& cplSiteName &

What if Capacity Isn't AT&T's iPhone Problem?

Craig Matsumoto
2/22/2010

AT&T Inc. (NYSE: T) is beefing up its mobile backhaul network to answer complaints about coverage and reliability for the Apple Inc. (Nasdaq: AAPL) iPhone. But what if 3G capacity isn't the problem? (See AT&T CTO: We Know What's Best for You and AT&T Says It's Ready for Wireless Growth.)

An analysis published by Ars Technica on Sunday suggests iPhone coverage suffers, not because it hogs bandwidth, but because the phone is continually shutting down and re-establishing its data connection. That would mean the problem is in the signaling network rather than the raw capacity.

Ars Technica came to this conclusion by talking to a source at Telefónica UK Ltd. , the only operator that appears to be having problems similar to AT&T's. (See 02 Felt iPhone Crunch Too.) The source says the iPhone saves battery power by killing its data connection whenever it's not in use. The result is that the iPhone is continually requesting new data connections, overloading the signaling network.

This scenario would also explain why other carriers don't see the problem: Other European nations saw text messaging get adopted quickly, and operators built more signaling capacity as a result. More specifically, those operators used equipment that can shift more spectrum to signaling when needed.

Whether this really is the cause of AT&T's problems is still uncertain. The carrier didn't comment for Ars Technica. (Light Reading has a query in the works.)

But it's something AT&T and other operators might want to keep in mind, because Ars Technica notes that phones using the Google (Nasdaq: GOOG) Android or Palm Inc. webOS platforms use the same kind of battery-saving trick.

— Craig Matsumoto, West Coast Editor, Light Reading

(7)  | 
Comment  | 
Print  | 
Newest First  |  Oldest First  |  Threaded View        ADD A COMMENT
stevedc3
stevedc3
12/5/2012 | 4:42:19 PM
re: What if Capacity Isn't AT&T's iPhone Problem?


I don't understand why people are surprised and this wasn't getting press ages ago. This isn't iPhone specific either - it has to do with chatty & signaling intesnive apps.


Throwing spectrum around doesn't help these fundamental issues. I wrote about this in depth several times recently:


http://stevecheney.posterous.c...


http://stevecheney.posterous.c...


 


 

Disinto
Disinto
12/5/2012 | 4:42:19 PM
re: What if Capacity Isn't AT&T's iPhone Problem?


I wonder how can RRC states timers be tweaked in order to alleviate the problem, i.e. quickly switching to CELL_FACH state for non real time scarce traffic like IM/unfrequent app traffic. It may help saving some sig ressources. I however guess that these mechanisms are already in place in AT&T network.

Honestly
Honestly
12/5/2012 | 4:42:19 PM
re: What if Capacity Isn't AT&T's iPhone Problem?


Craig, can you be a little more specific.  Is the coverage effected by signaling and service time outs effecting data only, phone calls only, or both.?


 


Can you also explore and let us know what it could potentially cost At&t to fix the signaling problem, and how long it could take.


 


If true, At&t has basically lied to its customers.  Why would any one be shocked about that.


They will only talk with you If they can control the spin.  


 


They are a monster.  So much for Telecom reform.


Thanks


 


 

Honestly
Honestly
12/5/2012 | 4:42:15 PM
re: What if Capacity Isn't AT&T's iPhone Problem?


Craig, can you ask a Heavy Reading analist about cost and time of a potential fix.  I Think you are onto something important
thanks

Pete Baldwin
Pete Baldwin
12/5/2012 | 4:42:15 PM
re: What if Capacity Isn't AT&T's iPhone Problem?


Hi Honestly -- Whether the signaling factor is contributing to dropped voice calls, I don't know offhand. Depends on whether the voice network is using the same signaling, I would guess, and i'm not enough of a 3G expert to have that answer at hand. (Anybody know?)


As for the cost, it depends on how elaborate the 'fix' would be.  I'm sure there will be a few vendors willing to volunteer their thoughts.

OldPOTS
OldPOTS
12/5/2012 | 4:42:14 PM
re: What if Capacity Isn't AT&T's iPhone Problem?


"As for the cost, it depends on how elaborate the 'fix' would be.  I'm sure there will be a few vendors willing to volunteer their thoughts."


 


Signalling boxes come in two types.


One is a single box with a few expansions, so this can require additional load sharing box.


The second is a more expensive distributed box than can be expanded by adding processors and disks.


Note this app, besides routing, requires a lot of disk lookups.


So this is a complex networking issue.


 


Also a lot of this signalling can be carried on IP links, especially 3G traffic, with other traffic and the small signalling packets then encounter contention in the network with other larger Eth/IP packets.



 


OP





phil-tw
phil-tw
12/5/2012 | 4:42:11 PM
re: What if Capacity Isn't AT&T's iPhone Problem?
Just to declare up front: I have a vested interest here as I'm at Nokia Siemens Networks and we actually have a unique approach to this exact issue.

Craig's right: smart devices do very rapidly put themselves into an idle state to conserve their battery when they're not actually transferring data.

To keep the appearance of being 'always on' - for example to keep e-mail updated, or for IM presence - they initiate a signalling message sequence to the network every minute or every two minutes. This means that even when not being actively used, each smart device on a network can generate the equivalent to something like 1000 short calls per day. Not much data is transferred with each of these background updates, but the combined signalling load is huge. And if a network reaches congestion with the signalling then voice calls for other users (including those NOT on smart phones) can't be set up either. So, the operator either has to throw more signalling capacity at the network to compensate or find ways to reduce the signalling load.

One way this can be done by implementing the paging channel feature that's been in 3GPP radio standards and in every smart device for some time but only we (Nokia Siemens Networks) have built the feature into our radio network solution. The paging channel feature cuts the background signalling message sequence by over 80% and consequently cuts the overall typical network signalling load to one third.

And - as a very nice by-product - doubles the smart device stand-by battery life.

That's what smart phone heritage in a networking organisation can do: far-sighted R&D. We've shipped this feature in every new base station since 2007.

We blogged on this topic here and our CEO highlighted this in our press conference at Mobile World Congress, a recorded webcast of this session can be seen here
Featured Video
Upcoming Live Events
October 22, 2019, Los Angeles, CA
November 5, 2019, London, England
November 7, 2019, London, UK
November 14, 2019, Maritim Hotel, Berlin
December 3-5, 2019, Vienna, Austria
December 3, 2019, New York, New York
March 16-18, 2020, Embassy Suites, Denver, Colorado
May 18-20, 2020, Irving Convention Center, Dallas, TX
All Upcoming Live Events