Boot camp for prospective Verizon wireless handsets

Only the best of the best can become Verizon Wireless (www.verizonwireless.com) handsets. The nation’s largest carrier, with 27.9 million subscribers, puts new handsets through the ringer before they can be sold to the public.

Lou LaMedica, Verizon Wireless director of CPE (customer premises equipment), evaluation and development directs 14 lab engineers at the carrier’s handset testing lab in Bedminster, NJ. LaMedica rarely rejects handsets; only six over the course of 13 years have flunked out, he said. The process is more of a give and take between Verizon and handset vendors.

“What we do is identify deficiencies and give them (handsets) back to the vendors or they will be here working with us and ultimately they will try to address them,” LaMedica said.

The team tests everything about the handset, from its propensity to drop calls to the wireless-Internet interface. The root of most problems, LaMedica said, is the handset’s software. For handsets’ wireless-Internet interfaces, the testing team primarily tests those sites on Verizon’s deck. Servers frequently make the handsets look bad.

“There is always some handset that has a problem with a feature in one of those servers that half the time turns out to not be the handset, but the server,” LaMedica said.

Stage I of Verizon’s handset testing begins with parametric tests that all handsets typically meet. Stage II involves interoperability testing and messaging compliance with the carrier’s infrastructure. Testing labs exist for Lucent (www.lucent.com), Motorola (www.motorola.com) and Nortel (www.nortelnetworks.com) infrastructure. Stage III runs through all the core performance and characteristics of the handset.

“In addition, we do a metrics route for call performance which baselines a product that we are looking at against a product that we currently sell,” LaMedica said. “If there is a lot of variation between those, we go back to the vendor and say, ‘you have a problem here, we think it’s in this particular area, tell us what you want us to do to help but get your people on it and come to a solution.’”

To test a handset’s audio performance, Verizon uses a dummy. Dubbed Mr. Head, the team uses a mannequin with silicon ears that closely replicate human’s hearing ability.

“A number of years ago we did not have a good way to verify audio performance with CDMA handsets,” LaMedica said. “We kicked it around and thought there’s got to be a better way than passing it around and asking everybody what they thought of the sound quality.”

Mr. Head measures things such as audio amplitude, frequency, noise and distortion. A test that mimics human speech goes into the handset and the team compares the results to the original.

These days, LaMedica is busy testing 1XRTT handsets. Thus far, they’ve performed surprisingly well.

“Considering that it’s basically a new technology, we are not having anywhere close to the same issues we had when we originally deployed CDMA,” he said. “In the past, when you had an issue, you’d have 20 PhDs standing around scratching their heads (wondering) what do we do next. Now, we’ve climbed the learning curve. It’s never routine, but problems tend not to be as big of a crisis as they were in the past.”

LaMedica said that early hiccups with 1XRTT phones have been in the call processing software, but added that those problems originate from the network as well as the phone. So far, LaMedica’s team has tested 1XRTT devices from Audiovox ( www.audiovox.com/ACC_home.html ), Kyocera ( www.kyocera.com), LG Infocomm (www.lgjoyphone.com), Samsung (www.samsung.com) and Sierra Wireless (www.sierrawireless.com).

“You are going to see a lot of 1X phones because I think anybody who got out of the business or stayed on the sideline sees this as the time to get back into it,” LaMedica said.