Blog

Comment


mike.keeley

8/26/2010 5:11 PM EDT

Thanks for the insightful comments.

Dan, I agree. Some of my ...

More...



Dan Mitten

8/20/2010 1:50 PM EDT

I have never found short cuts in testing to make economic sense. Correcting ...

More...

Getting LTE device design right

Mike Keeley, Director of Product Management, Spirent Communications

8/19/2010 3:31 PM EDT

Right now, several cellular network operators are hurtling towards “LTE (Long Term Evolution) launch day”. The upcoming large-scale launch of commercial non-proprietary LTE is our industry’s most ambitious initiative since the first cellular networks were launched years ago.

Designing mobile devices for LTE is very different from previous mobile design efforts, mainly due to three key factors:

  1. Complexity – LTE brings major, not incremental, changes to every link in the network from the core to the mobile device.
  2. Multi-technology support – Subscribers expect voice and data applications to run seamlessly as they move among networks or service areas. As a global standard, LTE drives demand for devices that simultaneously support CDMA, UMTS and, of course, LTE.
  3. The popularity of data applications and the demand for faster data rates.

Multiple-In Multiple-Out (MIMO) technology will significantly impact the success of LTE. It is a “game-changer’ with the potential to multiply data rates without incurring additional spectrum costs.

But when MIMO is used, the device’s physical orientation directly affects the system’s capability. The system must quickly adjust whenever the device is moving. Over-the-air (OTA) testing of device designs ensures that the system adjusts properly and quickly enough to realize the “headline” data rates promised in LTE.

Next, what happens when a subscriber launches a data application using LTE service, then moves out of range into a region covered only by, say, CDMA? Finding out requires mobility or Inter-RAT (Inter-Radio Access Technology) testing, which involves multiple emulated radio networks working in concert to provide realistic repeatable scenarios.

Since data services are a big part of the promise of LTE, data testing must be done with realistic network conditions. This includes not only Inter-RAT, RF fading, and MIMO scenarios, but adverse-condition (e.g. overloaded network) scenarios as well. In addition to throughput testing, data testing requires “data retry” testing, which ensures that a network won’t be overrun by service-connection requests when, for example, a network-based server goes down.

LTE is one of the few large-scale technologies to be deployed before the device certification process is finalized. Because of this and because of the new pitfalls outlined above, network operators and mobile device manufacturers alike are adding whole new areas of interest to their device test plans. Base largely on lessons learned from other large-scale wireless deployments, these new test plan philosophies are here to ensure that LTE delivers on its promise.

About the author
Michael Keeley is a Director of Product Management at Spirent Communications’ wireless test equipment division. He has led various teams involved in wireless network emulation and automated systems used for testing mobile devices. Prior to joining Spirent in 2000, Mike spent over two years at Lucent Technologies. He received his BSEE and MEng from Cornell University.





IqbalSingh.Josan

8/19/2010 4:14 PM EDT

In addition to UMTS, CDMA and LTE, users will also expect seemless operation with WiFi hot spots. For example, when a user is within the range of a WiFi hot spot, LTE is not needed. Therefore, a critical success factor for LTE devices will be how well they manage the transition from one wireless interface to another and it does not have to be annoying to the user. There are new and unprecedented challenges to test mobile devices that can interoperate over multiple wireless technologies.
Regards.
Visit us at uspurtek.com

Sign in to Reply



Dan Mitten

8/20/2010 1:50 PM EDT

I have never found short cuts in testing to make economic sense. Correcting problems in the field is terribly expensive! The wireless industry should never be without a serious trial deployment, for each technology.

Sign in to Reply



mike.keeley

8/26/2010 5:11 PM EDT

Thanks for the insightful comments.

Dan, I agree. Some of my customers have done their homework and quantified the relationship between proper testing and profits. Obviously, their conclusions are proprietary and I can’t share them, but the numbers would astound you. Consider all the areas impacted by a bug in a released device: customer support, subscriber revenue (renewals and new customers), and even the operational network-side costs of supporting marginally-performing devices. Multiply that number by hundreds of thousands and you realize that you are talking about significant amounts of money.

Iqbal, you raise an interesting point, but I’m afraid I’m not very closely involved with LTE/WiFi mobility. It does raise some interesting questions, though… for instance, will an LTE operators be willing to hand a viable call over to a home network, with little insight or control of its success? Should they, knowing that failures will cost them in customer support resources? Thanks for the idea.

Sign in to Reply



Please sign in to post comment

Navigate to related information

Datasheets.com Parts Search

185 million searchable parts
(please enter a part number or hit search to begin)
Jobs sponsored by

Feedback Form