Voice and SMS over LTE

The Voice (over LTE) Double Whammy

Yes, the feared "Double Whammy".  It’s a technical term.  

There are two things set to impact the investment in IMS-based voice over LTE.  

One item is an undisputed truth, the revenue per minute of voice continues its interminable slide towards rock-bottom.  As industry pundit Dean Bubley posted the other day:

“The growing elephant in the room is the growing realization in the industry that mobile voice telephony revenues are declining.”  

The point, from Dean’s perspective, is that it will be nearly impossible to find a return on investment for a massive new IMS/VoIP system.  As he concludes to eloquently:

“What operator will want to invest in more expensive technology in order to provide a worse service in a declining market?”  

Whammy #1.

Whammy #2 is one I don’t think gets the attention it deserves – namely that the generation of people, but particularly those under 30, would much rather text, email, Facebook, whatever… with their smartphone than actually make a phone call.   

To make the point, Nielsen recently released its Q3 2011 Mobile Media Report on US mobile habits.    

Across all age-groups/gender, the average number of voice minutes/month decreased 12% between 2009 and 2011. 

In the same timeframe, the average texts/month increased 35% and 44% for woman and men respectively. 

Perhaps adding insult to injury, the word “voice” didn’t even appear in the Nielsen report until page 24!  

Maybe it's just not that interesting any more.  The first 23 pages were filled with statistics about all the great over-the-top apps people use religiously. 

 

 

 

LTE on sale

As we head into the "buying season" also known as Christmas, there's no doubt mobile products/services are going to be big sellers.  

Verizon is maintaining a commanding lead in the US LTE market and the global market in general.  Just last week, analyst firm Telegeography published that Verizon had an estimated 60% of the total LTE subscribers worldwide with 3.1m.  

It sounds like LTE is compelling for Verizon's subscribers.  So I was a bit surprised when I heard an ad saying LTE was on sale at Verizon:  

To be fair, it's just a little sale, a holiday promotion where new subscribers can get 2x the data for the same price.  This got me thinking... if I can get 4GB of data/month for $30 on LTE, what does that cost on their 'old' 3G (EV-DO) network.  Hummm...

The standard EV-DO data pricing as of Dec 2, 2011:

• 2Gb ->  $30
• 5Gb ->  $50
• 10Gb ->$80

LTE data is 50% off.  But wait... I thought LTE was compelling, super fast, wonderful, why would it be 50% the price of EV-DO? 

Verizon's LTE is available in ~130 markets in the US, whereas AT&T has 'launched' service in 15, so they can't be facing competitive pressures.  

I will posit one of the problems is battery performance.  Verizon's 'dual radio' approach to addressing the voice over LTE problem is sucking electrons... fast.  This was a known and acknowledged problem associated with a dual radio approach.  I wonder if CS Fallback will do a better job of inspiring consumer confidence.

 

 

 

 

CS Fallback's slow call setup times

About two years ago, Kineto published an analysis of CS Fallback and, among other things, came to the conclusion that call setup times would be about 50% longer than a ‘standard’ UMTS call setup times.  

Kudos to Zahid over at 3Gand 4G Wireless Blog for recognizing the problem.

It will be interesting to see customer acceptance of CS Fallback devices when they are finally introduced into the market.  

 

Verizon doesn't support handover in VoLTE

In a very un-shocking piece of news, Verizon announced that their new SIP/IMS based VoLTE service won't hand calls back to their existing CDMA network.

Why?

Basically there's no compatibility between the CDMA network which Verizon uses today to carry hundred of millions if not billions of voice minutes, and it's new VoLTE voice network.  Verizon is building an entirely new voice network, and is struggling to find a way to make the old and new compatible.  

Of course a VoLGA solution supports seamless call handover to/from the existing GSM/3G networks.  It did almost two years ago.  

Is this causing a fragment in Verizon's market???

Thunderbolt needs a little lightening

The new HTC Thunderbolt, the industry's first LTE phone, made its debut just the other day.

Overall the reviews are extremely positive… particularly related to the data rates of Verizon's LTE network. The primary complaint: Battery Life.

One analyst I spoke with at CTIA said the phone was out of power by noon. Certainly some of this can be attributed to the newness of LTE as a RAN technology, and undoubtedly improvements will come.

But the Thunderbolt also uses a new technique concocted for providing voice and data simultaneously when on LTE. Cleverly named "SVD", or Simultaneous Voice/Data, this is something vendors developed to keep the CDMA voice channel (1x) on/open when the LTE radio is on.

This is not how CDMA/EV-DO devices behave. But I guess Verizon was catching so much flack that it decided to try a dual-radio solution.

As a result, those HTC Thunderbolt commercials are actually more accurate than one realizes. It turns out that the Thunderbolt device really does need a steady supply of lightening to keep it powered throughout the day.

Watch this video about how Smart Wi-Fi saves battery life.

Video calling over LTE brings promise of fresh revenues

I must admit I was a bit stunned when I read the headline for this article in Total Telecoms.  

I believe that Skype and Facetime have set the bar pretty high for video calling – works great, costs nothing.  

So where are the ‘fresh revenues’ going to come from?

The article points out that, to date, mobile video calling has failed to achieve widespread popularity “…for a number of reasons, including doubts over quality, prohibitive costs to the user, incompatible devices and a perception that consumers aren’t interested in it.”  

But Leslie Ferry, vp of marketing for IMS vendor Broadsoft, claims that Skype’s success shows “there is such a demand in the market for these services.” 

Perhaps there is demand because Skype has overcome the negatives listed above…

• Skype’s drive to provide the highest quality over packet networks is admirable.  
• Certainly one can’t beat Skype’s price point – free.  
• Given Skype is available on the iPhone and Android 2.2, plus RIM and Nokia, plus every PC platform available, I think it’s fair to say they have overcome device compatibility.  
• Consumer interest… that’s a tough one, but making it free, available everywhere and with good quality should spur whatever demand there is.  

I struggle with how mobile service providers are going to be able to compete with the likes of Skype and Facetime, both of which seem to have an insurmountable lead today, even with Broadsoft’s undoubtedly advanced IMS-based video calling solution.  

In a world where billable minutes of use is (was) king, the fact that competitive video calling services are priced at zero (even less than regular cellular calls), will make it difficult to charge anything, let alone a premium.

Even in the brave new world of bits, bytes and data, where services are just packets flying over the network, I think it will be difficult for operators to actually construct a viable video-calling business.  If the bar is set at $0.00/minute for video calls between clients on the IP network, is this a loss leader?

Are mobile operators planning to invest tens of millions of real dollars to develop a video calling service they are going to (be forced to) give away? 

Undaunted, Ms. Ferry goes on to say “In any market there are always going to be new entrants” and while mobile operators can take advantage of their broad base of subscribers, “over the top guys have to start from scratch.”  

With Skype pushing 700m registered users, nearly $1 billion in revenues, and ~40% of minutes (calls) include video, it seems to me that Skype’s been scratching along for quite some time.    

 

The Price of Admission to LTE = IMS

It appears that the price of admission to the LTE club is IMS.  

I’ve said before that one can’t build an LTE phone without providing a mechanism to make a phone call.

Now that the industry has ‘standardized’ on IMS (aka VOLTE), mobile operators are compelled to invest in an entirely new voice infrastructure so subscribers can make calls when attached to LTE.  

Never mind that the new infrastructure works *only* when the phone is on LTE, and that the ‘old’ voice network, which carries trillions of voice minutes a year, will still have be maintained.  

LTE is for big service providers with deep pockets, and the price just went up. 

VoLGA is dead… Long live Rhine?!?!!?

 Not surprisingly, Deutsche Telekom finally capitulated to the ‘rest of the industry,’ declaring they too would follow the path of CS-Fallback and on to IMS (VoLTE) for voice over LTE.  

Followers of the site already know the facts:  

• The VoLGA specs have been done for nearly 18 months now.  
• VoLGA systems were demonstrated at MWC 2010 (a year ago).
• VoLGA systems demonstrated support for ALL circuit services, plus all those ‘crazy’ advanced services like pre-paid, plus all those ‘boring’ USSD network services.
• VoLGA support in the handset is extremely simple.
• VoLGA is practically built in to the handsets Kineto is supporting with Smart Wi-Fi today.

But that’s not why I wanted to write this post.  There are three trends which lead me to believe that while VoLGA may be ‘dead’ today, there’s a new river preparing to rise up.  

First, in nearly every meeting with vendors involved with bringing voice over LTE devices to market, there was a universal look of panic/dread/despair.  It seems that while these vendors keep up a good front to their perspective (and current) customers, when talking off-the-record, it’s clear there are some major issues with CS-Fallback (not really a surprise), and that IMS is a lot farther away than anyone really wants so say.  

Second, we’re starting to have the panic discussions with the mid/low level folks at the service providers – the people who actually have to do the work.  They are under the gun to deliver something that looks increasingly difficult.  Who knew you may have to align your LTE cell plan with GSM cells to meet acceptable handover delays?  Who knew that you end up with a much higher signaling load on the HLR with CS-Fallback?  Who new you needed to upgrade your existing GSM MSCs to support CS-Fallback?  Who knew that ‘falling back’ to GSM would introduce unacceptable levels of call drops?  

Finally, I was amazed at the number of analysts who came by to say “Don’t worry, ‘VoLGA’ may be dead, but the concept is still the fastest, cheapest voice solution for LTE.”   “Just wait,” they said. “Give the industry another year or two to really work into a panic, and then re-launch it as a different river…”  

There is no doubt in my mind that VoLGA is the fastest, cheapest, most effective way to deliver voice over LTE.  While we’re waiting, I think I’ll start exploring other rivers in Europe.  

 

CS Fallback an LTE Innovation? I don't think so.

I was surprised by the announcement this week that a vendor has “successfully completed a circuit switch fallback voice call”.  

After reading the release, I’m shocked that making a call with CSFB is considered “a breakthrough demonstrating continued innovation in LTE/EPC”.

I’m not sure forcing a phone off LTE and back to GSM when the user wants to make a phone call is really a ‘continued innovation in LTE’.  

Am I the only one that sees CS Fallback as a step backward for the industry?    

It seems to me that a continued innovation would be giving operators the ability to turn their existing voice service into a VoIP service over LTE.  There’s no need to ‘fall back’ for operators to make ‘full use of their existing network to protect investments’.  

VoLGA, the simple solution which was demonstrated by two vendors more than a year ago, turns mobile voice into mobile VoIP.  No complicated interfaces, all proven 3GPP technology.

PS - kudos to James Middleton at Telecoms.com for using the image of the tin cans in his coverage of this 'announcement'.  

Verizon LTE's phones: Still no voice over LTE

I braved the mobs at CES last week.  I managed to squeeze into the Verizon stand for a look at their new LTE phones.  I saw two of the four devices, and both were exactly like you’d expect: Android smartphones with touchscreens. 

I spoke with the Verizon reps to understand how voice services would be delivered. 

But first, a word on CDMA network technology.  Of course Verizon has a CDMA-based network (as opposed to a GSM-based network like 90% of the world).  In CDMA land, 1xRTT is the radio equivalent of ‘2G’ (ie circuit voice).  CDMA operators added data with EV-DO, which is a ‘Data-Only’ 3G-esque technology.  In the GSM world, the 3G technology (W-CDMA) supports voice and data concurrently. 

The point is, Verizon’s entire 3G network is ‘data only’, there is no ability to have simultaneous voice and data.  If you’re on a phone call, you’re on 1xRTT.  If you’re surfing the web, you’re on EV-DO.  If a call comes in, the EV-DO radio stops and the 1xRTT radio starts.  This is a deficiency that AT&T has hammered home several times. 

Back to LTE… 

It turns out, Verizon is going to treat voice the same in LTE as it does in EV-DO. 

LTE = no native voice (use 1xRTT); EV-DO = no native voice (use 1xRTT). 

The rep I spoke with said they are working to come up with a way to have the 1xRTT radio on all the time while the LTE radio was on so there wouldn’t be a delay in making (or receiving calls).

I asked what impact there might be on battery life by having both the LTE and 1xRTT radios on simultaneously.  The rep said that while they are technically building this capability, in all likelihood the phones will ship with simultaneous V&D turned off.  Huh????

Then I asked about IMS… “Way off in the future”.  Great.

There is still no voice over LTE.