Knowing when to say when

European wireless operators had it right back in the day when they decided collectively to coalesce around a single technology: GSM. It made life simpler for everyone: the consumer, the operators, the network equipment providers, the regulators and perhaps most of all the test companies, who could build product to a single standard.

“There is definitely value in having fewer standards from a vendor point of view because it is very clear what needs to be done,” said Donna Bastien, marketing manager for OSS at Agilent Technologies.

However, right or wrong, that moment has passed. European operators, like those in North America and elsewhere always have, are facing a host of new air interface technologies and applications from which they can choose.

“The reality is,” Bastien said, “There are a lot of small wheels moving very fast down the road — each one being a different protocol or capability — and the more wheels you have, the harder it is for a vendor to decide which one to pursue.

It's not quite as madcap as chasing a wheel of cheese down Copper's Hill in Gloucestershire, England, but picking the right technology could make the difference between winning and losing in the test and measurement market.

From the vantage point of four test companies — Andrew Corp., Agilent, Comarco Wireless Test Solutions and Spirent Communications — we will see that the wireless universe looks pretty much the same in all directions. Picking the right technology can be tricky, but these companies all seem to agree on which new technologies to invest in: HSDPA and EV-DO. And they seem to agree on what technology not to invest in (just yet): WiMAX. The difference between these companies is more a matter of when to invest, not what to invest in, although the ‘what’ is still very important. And determining the proper timing may be trickier than picking the right technology.

“The answer would be different for smaller companies, but fundamentally for everyone, we are driven by what our customers are asking us to do,” Bastien said.

The problem with that, she added, is that there are always more opportunities than there is the bandwidth, or resources, to go build things. “The reality is, it is a constrained system. There is only so much R&D investment we can make; there isn't an infinite source of people or expertise and we, like any business, have to make certain calculations,” Bastien said.

Agilent makes calculations based largely on what its customers ask for, although with a bit more restraint than they have in the past. Comarco picks a sweet spot at the close of trials on a new technology and just before deployment. Andrew uses a five-step process for gating its decision to invest in new technologies, which is based largely on availability of testable commercial products. And Spirent's device testing division tries to be first-to-market with its gear for testing new technologies.

Is one method better than another? No, at least not between these companies or divisions of companies. Each is addressing its own niche (with only slight overlap) and each company has to make its own decision.

“It is one of our toughest questions,” said Nigel Wright, vice president of client services for Spirent. “It's a huge problem because not only are the air interfaces mushrooming, but within each interface the scope and depth of testing required gets more involved.”

Wright works in the handset test division of Spirent where the company enjoys a leadership role in the CDMA market. Wright says one of the best ways to determine the right technology and the right time is to trust the relationship with existing customers, which is good, provided one has existing customers with a wining track record. Fortunately for Spirent, it does.

“With some [customer] relationships we get the inside track on what they are thinking 12 to 24 months out. That enables us to make some pretty informed investment decisions,” Wright said.

The sharing of roadmaps can help a company judge the risk of investing in new technology, Wright said. However, sometimes you have to make your own calls. One carrier customer was pushing Spirent to support TDS-CDMA, a low chip-rate version of W-CDMA in China for which hype was in high gear.

“But our own on-the-ground experience told us it was not going to be the dominant player. Customers wanted us to invest, but we didn't see the evidence,” Wright said.

Spirent also held back on WiMAX investment despite already having success with a fading simulator for the technology. “There was more hype than substance in terms of what was going on in the R&D labs,” Wright said.

Much of Spirent's wireless R&D budget, which Wright says is at the high end even for a technology company, is going into maintaining its presence in the CDMA market and developing tools for W-CDMA and HSDPA. “You can't ignore the emerging opportunities,” he said.

While CDMA customers are trying to drag Spirent into the application-testing arena now that the air interfaces are mature, Wright says the trick to allowing oneself to be dragged there is to time the market.

“If we offer a unique and timely solution, say when a new handset is at the design and verification stage and you are first to market, obviously you can achieve some high-margin objectives,” he said.

Timing, however, is a little trickier for companies such as Andrew and Comarco, which try to time their products to coincide more with actual service launches. For Comarco, it helps that building support for new air interfaces takes place 90% in software and only 10% in hardware.

The company specialized in optimization and quality of service testing. It admittedly doesn't have the resources of an Agilent that would allow it to get involved in every new technology, however it has placed its bets firmly with HSDPA and EV-DO.

“EV-DO has been born and it is kicking butt,” said Francis Sideco, product manager at Comarco Wireless Test Solutions. “And HSDPA is the only technology right now that I would say is in the ninth month of the birth cycle.”

Still, timing is most important. Comarco tracks new technologies through the standards bodies, but mostly tries to get a feel for what operators are serious about and where their money is going.

Wireless growth

2 billion	total number of wireless subscribers by the end of 2005
735 million	number of mobile phones sold in 2004
45 million	number of 3G subscribers by the end of 2005
113	number of operators to date who have launched 3G services
4%	rate at which wireless voice services are growing year over year
30%	rate at which wireless data services revenue is growing
Source: Spirent Communications