Spare a Dime for Analog?
4:45 PM -- Analog engineering is hard. Digital isn't always easy, but it's more methodical; analog tends to be a black art, based on gut feel and experience. So, it's not surprising to hear Sanjay Jha, former CEO of Motorola Mobility, say that analog chips -- antennas and the analog front-end -- would be good territory for startups to mine when it comes to LTE-Advanced. Jha was on an LTE-Advanced panel at Thursday's Trend Spotting2 Technology Summit, put on by Cowen and Co. (The "2" is there because it's the second time they've put this on.) LTE-Advanced is the next step in 4G wireless, of course, and it lets a mobile device combine up to five carriers (a "carrier" in the sense of one radio signal, or a chunk of spectrum) for speeds up to 100 Mbit/s. (See Meet the Next 4G: LTE-Advanced.) And it sounds like it's going to come with some difficulties, particularly as more combinations of carriers come into use. The point wasn't to discourage the audience of investors. In fact, it was the opposite: showing them that LTE-Advanced presents questions that new companies could try to answer. "It is possible to continue to make money in wireless business, and therefore, it is worth taking all the risk and all the complexity this industry is facing," Jha said. But to his point about chips: The smartphone is dominated, in terms of physical layout, by what Jha grouped as "passives," meaning, antennas and front-end analog components. These items can't be integrated yet. "If I were looking to invest a dramatic amount of money to improve the performance of a handset right now, that's where I'd look," Jha said. It sounds like a good theory. But there's something else to consider. At the same conference, I ran into Yoon Choi, who's in corporate business development for Maxim Integrated, an analog chip company with a market capitalization near $9 billion. She's helped start a venture-funding division of Maxim, partly because venture capital for semiconductor startups has been so difficult to come by. Hardware remains a hard sell among VCs. It's true that there are interesting problems waiting to be solved, but I'm not sure that means something like analog chip design is going to get the funding it deserves. Jha had a good point about the potential, but it's more likely that established companies would have to foot the bill for the innovation he's seeing. In fact, he mentioned a second LTE-Advanced category that probably has more venture-funding appeal: the management of all those small cells that are going to be scattered around the network. — Craig Matsumoto, Managing Editor, Light Reading