Chips are holding back 5G private wireless

Private wireless seems to be booming, so it was a surprise to see an ABI Research report predicting that private 5G would not take off in 2023.

Many private wireless deployments to date are 4G, while 5G is still waiting for next-generation chipsets.

Leo Gergs, senior 5G markets analyst at ABI Research, said all the current 5G chipsets are still based on the 3GPP’s Release 15. They do provide enhanced mobile broadband. But important 5G features, such as ultra-reliable low latency communications (URLLC), were standardized in Release 16 in 2020, and there are no chipsets on that release yet. Devices with those chips will start to appear by the end of 2023.

“What I have been seeing in the last year or two is the moment you talk about private 5G for indoor applications – like industrial or manufacturing – what these deployments require is high availability and reliability and to some extent low latency, which is what the chipsets that are in the market at this moment cannot provide,” said Gergs.

He predicts Release 16-capable industrial-grade devices will be available at scale by the first quarter of 2024. And once that occurs, the large-scale availability will drive down device costs, causing 5G private wireless to pick up steam.

What devices need 5G chips?

The hardware for private wireless network infrastructure includes the small cell as well as end-user devices.

But end-user devices can include myriad things specific to the needs of different industries.

“I think that’s what is adding to the difficulty of designing these chipsets; they could end up in a number of devices,” said Gergs. “In a factory, they could be a robot or mobile work station, it could be a camera for surveillance or quality control.”

He said 5G phones do have URLLC chipsets “to an extent,” but industrial deployments need a different set of performance indicators because if connectivity drops for even an instant it could have dire effects.

For instance, a factory might present a hazardous environment with concurrent processes. “If you even lose connectivity for a second, a worker’s life could be at risk or production at risk,” said Gergs. “If you use [the devices] for preventive maintenance you could experience some down time and face some real revenue loss.”

The timeline

Based on experience from prior 3GPP releases, the industry was accustomed to waiting for about 15-18 months after a release was frozen for the first chipsets to emerge in the market. But things are taking longer now.

Of course, everyone is well aware of global supply-chain issues and especially shortages of semiconductors.

“Another aspect is: designing chips that meet industry requirements are proving to be more complicated than in previous releases,” said Gergs. “You have to have tight consultation with enterprises to learn what their requirements are and how to develop products that address these requirements.”