Raouf Halim, CEO of chipmaker Mindspeed Technologies Inc., has worked for three companies in the past 21 years, but, as he puts it, his desk hasn't moved more than 100 feet.

The 52-year-old has been carried along by a succession of three technology spinouts, all of them aimed at returning a degree of focus and nimbleness to the companies he's worked for. Halim started in the Newport Beach, Calif., offices of electronics giant Rockwell International in 1991 and landed at its communications chip unit Conexant Systems Inc. when it was pared from its parent in 1999. Just months after its spinout, Conexant launched a raft of acquisitions in an attempt to become a one-stop communications chip shop. But that again blurred the focus of the company, which was then led by Conexant CEO and chip industry luminary Dwight Decker.

Decker hatched a plan to break up Conexant into four parts: specialty wafer maker Jazz Semiconductor, which was acquired by Tower Semiconductor; publicly traded wireless chip specialist Skyworks Solutions Inc.; Conexant, which was taken private last year in a $197 million buyout by Golden Gate Capital; and Mindspeed, which Halim has headed since 2003.

Halim has helped build Mindspeed into a feisty, publicly traded midcap semiconductor player with a strong position in componentry for voice over Internet protocol and optical networking. But the company has long aspired to grow its wireless offerings in a way that dovetails with its existing technology and the customers that buy it. Mindspeed made its first big move to do so this month with an agreement to acquire Picochip Ltd., a private U.K. maker of chips for cellular base stations. The deal, including an earnout, carries a price tag of $77 million. It was handled by Halim and involved a challenging set of negotiations with the target's investors that lasted more than a year. Halim spoke with The Deal magazine's Olaf Domis about Mindspeed's growth strategy and the Picochip deal.

The Deal magazine: How significant would you say is the Picochip deal to Mindspeed strategically?

Raouf Halim: It's the one and only company acquisition Mindspeed has done since being public. Mindspeed, as well as all the spinoffs, have some DNA in the wireless arena because of Conexant and Rockwell, but the bulk of it went to Skyworks. We're a leader in VoIP technology -- services like Skype and others are based on Mindspeed core products. We also branched out into optical transport and other fiber-based areas. But for many years, we've been eyeing the immediately adjacent wireless base-station market. We started evaluating whether to develop this in-house or partner, and then Picochip bubbled to the front.

How did the deal come together? You acquired it for significantly less than the $110 million or so in funding it received.

To say the process was challenging would be an understatement.

We've been talking to them since late 2010. Throughout 2011, it wasn't really clear they were open to selling the company. The investors were examining a range of options, everything from an IPO or recasting the company or a sale. They had certain expectations on the value of the company, but there wasn't a good marriage of where they wanted to go with the valuation and where we wanted to go with it.

But it became clear to investors the company would require a significant round. They didn't have [a long-term-evolution] platform. They could roll the dice, do another round and dilute earlier investors, or they could sell to someone who would give them cash but also a piece of the upside.

Our market cap isn't very big, so we ended up giving them about 15% of the company's equity.

Who ran the negotiations?

We don't actually have an M&A team. From a deal perspective, it was the CEO -- yours truly -- who did all of that. It was extremely time-consuming, especially this past November and December. Picochip was down to a Series F. The Series A, B and C guys were not necessarily in sync with the Series F guys. Then there was the management team. There were many times I didn't think this deal would get done.

They had a couple strategic investors involved, like Intel and Samsung, and sometimes it's the nonfinancial guys who have the most tricky needs in a transaction like this. They might have their own strategic internal plans, or see value in the assets you're trying to buy.

Did the earnout help break the logjam?

They had one view of what their business and revenue growth would look like, but when you conduct due diligence, you have a different view. The earnout was something that resolved that.

How do you keep Mindspeed from spreading into too many areas?

It's a constant balance, and a particularly critical question for us because of our scale. We did $160 million in revenue last fiscal year. We're not huge -- $50 million for a research and development budget is not a big number. We're a midcap player.

But this deal brought us 150 people, largely engineers, who are very wireless-literate. This shot of wireless DNA, it was paid for, essentially. They have a significant revenue stream and growth. Once you take out the cost synergies from this deal, the R&D essentially pays for itself.

Picochip has exactly the same customers that leverage Mindspeed's wireline products, so we already have a great channel into these customers, which include Alcatel-Lucent, Nokia Siemens, Huawei and all the Japanese OEMs.