LTE Signaling Woes Ahead?

The threat that some smartphone operating systems and data applications could flood 3G networks with signaling traffic has been making headlines recently, but 4G networks look likely to up the ante on control plane issues.

Signaling has become a hot topic with operators, vendors, editors and bloggers as the growth of data on 3G has changed the nature of traffic on the network. Voice calls require a narrow control channel for setting up and maintaining a conversation, but some data applications can flood the narrow signaling channel by constantly pinging the network to maintain the session.

Light Reading Mobile has highlighted work from Nokia Networks that indicates Android and BlackBerry 's OS are "signaling bad boys" because they don't yet support a 3rd Generation Partnership Project (3GPP) standard that recognizes when a device is idle and reduces network chatter accordingly.

Other publications and blogs have been looking into how current and future networks and applications will handle signaling traffic:

They say
The evolution of data traffic on 4G Long-Term Evolution (LTE) networks should be a concern, argues Richard Quinnell at Test and Measurement World. Repeated data connections on the signaling channel "can overload an LTE network's control plane long before straining the network's communications bandwidth if enough smartphone or tablet apps are placing frequent, low-traffic calls," he writes.

Industry people suggest the little-known Diameter protocol will help deal with the complexity of signaling traffic created by always-on data devices on LTE networks. "Diameter is the only signaling protocol that is capable of managing the constant flow of core network signaling in an environment that has become far more complex with many more network elements needed to fulfill the promises of 4G," writes Susan Becker, marketing manager at Traffix Systems on the snappily-titled Diameter Blog.

Certainly, Tekelec has started to get some traction with its commercial Diameter Signaling Router (DSR), recently winning an LTE deal with an unnamed "tier one operator" in the U.S. Tekelec says its DSR "provides up to 500,000 Diameter messages per second in a single frame." (See Tekelec Wins LTE Deal.)

Long-time telecoms guru Andy Seybold, meanwhile, reminds developers in a column at Fierce Mobile Content that dealing with the signaling problem shouldn't be all down to the operator. App creators have a responsibility to ensure that chatty apps don't flood the signaling plane too, he writes:

    What should developers do to minimize the impact of their application on a network's signaling channel? The first thing is to look at other ways to update your customers' information. Can you do it with push technology that does not use the signaling channel? If you cannot, then look at how often the data really needs to be updated. Don't assume that you have infinite bandwidth, and understand that you are sharing it with many others. Write your applications accordingly.

Signaling traffic spikes can already be seen to have caused some serious problems on mobile networks. The "News and Views from Norway" website reports that in Norway this June, Telenor Group (Nasdaq: TELN) experienced a major outage, likely because of signaling spikes when it rebooted some servers on the network.

We say
Trace the growth of the signaling issue on LR Mobile below:

— Dan Jones, Site Editor, Light Reading Mobile

Page 1 / 2   >   >>
joset01 12/5/2012 | 5:00:36 PM
re: LTE Signaling Woes Ahead?

Make the control channel in LTE wider? Or is that too easy?

sarahthomas1011 12/5/2012 | 5:00:35 PM
re: LTE Signaling Woes Ahead?

That would definitely help, but I agree that developers need to create more network-aware apps. I've suggested this before and most people have said the problem is entirely the operators. It is, primarily, but it seems like there's more apps makers can do too.

krishanguru143 12/5/2012 | 5:00:35 PM
re: LTE Signaling Woes Ahead?

How many carriers and vendors were stating that LTE would reduce the signaling load that has plagued 3G?

lshannon 12/5/2012 | 5:00:25 PM
re: LTE Signaling Woes Ahead?

Hi, I'm fron Nokia Siemens Networks, and making sure that signalling isn't a problem in LTE is one of the things that we're working quite hard on at the moment.  The author of the article is right in that Diameter does offer scalable signalling support and its widespread use today will only expand under LTE.  However, Diameter addresses signalling in the core network, and where the real killer volumes of signalling need to be handled is in the radio network.  Diameter doesn't help there. 

The LTE specification was designed with the 3G experience fresh in everyone's mind, so the signalling learnings that we've been experiencing with smartphones have certainly been taken into account and LTE networks have been designed for much more efficient signal handling.  However, we're not going to know for 100% sure that there won't be a signalling issue in LTE until we see what the handsets do.  After all, signalling wasn't an issue in 3G until the smartphone manufacturers came up with their specification-defying proprietary Fast Dormancy software, which is what has blown 3G signalling out of the water for all of us.  If the handset vendors take a similar approach in their build of LTE devices, well . . . here we go again?

manju.kori 12/5/2012 | 5:00:25 PM
re: LTE Signaling Woes Ahead?

LTE has been designed to cut down the signalling load in the network, by having some of the signalling messages encapsulated into other messages. But its up to the developers to make sure that they don't pump in the extra signalling traffic by wisely utilizing the benefits that specifications have provided.

digits 12/5/2012 | 5:00:24 PM
re: LTE Signaling Woes Ahead?

Good points.

If I was an operator, though, I'd be presuming that the handset and apps folks are not going to be thinking about how I am going to manage the traffic on my networks - I imagine the "it's a network problem - let the network guys sort it out' mentality will prevail.

lshannon 12/5/2012 | 5:00:23 PM
re: LTE Signaling Woes Ahead?

Yeah, though it can be frustrating trying to keep the network boat afloat when the other occupants are busily pounding holes through the bottom of it.  But you know, if the network boat goes down, the handset and app guys go down with it, too.  I don't know how many more signalling-driven major network failures (Telenor Norway just last month, for example - an 18 hour outage?  Ouch!) it's going to take for the industry to realise this.

krishanguru143 12/5/2012 | 5:00:13 PM
re: LTE Signaling Woes Ahead?

Isn't that what the standards body should be doing?  if the phone manufacturers want to go outside the standards, then they standards body should be denying them from using the term LTE.  Much like the WiFi side does; don't meet the standards, can't say it is compatible.  That would be a much better solution as the carriers selling the branded handset would be looking at handsets that meet the standards they are using.  Sure you might get the unlocked handsets not sold to providers, but how many major handset manufacturers will be selling non-complaint handsets?  Then all you have is the knockoff models, but even then, the hardware is probably more or less using off the shelf components.

lshannon 12/5/2012 | 5:00:12 PM
re: LTE Signaling Woes Ahead?

Good point about the standards body, but I'm afraid that's probably not the answer.  Handset-proprietary Fast Dormancy software came about in 3G when all the major handset manufacturers took advantage of a loophole in the standard.  So it's not that they're not compliant, it's just that they're doing something that the original standards writers never imagined and did not prevent.  (It reminds me a bit of the situation in the early 80s, when the drug ecstasy was legal because it was new and lawmakers hadn't gotten around to making it illegal yet.)  The problem now is that **all** of the major smartphone manufacturers use Fast Dormancy, and what operator is going to voluntarily ban Android phones from its networks?  The revenue gain far outweighs the network pain in the long run. 

Now we just have to figure out how to minimize that network pain part.

krishanguru143 12/5/2012 | 5:00:08 PM
re: LTE Signaling Woes Ahead?

The carriers do test the phones and then request changes to them.  Many phones have a branded firmware that is customized for each carrier.  The carriers well could have rejected the phones and the manufacturers would have gone and made the changes.  Google would have fixed it in an instant, they want people to use the phones and that doesn’t happen unless it is available to the users.  It took the carriers to get Android where it is in the marketplace today, not Google.  Google tried the bypass the carrier route and it didn’t work out for them as many wanted a subsidized phone.  How many carriers would have wanted fast dormancy removed?  Probably almost all of them.


There will always be companies trying to use a loophole for a competitive advantage and that is where the standards body comes in.  They cannot prevent it, but they can bar them from using any trademark or saying it is compatible.  The GSM Association could have declared those phones as not being able to be called GSM, UMTS, HSDPA, etc.

Page 1 / 2   >   >>
Sign In