More people spend more time doing wireless design in Europe than anywhere else in the world, according to the chief economist of STMicroelectronics (ST), Jean-Philippe Dauvin. “Thirty five per cent of the world’s wireless design is done in Europe,” states Dauvin.
A figure that Eric Janson, senior vice-president at Cambridge Silicon Radio (CSR) is not surprised by. “If I think of all those Texas Instruments designers at Sophia Antipolis, and all those at Infineon and ST, and at Lund and Kista in Sweden, and in Finland, and at Aalberg in Denmark, then I’m not surprised.”
Jerry Loraine, CTO of SiGe Semiconductor, says he would add “Grenoble and Caen in France, Philips in Eindhoven, and Cambridge and Bristol in the UK” to the list.
These clusters are self-seeding. “We have a design centre in Sophia Antipolis because there are good engineers there,” says Janson. “The universities have caught on to wireless and have good courses. We’ve also got a design centre in Lund in Sweden for the same reason.”
But what exactly does Europe bring to wireless?
According to Richard Traherne, head of the wireless business unit at Cambridge Consultants (CCL), there is knowledge of the system design of chips and wireless systems.
“That’s what influences a lot of the work we get and win. There are a lot of silicon companies outside Europe capable of taking a spec and turning it into a chip,” says Traherne. “But inside Europe we have the system knowledge, with people who can think across the disciplines and know how to partition the product, how to have different variants, how to design it so it works with legacy equipment. It needs people with knowledge of control electronics, and sensor electronics and manufacturing techniques.”
CSR has experience of hiring engineers from the US and the Far East, and thinks Europe has important advantages.
“One is that the basic electronics environment in UK, Scandinavian, German and French schools is much stronger in wireless design,” says Janson. “The essentials of communications theory are not simple concepts, and the schools do a good job in preparing people for the challenges of design.”
Janson believes another advantage is the supportive environment for designers in European companies. “That’s in contrast to Japan, the Far East and the US, where the companies are focused on executing in a timely fashion. Good RF just can’t be rushed,” explains Janson. “European companies are prepared to spend more time in getting it to work than trying to rush it through and re-patch several times later. The European approach is a more deliberate and thoughtful approach in the basic planning stage of IC design.”
Jamie Urquhart, general partner in venture capital firm Pond Ventures and a co-founder of ARM, believes RF designers are more conservative than digital designers. “Possibly because making RF chips work is much more difficult than getting digital designs to work. There’s a lot of confidence in the UK that we can produce innovative and successful design. RF design needs bold steps to be successful.”
CCL’s Traherne thinks it is not just because “we’re in Europe” but also because they are “talking the same language”.
“We don’t have to ask: ‘What’s the input noise on that pin?’,” he explains. “We solve the problem at the system level rather than at the detailed level and work out the input noise on that pin.”
Another reason for Europe’s strong position in wireless design is its failure in computers. “Wireless is analogue. Europe has gone to analogue because of its lack of success in the computer industry which is all digital. We haven’t been able to go down the easy street of digital,” says Steve Collis, technical director for Europe at Mentor Graphics.
A key ingredient to Europe’s success in wireless is its attitude to standards. “When your radio wakes up and starts sniffing around the airwaves looking for a signal, it finds a lot of signals and that has forced standards bodies to respond with more meaningful, robust standards,” says CSR’s Janson.
“Wireless is built around a raft of standards many of which were drawn up by Europe - Bluetooth, GSM, wireless LAN - with the systems expertise coming out of European companies,” says Mentor’s Collis. “Europe is more consensual in creating standards. It’s different in the US where the same desire to create standards is not present, with each company wanting to become dominant.”
This is a sentiment which CSR’s Janson agrees with. “If you look at the actions, not the statements, of US companies, a lot of US standards work is very contentious and competitive,” he says. “Look at 802.11n, (MIMO-enhanced WiFi), there’s still no standard, but companies keep coming out with pre-standard products to get into the market.” CCL’s Traherne puts it as, “we toe the line in Europe”.
As is to be expected when something is attractive, there are a few bees buzzing around the honey pot and in this case the strength of European wireless engineering has attracted the venture capitalists (VC).
“You could argue that wireless start-ups in the UK have a better chance of success than a wireless start-up in Silicon Valley,” says Pond’s Urquhart. “Many of the customers are based in Europe, and there’s nothing like having large companies around which can give vital feedback when they are involved in the early stages of development.”
However CSR’s Janson thinks a US start-up still has an edge on a European start-up.
“US companies find it easier to tie into the VC community and get going because a lot of OEM design of equipment gets done in the US. It gives them a leg-up in getting early design-ins. The chip’s probably not going to work, but they get early experience of getting to make it work,” says Janson.
“In Europe, the wireless OEMs are large, and well-established, and find it more difficult to go with a start-up because of issues like supply and patent protection, and they’re less inclined to go with a start-up than the US companies.”
European OEMs can be sensitive about the extent of their IC design activities. Asked how many chip designers it has, a Nokia spokesperson replies: “Unfortunately the answers to your questions are considered business critical information and therefore the information is classified as company confidential.” The number could be huge. Two years ago STMicroelectronics revealed it had some 300 design engineers working exclusively on chip designs for Nokia.