San Diego-based Qualcomm has come a long way from its humble beginnings as a startup founded by a handful of Cornell University and MIT alumnus and University of California at San Diego professor Irwin Jacobs in 1985. Today the company, which is known for its wireless chipsets, employs about 33,000 workers worldwide and makes more than $22 billion in revenue annually.
The company’s chipsets are its primary business — it powers a large majority of the world’s smartphones and tablets. It also has a strong licensing division and a vast portfolio of patents that make up a large part of its revenue stream.
Qualcomm made a name for itself by designing the first code division multiple access (CDMA) cellular base station that was derived from U.S. military technology. The company used the technology to transmit digital signals over cell phone equipment. CDMA was later standardized and used by operators around the globe, including U.S. operators Verizon and Sprint. Qualcomm went on to work on several more generations of CDMA before switching to 4G LTE and now 5G.
But the past year has been a tough one for the firm. In November 2017 Broadcom made an unsolicited $100 billion-plus offer to acquire Qualcomm, which the company rejected, saying it undervalued its leadership position and its future prospects. Broadcom persisted, and even sweetened the deal to $121 billion. But ultimately the deal was quashed after President Donald Trump issued an executive order blocking the hostile takeover bid.
But that’s not all. Before Broadcom’s unsolicited bid for Qualcomm, the San Diego chipmaker had been in the midst of acquiring the Dutch chip company NXP Semiconductors for $44 billion. After Broadcom’s unsolicited offer to buy the company was nixed, Qualcomm continued to pursue NXP but struggled to get approval from China. In July it terminated the deal, paying a $2 billion termination fee to NXP.
5G and New Growth
Now that all that drama has subsided, Qualcomm is setting its sights on new areas of growth such as 5G, IoT, and the automotive arena. During a recent Citi Group Investor conference, Qualcomm CFO George Davis said that the company is banking on 5G, which he called a “very powerful force,” according to a Seeking Alpha transcript. “We are well ahead of our merchant competition in 5G …. And remember, to be good in 5G, you’ve got to be really good in 4G.”
Qualcomm is heavily invested in 5G, an area that is true to the firm’s roots. In a report commissioned by the company and conducted by IHS Markit the analyst firm predicted that 5G will impact virtually every industry section and could result in $12.3 trillion in global sales by 2035.
Not surprisingly, the company has been very aggressive in its work in spectrum aggregation, antenna technology, and modulation and coding, which are key to 5G deployments. The company has participated in interoperability testing for 5G New Radio (NR) with Nokia in the 3.5 GHz and 28 GHz spectrum bands. It’s also working with operators such as BT, Deutsche Telekom, SK Telecom, NTT DoCoMo, and Vodafone Group on 5G NR testing.
Plus, Qualcomm late last year released a 5G smartphone reference design for testing. The company has repeatedly said that it believes the first standardized 5G devices will be available in early 2019.
Servers or Edge?
A few years ago Qualcomm decided to expand beyond its core business and dipped its toe into the server market with the intention of competing head-to-head with Intel. In December 2016, the company started sampling its Arm-based Centriq 2400, a 48-core processor that was targeted at the enterprise market.
But Qualcomm’s move into that market was apparently short-lived. Last June there were rumors that the company might divest of that business and that it had laid off some of its employees. Qualcomm executives have maintained that they are keeping the Centriq product line, just retooling it.
According to Patrick Moorhead, president and principal analyst with Moor Insights and Strategy, Qualcomm no longer is going to go after the large Tier 1 and Tier 2 hyperscalers, and instead will redirect it energies to edge compute for service providers. “We [Moor Insights & Strategy] do believe that the carrier edge compute opportunity is large, but it likely doesn’t start driving revenue for a few years,” Moorhead said in an email to SDxCentral.
Of course, Qualcomm has always worked closely with the operators on its wireless device chipsets. And it continues to work closely with them on 5G chipsets and reference designs so it makes sense that the company would view edge compute as a natural expansion area.
“I have personally talked to carriers who say they will be investing in the space. The reason is simple — latency-prone apps like augmented reality, gaming, and self-driving cars require very low latency communications and computing to make it work well, and this can only happen at the edge,” Moorhead added.
The company has said its first chipsets with its wireless edge services will be available in the second half of this year.
IoT is another big area of focus for Qualcomm and one that the company believes will have big return on investment. According to Qualcomm spokesman Ignacio Contreras, the company expects to have more than $1 billion in revenue from IoT this year. But much of the IoT focus is on wearables and consumer devices such as voice-enabled home hubs, smart speakers, and connected appliances. Earlier this year the company told analysts during its analyst forum that it has more than 30 production-ready reference design platforms available for a network of original design manufacturers.
Analyst Mike Krell, head of IoT strategy with analyst firm James Brehm & Associates, said that IoT is a good focus for Qualcomm because processing of IoT systems is already distributed and as more IoT nodes are deployed this is going to become even more critical. “Qualcomm is well positioned to take advantage of both computing on the IoT device itself, as well as the aggregation points like IoT gateways,” he said.
As part of its IoT strategy, in April Qualcomm unveiled a new IoT platform called Vision Intelligence. This platform, which is based on a system-on-chip (SOC), uses an artificial intelligence engine coupled with a heterogeneous compute architecture and an Arm-based CPU to make it possible to do processing at the network edge instead of in the cloud.