alternatives.
"Look at the companies that are rolling out massive MIMO – they're very entrenched with the customer," said Matytsine.
He is, however, prepared to concede that massive MIMO has certain attractions over his passive multibeam antenna technology.
Where MatSing excels at capacity, massive MIMO looks superior for throughput.
Massive MIMO cannot evade the fundamental laws of physics.
A Butler matrix, despite the curious name, is not an array of domestic servants but the critical, passive component of an antenna panel. Used in a 5G mobile network, it can send out multiple beams to support device connections. The drawback is the phenomenon of beam "squint." When a beam fired from a neat lineup of antenna elements is angled left or right, its shape alters and it begins to intrude on its neighbors like a passenger leaning into another as a car rounds a bend.
But what if a lens could be shaped like an eyeball so that beams could point in many directions without such interference? The concept of a refractive, spherical lens was proposed in the 1940s by scientist Rudolf Luneberg, who gave his name to the design. More than 60 years later, Dr. Serguei Matytsine and his sons Leo and Michael, who had grown up in Singapore, made the Luneberg lens a foundation of MatSing, the company they founded in 2005. The brand, a portmanteau of "materials" and "Singapore," also carries echoes of the family name. Today, its lenses are visible at stadiums across the US, where MatSing is headquartered.
Related:Why Verizon hid its 5G network in the Super Bowl stadium
The sports arena is where MatSing has enjoyed most success, but it is not the environment in which it started. Back in the late 1990s and early 2000s, several governments were trying to build a large array of antennas to receive weak signals from space. A presentation by the intergovernmental project that eventually became known as "Square Kilometre Array" caught the attention of the Matytsine family.
"Somebody said there's this technology called a Luneberg lens that actually has quite a few advantages compared with dishes because it can see a wider field of view," said Leo Matytsine, a co-founder and executive vice president at MatSing. "You can have multiple independent beams."
The difficulty of making a Luneberg lens that could be used instead of a dish centered largely on the necessary materials. "You need special materials, dielectric materials that have certain properties, in order to construct such a lens," explained Matytsine during an interview with Light Reading. "Nobody previously was able to find a material that would be suitable, and traditional materials don't work so well for communications."
One of MatSing's original innovations was a new type of lightweight metamaterial invented by Leo's father. It drew MatSing into the satellite sector, where a single Luneberg lens could provide an alternative to the use of several dish antennas for tracking different satellites. From there, MatSing has moved into terrestrial communications.
Related:Balls don't lie: Ericsson and Nokia's big 5G bet meets multibeam threat
Stadiums hosting tens of thousands of people seem a natural fit for MatSing and its Luneberg lens antennas. Lodged in the steel skeleton of the building, they are described as looking like beach balls by Earl Lum, the founder of analyst company EJL Wireless Research, who sounds impressed by what he has observed.
Avoiding the problem of squint that is common to panels, a MatSing sphere with its many beams should theoretically be able to provide blanket coverage for everyone at a game or concert. "Lenses are ideal for multiple beams, and multiple beams are perfect for sectorizing and therefore producing high capacity from a single antenna," said Matytsine.
MatSing ball spotters would today be able to see the equipment at many of the biggest US venues, he said. Those include around 70% of the stadiums used by the NFL, various racetracks and concert locations such as Coachella in Indio, California, just a couple of hours' drive from MatSing's headquarters in Irvine.
Material science
But the company is facing competition in this sector. At the front of the pack is Amphenol, a giant equipment maker that reported sales of more than $23 billion last year. Through a unit called Andrew, acquired from CommScope as part of a $2.1 billion deal in July 2024, it also offers passive multibeam antennas. Originally, those were based on Butler matrices, but Andrew has more recently begun to market a metalens under its "CleanBeam" brand. It appears to work on similar principles to MatSing's Luneberg lens.
Related:Massive MIMO attacked as Galtronics bets on multibeam 5G growth
A much smaller competitor from Canada is Galtronics, whose product looks more like a conventional panel. Acknowledging MatSing has been the dominant force in the market, CEO Leighton Carroll struck a bellicose tone during a recent discussion. "We came up with a patented, panel-based multibeam," he told Light Reading in September. "We have a product that weighs half as much as an equivalent MatSing, and, when we launched, cost a quarter to a third of the price."
Obviously conscious of his rivals, Matytsine shows little interest in badmouthing their technology claims. But he clearly thinks his company's technology, protected as it is by more than 60 filed patents, would be hard for anyone to replicate. "It's not electrical engineering, it's material science, and our patents cover a lot of these key new materials we've invented," he said. "It's possible to use alternate materials, but the question becomes what kind of performance you'll get out of that."
Anything subpar could struggle to provide 120-degree coverage, the typical cone-shaped measurement of a single site sector, or handle appropriate frequency bands and power constraints, he reckons. If alternatives were readily available, MatSing would probably not have observed violations by Chinese companies. "We've seen companies, primarily in China, that are straight up copying our patents, and trying to replicate, and trying to push those products outside of China as well," said Matytsine.
He also wants to dispel the notion that MatSing is synonymous with balls. The first products it developed for the telecom market in 2011 were actually cylindrical, as Andrew's competing products for stadiums are now. Other products in the expanding MatSing range look squarer and more like standard panels. Those designs came partly in response to customers trying to install spheres outside stadiums.
"People would put them on rooftops and start using these large spheres to cover busy streets," said Matytsine. "They liked the results, and it worked. But the only downside is that when you switch out a regular panel antenna to a large sphere, especially in the US, in a lot of places you have to change the planning." Reverting to more typical shapes and dimensions has helped MatSing satisfy other needs.
On the massive MIMO attack
The association of MatSing with large and cumbersome balls, nevertheless, might explain why the company is far less visible outside stadiums than in them. For all the makers of passive multibeam antennas, a much bigger obstacle is the 5G industry's current obsession with active antennas and massive MIMO, a high-profile technology that stuffs dozens of transmitters and receivers into radio units to boost network performance. Having spent heavily on massive MIMO development, both Ericsson and Nokia have a vested interest in promoting it over alternatives. And they remain the radio access network (RAN) vendors with which US and European telcos mainly interact.
"Look at the companies that are rolling out massive MIMO – they're very entrenched with the customer," said Matytsine. "They're large companies, and they're pushing out that technology, and they provide not only the antennas but the services and everything else. They want to sell the whole package." All that has made it much harder for a family-run business such as MatSing to be considered.
He is, however, prepared to concede that massive MIMO has certain attractions over his passive multibeam antenna technology. Mobile operators ideally want to offer both throughput and capacity in a busy area. Throughput essentially measures the speed of the connection delivered to a customer. Capacity is more about the volume of users a cell site can support. Where MatSing excels at capacity, massive MIMO looks superior for throughput.
The technology has succeeded by isolating beams to prevent so much interference and squint when they are angled, said Matytsine. "You don't have side lobes or other issues from neighboring beams, and you're going to have a very clean single beam that can create fast speeds, and that's what it solved," he explained. "The question becomes capacity, because then you only have one radio plugged in, and how do you scale that?" The argument is that when beams are added, squint becomes a problem. Massive MIMO cannot evade the fundamental laws of physics.
Even for throughput, it does not appear to have lived up to some of the grandest claims made in the early days of development. "Massive MIMO said it was going to get eight times the throughput, because we'd do beamforming that concentrated the energy," said Marcus Weldon, a former chief technology officer at Nokia, during a recent conversation with Light Reading. Beamforming is a technique intended to ensure a signal accurately pinpoints a device.
"The practical reality is that because it's so hard to do, it's two times," continued Weldon. "That's why massive MIMO has sort of dropped out of sight a bit, because the costs of the radios and the complexity of managing that beamforming didn't prove in reality to be manifestly better. Two times is good, but it's not worth the cost and expense – you might as well just deploy more spectrum on another band."
Such words seem like potentially helpful marketing collateral for the provider of a competing technology. MatSing's antennas can make do with basic RAN products that include as few as four transmitters and receivers per radio unit (so-called 4T4R). But the probable apathy of Ericsson and Nokia might present a hurdle. "It's one of those things where they don't make money selling 4T4R radios because they're commoditized," Lum previously told Light Reading. "And the multibeam antennas from Amphenol and MatSing only need a 4T4R radio."
On the plus side, a MatSing antenna can be easily paired with a radio from any provider. Someone doing the requisite joinery would not face the same difficulties that have beset companies trying to integrate parts from multiple suppliers in an open RAN. "So long as you're using standard 4T4R or 8T8R radios, it works," said Matytsine. "It's like a plug and play."
The dream for him or any of the other passive multibeam antenna makers would be a mainstream telco backlash against massive MIMO that translated into later-stage 5G or 6G opportunities. "Imagine entering a market where you're competing against the big OEMs that sell into telecom companies, and you're just a small business," said Matytsine. "It takes solving a unique problem, or the customer seeing that things aren't working." Stadium success for MatSing has garnered attention. But other things not working could take it into a much bigger field.
