Industry Voices—Madden: Can Huawei survive without American chips?

Joe Madden

Huawei executives had a bad week. Their supply of American technology has been cut off. There’s been a temporary reprieve to support existing networks and existing handsets, but new production is still cut off. For ZTE, a similar action was nearly fatal last year, and ZTE only survived because President Trump granted a stay of execution.

Huawei is a stronger company than ZTE. Huawei makes its own ASICs and baseband processors, and has armies of engineers with deeper capabilities. Here’s our diagnosis of their condition:

Mobile Infrastructure:  (40% of revenue)

  • HiSilicon can deliver application-specific integrated circuits (ASICs), but they rely on ARM for processor core technology. Without a renewal of licenses from ARM, they are open to legal action in Europe, Latin America, and other countries that use Huawei equipment.
  • Field-programmable gate arrays (FPGAs) are an important part of 5G base stations, because they offer the programmable logic that original equipment manufacturers (OEMs) need. Instead of hard-wiring everything into an ASIC, the FPGA can adapt as the OEM learns about new stuff like beamforming. Both major FPGA suppliers are American, so any work-around that Huawei tries will be inflexible and awkward.
  • Mobile infrastructure relies on integration of high-performance analog-to-digital (ADC) and digital-to-analog (DAC) converters with radio processing. U.S.-based Xilinx is in position to supply this. Huawei/HiSilicon can integrate ADCs and DACs in their ASICs but our understanding is that their performance is low for dynamic range over 5G bandwidths. Translation for non-techies: They can do it, but they will need a waiver from the Chinese government to ship it to the field. The Europeans won’t like that.
  • RF device vendors exist in Japan and Europe, so Huawei can probably get the specialty radio parts that they need.

Handsets/Consumer devices (50% of revenue)

  • Huawei uses their internal HiSilicon team for most baseband modems, but they also buy from Qualcomm because HiSilicon lags behind the leading edge. HiSilicon launched a ‘multimode’ 5G modem in January 2019, but it relies on ARM cores, so outside of China they’ll have problems with licensing the key IP.
  • RF devices in the handset are dominated by American companies. Broadcom, Qorvo, Skyworks, and Qualcomm offer the best options for small size, large numbers of bands, and long battery life. Murata (Japan) is the only option left with the capability to integrate a complete RF front end in a compact size, and Murata has limited capability in the high frequency bands. No Chinese RF supplier has the ability to integrate 20+ bands in a small module.  
  • Lack of support for Android updates and the Google Play Store will be significant.   Nobody knows whether Huawei’s homegrown operating system is ready for prime time, but nobody has a ‘Play Store’ full of apps for the Huawei OS.  

The bottom line: We believe that Huawei can scramble and put together work-arounds, but their products will be janky. All-China base stations may not meet global standards for the fidelity of radio emissions, at least for the first year or two. All-China handsets will probably drop back to covering only 5-10 frequency bands, or will have to get 20% larger to support 20 bands.

I’ve seen a similar story play out before. In 1998, my company (Spectrian) decided to sell 2G amplifiers to a small unknown company called Huawei. I visited multiple times with demonstration hardware, and watched the Huawei engineers swarm over my product. We thought that we could sell them amplifiers for 2-3 years before they learned how to overcome the difficulties of high linearity and high efficiency. In fact, we sold our product for only a few months before Huawei dropped us, and started manufacturing their own unit. Poor performance? No problem. Huawei received a waiver from the Chinese government; they didn’t need to meet stringent linearity requirements after all. We decided that a lawsuit would be a waste of time, and we simply walked away.

This time around, a waiver from the Chinese government won’t cut it. Half of Huawei sales originate outside of China. Without ARM licenses and American chips, Huawei might survive in the domestic China market, but struggle for 2-3 years in global markets before achieving high performance again. Even then, it will be difficult for China, Inc., to reproduce American technology without violating American IP, and their legal battles for the world market could be unsustainable.  

Can Huawei catch up? Sure, they can spend the next 2-3 years to achieve the level of performance that ARM, Xilinx, and Qorvo can achieve now.  But those companies will continue to innovate, and by 2022 they will have better technology than they have today.

No matter what Huawei says, they are lacking key technologies, and their future is directly tied to negotiations with Mr. Trump.

Joe Madden is principal analyst at Mobile Experts, a network of market and technology experts that analyze wireless markets. The team provides detailed research on small cell, base station, carrier Wi-Fi, and IoT markets. Madden currently focuses on trends in 5G, IoT, and enterprise markets for wireless infrastructure. Over 26 years in mobile communications, he accurately predicted the rise of digital predistortion, remote radio heads, small cells, and a mobile IT market. He validates his ideas with mobile and cable operators, as well as semiconductor suppliers, to find the match between business models and technology. Madden holds a physics degree from UCLA. Despite learning about economics at Stanford, he still obeys the laws of physics.

"Industry Voices" are opinion columns written by outside contributors—often industry experts or analysts—who are invited to the conversation by FierceWireless staff. They do not represent the opinions of FierceWireless.