LTE Plus / Enhanced LTE (was "Sprint Spark" - Official Name for the Tri-Band Network)

[Deval]

So for whatever reason, my Mini has not been able to connect to B41, so I'm a sad clown at the moment.

[Paynefanbro]

On collocated sites, can Sprint hook up NV backhaul to the Clear tower? One of the towers that services my home has 3G and 4G upgraded, and happens to be the location of a Clear tower too.

 

Would Sprint increase backhaul to this site sooner rather than later?

[transitwatch889]

So for whatever reason, my Mini has not been able to connect to B41, so I'm a sad clown at the moment.

Are any of the sites you visited as listed as being active on sprint online mapping?

[ericdabbs]

On collocated sites, can Sprint hook up NV backhaul to the Clear tower? One of the towers that services my home has 3G and 4G upgraded, and happens to be the location of a Clear tower too.

 

Would Sprint increase backhaul to this site sooner rather than later?

 

I think that is the plan eventually on colocated sites that already have the appropriate NV backhaul installed. However at the same time I feel that Sprint is only going to hook up NV backhaul to the LTE 2600 radios/antennas to the their cabinet if they have sufficient backhaul to support about three 20 MHz TD-LTE carriers.  

 

Until then I think Sprint might just use Clearwire backhaul for just LTE 2600 and current NV backhaul for LTE 800/1900 until it can receive a huge boost in backhaul on the Sprint NV side to support Sprint's ultimate vision of LTE 800/1900/2600 on the same platform.

[iansltx]

Hate to rain on everyone's parade, but Sprint promises peak speeds of 50-60 Mbps on TD-LTE, not 70+ Mbps. Even with proper backhaul. Maybe you can hit higher speeds, but the demos Sprint was doing with the Galaxy Mega hit a bit over 50 Mbps in lab conditions.

 

On the other hand, Sprint's network (due to small cell sizes) will be hitting near-ideal speeds a much larger percentage of the time than AT&T or Verizon, until those guys do an AWS overlay in an area.

[Deval]

Are any of the sites you visited as listed as being active on sprint online mapping?

 

I'll take a look

[Paynefanbro]

Hate to rain on everyone's parade, but Sprint promises peak speeds of 50-60 Mbps on TD-LTE, not 70+ Mbps. Even with proper backhaul. Maybe you can hit higher speeds, but the demos Sprint was doing with the Galaxy Mega hit a bit over 50 Mbps in lab conditions.

 

On the other hand, Sprint's network (due to small cell sizes) will be hitting near-ideal speeds a much larger percentage of the time than AT&T or Verizon, until those guys do an AWS overlay in an area.

 

I still think they are being modest because we know the speeds that this is capable of. They even talk about aggregating spectrum next year to had 40Mhz and provide over 100 Mbps. Isn't it obvious that they'll need to increase backhaul to achieve this and wouldn't they do this sooner than later to avoid people getting slower speeds than can be delivered? 

 

Ex: Verizon's AWS LTE network in NYC runs at about 80Mbps because of backhaul constraint.

[mmark27]

I've read through the last 4-5 pages, and don't see this addressed.  I have 3 things/questions/comments: 

 

#1. It appears that the 4.3 update for the S4 does not show equivalent signal bars to LTE strength, IF one can believe the Signal Check depiction of bar strength, which I'm pretty sure is close to how OS determines the signal. 

 

#2. I notice that if I have strong LTE outside (-80-85dBm) and go into a building, I'll lose the LTE and pick up eHRPD quite strongly (-55dBm). Since I have a S4, I'll only get the band 25 1900LTE and the eHRPD broadcasting is also off the 1900 band.  Are the radios just that much stronger for the 3G than the LTE? Will this eventually give me 1900LTE inside the building equivalent to the 3G that I'm seeing? 

 

#3. I also see that if I drive around the city with the LTE Engineering mode on, it'll report that I'm always connected to LTE just very low at -117dBm.  At this point, the phone/signalcheck report that I'm getting 3G because that signal is stronger and it bumps to that for data connectivity. 

 

SO, will 1900LTE be equivalent to eHRPD 1900 as far as signal and connection?  Or are the LTE radios just weaker?  Will they be "turning up" the strength of signal on towers as a market matures?  Or it is what it is and the LTE will not be any better on 1900 than it is when they first turn on the tower?  Perhaps I assumed that I'd at least be getting LTE where I had 3G before, inside or out.  

 

Feel free to link me to an article that discusses this, I couldn't really find one that discussed this explicitly. 

[VW Maverick]

Never mind.  

Edited November 18, 2013 by VW Maverick

[mhammett]

there's no way in hell they can get the max speeds for multiple 20mhz carriers from one fiber air link after 2-3 hops

What is the max capacity in either direction from a fully loaded tower? Gig?

 

Sent from my EVO using Tapatalk

 

 

[WiWavelength]

Perhaps I assumed that I'd at least be getting LTE where I had 3G before, inside or out.

Nope. In terms of robustness, CDMA1X > EV-DO > LTE. You should have already noticed that EV-DO drops before CDMA1X. The same holds true for LTE and EV-DO. Each generation of wireless airlink grows more efficient but also more fragile.

 

Feel free to link me to an article that discusses this, I couldn't really find one that discussed this explicitly.

You have been a member for quite a while. Do you read our articles posted on The Wall? I wrote this one over a year ago. It has received over 40,000 views. Plus, it is probably the most cited and copied article on LTE signal strength. I am sorry if you missed it previously.

 

http://s4gru.com/index.php?/blog/1/entry-308-rssi-vs-rsrp-a-brief-lte-signal-strength-primer/

 

AJ

[milan03]

I still think they are being modest because we know the speeds that this is capable of. They even talk about aggregating spectrum next year to had 40Mhz and provide over 100 Mbps. Isn't it obvious that they'll need to increase backhaul to achieve this and wouldn't they do this sooner than later to avoid people getting slower speeds than can be delivered? 

 

Ex: Verizon's AWS LTE network in NYC runs at about 80Mbps because of backhaul constraint.

Carriers like to initially under provision their backhaul service until they absolutely have to pay for more bandwidth. Verizon has fiber to pretty much all of their cell sites in NYC, but they're definitely not paying extra $$$ to support the entire 150Mbps + 73.6Mbps per sector, yet. Their 10Mhz Band 13 is constantly loaded, but Band 4 is consistently above their advertised 5-12Mbps, so I guess they're saving money...

 

T-Mobile is doing pretty much the same. They just went from 5Mhz FDD LTE to 10Mhz FDD LTE in NYC, and peak speeds went from 35Mbps to only ~40Mbps on an unloaded sector. The upload has doubled to about ~20Mbps.

 

The good news is that it's all scalable and only requires remote provisioning if the equipment is in place. It is an extra expense for wireless operators, but with sufficient backhaul service they're effectively increasing spectral efficiency at their cells by utilizing extra capacity, and potentially shortening data session time at the UE.

[jamesinclair]

Just saw the Sprint Spark icon. Its hilariously bad. Hope its not distracting or easy to disable.

 

 

Probably a GS4 triband phone refresh.  I am not sure if Sprint is in the process of clearing out their single band LTE GS4 inventory to make wave for the GS4 triband phone.

 

Lots of black friday GS4 sales.

 

 

[nexgencpu]

I would be interested to know where that is. I went to 3 B41 sites last week.....got nothing....

You might want to get on XDA and ask, but as far as I know, no one has been able to connect to B41 using Nexus 5. Too bad because NYC is chock full of "Turbo" B41 that I cannot access.

[ericdabbs]

You might want to get on XDA and ask, but as far as I know, no one as been able to connect to B41 using Nexus 5. Too bad because NYC is chock full of "Turbo" B41 that I cannot access.

 

Hmm that is interesting.  I wish I could tell other S4GRU members on this forum to test the same confirmed B41 LTE tower in LA/OC to see if they can connect on B41 LTE with their Nexus 5.

[nexgencpu]

Hmm that is interesting.  I wish I could tell other S4GRU members on this forum to test the same confirmed B41 LTE tower in LA/OC to see if they can connect on B41 LTE with their Nexus 5.

So, you have been limited to just one tower? That would seem somewhat bizarre, and would indicate that its not just a device level compatibility issue. Another thing I find interesting that no reports have come in of other devices connecting to B41 besides the Hot spots. Like the Mini, Mega or Max. 

[ericdabbs]

So, you have been limited to just one tower? That would seem somewhat bizarre, and would indicate that its not just a device level compatibility issue. Another thing I find interesting that no reports have come in of other devices connecting to B41 besides the Hot spots. Like the Mini, Mega or Max. 

 

Technically 2 B41 LTE towers.  Sorry I forgot to mention the other one. I just haven't gone out to test out more B41 LTE towers since they are not along my route.  I have confidence that it should work on the B41 LTE towers in the LA/OC area that have been approved so far.

[mmark27]

The Wall doesn't seem to display all articles, without several clicks, that's the weakest part of these forums, IMO. There's no "main page" that opens first or at top level with articles. My apologies for missing that article, thanks for the link.

 

Sent from my SPH-L720 using Tapatalk

 

 

[DaQue]

When they turn on carrier aggregation some time down the road will the phones have to have multiple radios or can it be done with just one?

 

 

 

[mmark27]

Also my question was, will there be an increase in strength after the market matures. Or is it "as is" once it's on. Not talking speeds here, but signal strength.

 

Actually, your article, while very informative, didn't answer much of my question. Why in the case of a very strong signal for LTE do I not get anything just inside a building? I mean despite the variance in strengths, penetration and fragility, 1900 should still give me something you'd think. Oh well, hopefully 800LTE fixes some of this eventually.

 

Sent from my SPH-L720 using Tapatalk

 

[digiblur]

When they turn on carrier aggregation some time down the road will the phones have to have multiple radios or can it be done with just one?

 

The carrier aggregation with the current single transmission path devices would require the channels to be in the same band. Much like the first docsis 3.0 cable modems required the channels to be adjacent to bond them since it was done with the same tuner.

[mhammett]

150Mbps + 73.6Mbps per sector, yet.

(150 + 73.6) per sector or 150 + (73.6 per sector)?

 

Sent from my EVO using Tapatalk

 

 

[milan03]

(150 + 73.6) per sector or 150 + (73.6 per sector)?

 

Sent from my EVO using Tapatalk

Was referring to two live FDD airlinks they have to support per sector. 10Mhz of Band 13, and 20Mhz of AWS.

[tommym65]

What is the max capacity in either direction from a fully loaded tower? Gig?

 

Sent from my EVO using Tapatalk

 

Ideal, perfect-condition, optimum, everything-is-perfect combined download/upload speeds:

 

1900 LTE = ~37 Mbps/sector = 111 Mbps

 

800 LTE = ~37 Mbps/sector = 111 Mbps (although I think AJ has said it is slightly less, because of propagation characteristics)

 

2500/2600 LTE = ~74 Mbps/sector = 222 Mbps

 

Total LTE = ~444 Mbps/sector under optimum conditions

 

Corrected:

Total LTE = ~444 Mbps/site under optimum conditions

 

Voice/CDMA = ??? (but far less than LTE)

 

So a totally loaded site with all 3 bands could need about 500 Mbps if each technology had a single carrier and each sub was virtually on top of the site.  If you add carriers (as Sprint is supposed to do on 1900 in Chicago in early 2014), figure you add ~100 Mbps per 800 or 1900 carrier, about 200 Mbps per 2500/2600 carrier.  In the next couple of years, no Sprint site would be likely to need more than 1Gbps, well within the capacity of a single fiber, and (mhammett please correct me) also within the capacity of a well-implemented microwave link.

 

My numbers are fuzzy, and may not properly account for download versus upload, but they are close. They also do not account for the fact that most subs will generate far lower speeds due to distance and attenuation.  Anyone who can improve them, please do so.

[mhammett]

Was referring to two live FDD airlinks they have to support per sector. 10Mhz of Band 13, and 20Mhz of AWS.

The price difference between 225 of backhaul and 675 of backhaul is minimal. Over 150 or 200, just get a GigE.

 

Sent from my EVO using Tapatalk