Broadcom's new LTE modem plan

MESA, USA: From a wireless chip standpoint, Broadcom's acquisition of Renesas Mobile's LTE modem properties has to be our lead story. It is ironic that the Renesas LTE modem was the only solution certified by five major carriers and could never get a cellphone socket.

Broadcom stated that Renesas Mobile really did have a tier-one socket lined up, but their parent company's financial difficulties precluded them from completing the transaction. That coincides with Renesas earlier statement to me at MWC that they really did have a tier-one socket that was "soon to be announced."

Broadcom is getting a multimode 2G/3G/4G/LTE modem with RF transceiver that is already validated at five major wireless carriers (AT&T, NTT DoCoMo, Orange, Vodafone and EE). In fact, multimode LTE modems from both Renesas and Qualcomm were the only ones employed by test equipment houses for their own test and calibration for almost two years.

So the question arises: If the Renesas modem was so great, why didn't it get into any cellphone sockets? What's more, what makes Broadcom think that they can get the same modem into any cellphone sockets? I don't have the answer, but I suspect that poor marketing by Renesas had something to do with its poor showing. Hopefully, Broadcom will be better at marketing.

If I correctly understood the CEO's comments on the acquisition, the current LTE development effort will continue in parallel with fielding the acquired LTE product. The current LTE modem development is based on Broadcom's earlier acquisition of Beceem and their CEVA-XC DSP-based OFDM product. The Renesas LTE modem, developed some time ago by Nokia ASIC designers, is based on custom programmable processors, accelerators and a small ARM core for handling the low level stack.

The CEVA-XC DSP approach is apparently aimed at a superior future LTE (c.2015?), since it is a newer processor architecture and better long-term processor support by CEVA is likely (perhaps constituting a separate developer support team).

The company's 2014 roadmap is good, with the addition of carrier aggregation, envelope tracking and TD-SCDMA, with the latter necessary to be successful in the China Mobile market.

Broadcom promises to field its acquired LTE asset in a commercial product in Q1/14. We think that's achievable, but it may be in a Tier-2 smartphone, since the Tier-1s seem to be spoken for at the moment. But, like Intel, they have to start somewhere if they want to take on the 800-pound gorilla of the LTE modem market.

The author is president and principal analyst, Forward Concepts, USA.