Slashdot Mirror
Verizon CTO Says 4G Service Is On Track
Verizon has announced that it is on track to roll out their new 4G LTE service using the 700 MHz band that it acquired in the recent FCC auction. Targeted first towards USB air cards for laptop customers, the service will be extended to cell phones and other mobile devices with embedded LTE eventually. Testing in Boston and Seattle should conclude in the next couple of months and commercial deployments should follow soon thereafter. "Lynch said getting voice to work over LTE has been particularly challenging. But that challenge is getting resolved as Verizon and other members of the GSMA announced Monday they are supporting a standard that uses IMS technology to deliver voice services over LTE. Still, more work needs to be done. Until a solution is complete, Verizon will use its CDMA network to provide voice services. And the LTE network will be used for data. Eventually, when voice over LTE becomes a reality, Verizon will use that technology. Verizon will also have to integrate EV-DO into its LTE offering to ensure that customers can switch to the 3G EV-DO network when the 4G LTE network is not available. Even though Verizon is being aggressive in building its network, it won't happen overnight."
So, everything is hunky-dory, going right according to plan.
But the phone company doesn't actually have any way of making the new technology make voice calls, so they'll be retaining the legacy CDMA technology. And, of course, they'll be building the intermediate legacy EV-DO technology for the forseeable future to deal with places where the new hotness is not actually available. Oh, and support for mobile devices is planned for "eventually"...
I wish my standards for success were this achievable.
Damn it, I should have waited to buy my phone, instead of just buying it yesterday!
If they used VoIP, they wouldn't be able to justify the price they charge the end users.
It is worth nothing that while LTE is still in development Sprint and Clearwire have already deployed 4G services that are operational and covering 30 million people in the US. Wimax is deployed in around 145 countries worldwide. Sprint will have a 4G device in 2Q or 3Q this year, and will likely have 120 million people covered by 4G before LTE is even deployed here.
Having worked on LTE and LTE Advanced these last two years of my university degree, I can't wait to see and use the actual network... Well, guess I'll have to wait a bit longer to see it in Europe...
Regards, Boyan
Who says they have to justify? Not a lot of alternatives out there.
And they do VoIP on the back end ...
---- Booth was a patriot ----
Timing over a packet network is not trivial for voice or video...
An obscure unpopular candybar-shaped brick, preferably with a non-standard keypad and an external antenna and a kick-ass field test app pre-installed. Kinda like the Nokia 6650, but with LTE. I'd cringe if the first production LTE phone was, for example an iPhone 4G or some Google Touchscreen phone that uses LTE to supply a continuous barrage of text based ads
VoIP has large problems with latency. IP over wireless latency is all over the place. One tower can be 150ms the next 3000. Yet there will still be voice slots open. I know you will say go all IP. What about the *MILLIONS* of people out there with working cell phones who could give a crap about IP? They are in a bit of a pickle. They scavenged up all the slots for voice. Now they do not have enough for IP itself. Never mind the 8 different ways they have of talking over the air (all backwards compatible with the old AMPS network). LTE or WiMAX are the future (my money is on LTE due to who is going for it At&t, vodaphone, and verizon). But they do not have have (at least right now) the capital to spend on new equip. Especially when they have a depreciation sched of 5-10 years and just upgraded the whole show 3-4 years ago. They havent finished paying for the old equip yet and are upgrading when it makes sense and they think they can charge you more for it.
Just switching over to IP doesnt change their pricing model at all. The over the air cost is actually quite small (think cents per hour per hundred users). The real costs are the million dollar radios/backhaul lines. Then the tower rentals. Then all the people to support that. THEN and this is most important what the market will bare. People seem to think if it costs 2cents to make they should charge 2 cents. They are not a 'non profit' they will charge what they think they can get out of you (if its 2 cents or less they probably will not bother to do it at all). They will set their prices to the 'sweet spot'. So even if magically they go all IP they will still charge just as much. Take text messageing they literally get the over the air for free as the message is sent padded out all 0's anyway. It is the same message so the tower can see where you are at. Yet they charge what the market will bare in 15 cents to 50 cents a message.
VoIP over a cell network would be a bad idea. At least in the traditional sense. Yes, it works, barely, but it's like trying to fit a square peg in a round hole. The problem stems from the fact that voip infrastructures are usually designed around a mostly reliable network, that very occasionally drops entire packets. A cell network is designed to cope with an unreliable network, where bit errors are common. Everything, from codecs to protocols are designed with that in mind. Is the reason why G729 can get the same quality at 8kbps as AMR at around 12kbps. The extra bits are there for redundancy. In addition to that, you definitely want traffic shaping and QoS guarantees when doing voice. Otherwise your neighbor's porn downloads might crowd out your calls. You don't really notice that in broadband based voip installations, simply because there's usually a ton of bandwidth to go around. But a shared radio connection is an entirely different ballgame.
They will probably use packet switching (read IP) on the backend though. Once the bits are safely tucked in some fat fiber pipes.
If con is the opposite of pro, is Congress the opposite of progress?
Get with it. Here in Nebraska we already have 4-H
Give me Classic Slashdot or give me death!
That's the plan. LTE will use SIP with QoS to guarantee maximum latency and jitter. To all of the people saying you can't use VoIP: how stupid are you? Almost all voice calls in the western world go over packet switched networks and have for most of the last decade, and most of the last three decade in some cases. Do you think things magically get worse because those packets have an IP header? If you make a landline call anywhere in the UK or Canada, you are using VoIP.
The problem is that some of the standards for telephony services over LTE (which is an all-IP network) have not quite been finalised yet. This includes things like SMS bridges and the standards for mapping SIP addresses to phone numbers.
I am TheRaven on Soylent News
According to Wiki, 4G is packet based only. It's assumed that by the time 4G is rolled out, IP4 addresses will have been exhausted. So does that mean all new 4G phones will use IP6 by default? Sounds like a good idea to me. If your going to make a move to IP6, handheld devices are the perfect place to start rolling out the new IP standard.
Life is not for the lazy.
Actually calls will probably sound better over LTE.
Until you call someone on a different network, such as the landline network, at which point the voice quality goes back to narrowband.
No. I don't think so. I can't afford Verizon any more so this spring I close the account and smash the phones to move to another carrier. The phones are junk and the plan costs too much. I'll get new phones if or when I choose a replacement carrier.
bob@Osprey:~>
LTE *is* VoIP
Climate Progress - Hell and High Water
how stupid are you?
Wait, so you're saying my personal experience using Skype at a coffee shop doesn't scale to massive international networks? Shocking, shocking I say.
My God, it's Full of Source!
OUTSIDE_IP=$(dig +short my.ip @outsideip.net)
Water is wet, which is important for someone who goes boating. Likewise, calls routed through the PSTN are narrowband, which is important for someone who calls mostly customers on another network.
No, I mean maximum latency. QoS guarantees a maximum latency, not a minimum latency for each packet. It doesn't need to guarantee a minimum latency; that is defined by the physical conditions of the network.
I am TheRaven on Soylent News
That's not the entire problem. How does the handset find the SIP to POTS gateway, for example? Do you hard-code it in the SIM? Do you use anycast (in which case, is there some authentication to prevent spoofing)?
I am TheRaven on Soylent News
The Cellphone doesn't need to know the SS7 gateway(aka pots gateway), that's SIP proxy job... However you may ask how does the cellphone get the registar and proxy addresses, and the anwser is:or via a field in DHCP or a field in the PDP activation context..