wispiANt

1 hour ago, Paynefanbro said:

Assuming we're gonna start seeing these getting deployed on the T-Mobile side later this year/early next year too: 

 

I believe these are dual-radio units, so I don't think we'll be seeing T-Mobile deploying this exact model.

The permit almost make me think this might just be a 6419 and 6449 bolted together? The listed dimensions/weight work out to be a little over twice that of a 6449. They also state that the unit can be broken down into (at least) two parts.

Looks as though Sprint eNB 79117 (40.701608578984036,-73.95875047577074) was converted to T-Mobile eNB 87493.

Another Verizon deployment, this time at the corner of Graham and Scholes (40.709250369767524, -73.94362128284861). Didn't have my Verizon SIM on hand, but it either hasn't been turned up or it's mmWave-only.

6 hours ago, radem said:

This will significantly simplify T-Mobile's operations from operating 5 networks and all the associated equipment to operating just 2 networks consisting of LTE and 5G.

And 2G! No shutdown date just yet.

For the T-Mobile network, at least, the UTMS shutdown shouldn't spell much of a change on a per-tower basis. In most cases, the same radios and antennas that are broadcasting midband (B2/B66) LTE are also being used to for 2G/3G. RAN shutdown and spectrum reallocation shouldn't be more complicated than pushing some changes in firmware.

16 hours ago, Paynefanbro said:

Seems like ideal spacing for mmWave is something like one site every couple hundred feet if you want contiguous coverage.

Wouldn't be surprised if Verizon is aiming to do just this. I believe they're still planning on eventually carrying 50% of urban traffic on mmWave and this NYC small cell situation has been a major roadblock. They're probably chomping at the bit to get mmWave deployment fully underway in NYC.

1 hour ago, Paynefanbro said:

Finally, I also got to visit the Extenet small cell by the Staples in Williamsburg. It's definitely a Verizon small cell broadcasting mmWave but nothing else. I wasn't picking up LTE nor C-band from it. Speeds were around ~2.2Gbps coming from it and it had an outdoor coverage radius of about a block and half before my phone dropped it completely.

I feel like this kinda makes sense to me? If you look at the technical drawings, it appears as though the deployment strategy is a cantenna-style antenna up top and then the accompanying RRUs below. I'm assuming Verizon provided these drawings (Nokia gear), but in T-Mobile's case the radios would likely be an AIR 4435 (n41 4x4), dual Ericsson MicroRadio 6503s (B2/B66 4x4), and an Ericsson MicroRadio 4408 (4x4 LAA and/or CBRS). Or something equivalent - it'd be nice to see something like the Ericsson RRU 4460 instead of those older 6503s but that might be a bit too big.

Anyway, I don't see how they could fit both all their sub 6GHz and mmWave gear under a single-carrier shroud. Verizon is more invested in mmWave so it wouldn't surprise me that they went the mmWave-only route more often. 

And I'm kinda impressed by the range! That seems to be the typical range of the current oDAS nodes.

1 hour ago, Paynefanbro said:

I'm also curious as to how many of these Sprint small cells they have broadcasting the keep PLMN citywide.

I'm guessing very few - I'm kinda under the impression that there are not many active oDAS nodes left overall. I haven't found an active one in Bushwick/Williamsburg/Harlem in ages. 

Another one!

Spotted my first Sprint -> Dish permit, 4318 Broadway in Upper Manhattan. Sprint decomm is Job #M00611880, Dish deployment is Job #M00616806.

The Sprint site at 102 North End Ave is another convert site, but the supporting documents haven't been uploaded and the description doesn't name the new carrier. Job #M00626206. Gonna lean towards T-Mobile on that one.

Wasn't able to grab a pic, but I believe the AT&T site (eNB 113160) just North of the Hewes St (M/J) stop has Ericsson C-Band antennas deployed. T-Mobile has also filed to remove the Sprint equipment on that collocation. 

On 3/18/2022 at 10:45 AM, Paynefanbro said:

eNB 9097/9098/9099 (Location: 40.73731141530291, -74.00194364096748)

T-Mobile has to be looking to decomm eNB 41151 if they convert this site. I just don't see how it would make sense to keep them both.

I'm looking forward to seeing what T-Mobile does with these on-building "small cells", they have a good number of them in Manhattan and IIRC they are still only broadcasting 2x2 MIMO B2/B66. AT&T has a whole bunch, as well, but it seems like they've upgraded quite a few of them with mmWave.

On 3/14/2022 at 11:01 AM, Paynefanbro said:

I visited this site and visually nothing has changed about it. I don't know how they managed to get 600MHz antennas to fit, but they did. Unfortunately I couldn't force my phone to connect to n71 at all. I cycled airplane mode a bunch but my phone would quickly connect to n41 from eNB 42917 instead of connecting to n71 from this site.

Speeds on both LTE and 5G weren't that impressive. On 5G I was seeing speeds of 50 down and 60 up and on LTE speeds were 30 down and 25 up. I think it may be a combination of backhaul being limited and just not being enough capacity being available in the area. The next nearest site to this one is eNB 53741 and it's also a flagpole site with only Band 2/12/66 so T-Mobile has their work cut out for them and increasing backhaul likely won't be enough.

I couldn't find anything on this site in the DOB database, but I did a bit of an analysis on flagpole sites in CT a few months ago:

https://old.reddit.com/r/cellmapper/comments/rj02qv/new_stealth_site_in_progress/hp3jbj3/

I'm guessing that the diameter of the shroud is a bit smaller than some of the sites on that list, so they likely didn't deploy 4x4 MIMO on lowband. But something like the FVV-65A-R3 (the smaller sibling of the FVV-65C-R3 on my list) has almost the same exact dimensions as the LNX-6513DS-A1M they probably had deployed.

Who needs 100MHz of n41 when you can aggregate 8 n71 carriers?

I'm now seeing the second n41 carrier @T-MoblieUser207 spotted a little while back, probably went live in Williamsburg within the last week or two. The Engineering Mode on my S21U won't show the bandwidth but I'm guessing it's only 20MHz or so. Seeing a bump in speeds of ~75Mb/s. 

Also, I believe Sprint eNB 74311 (at 40.69489559347538, -73.91177371321197, bordering Bushwick Playground) has now been converted and is broadcasting T-Mobile eNB 875607. Haven't confirmed it, though, and no permits have surfaced just yet.

https://old.reddit.com/r/cellmapper/comments/t25z8d/site_identification_request/

Interesting post from the CellMapper subreddit. It's getting tough to tell all these AIR 6449 variants apart. My guess is it's a B48-capable AIR 6449 variant deployed in the Bronx (similar to this deployment), not C-Band.

God, they're so ugly. I can't believe they were approved with such minor changes (though I do like the new floating screen design). I would so much rather have a couple single/multi-tenant streetlight deployments on each corner.

The expected coverage map is pretty cool, but it's a bit of a bummer that you'll see these things everywhere and Wi-Fi coverage will still be total garbage.

9 hours ago, T-MoblieUser207 said:

I'll need to double check, but B2 seems to be widened to 20 MHz on the subway DAS at one station, 125th St/Lex (4,5,6 trains).

Also widened to 20MHz at 59th St, 42nd St, and 34th St on the Blue line. Widened to 15MHz at 14th St - Union Sq.

9 hours ago, T-MoblieUser207 said:

Verizon eNB 81005 has a small little panel next to its C-band panel, anyone have an idea what it is for?

Checking Streetview, this his been up since 2016. First thought was backhaul, but there are no records of backhaul to that building. Second thought was LAA, but Verizon didn't start deploying LAA until late 2017. So this actually might be LTE-U, which they started deploying in late 2015.

13 minutes ago, T-MoblieUser207 said:

For eNB 41177, I passed by the area and I didn't see any panels on top of the buildings, so I restored its location, but its been located back. Is there something I'm missing? T-Mobile has blanketed the area with small cells, and I think this eNB is part of it, but don't have enough data to be certain.

There's a T-Mobile site on the building, DOB Job #320012008. Hard to link to the old DOB permit site but here's the job description:

"CREATING TELECOMMUNICATIONS ROOM WITH RELATED HVAC ON 1ST FLR AND CONDENSER ON ROOF RELATED ANTENNAS MOUNTED TO FACADE OF BUILDING"

And proof that it's T-Mobile:

27 minutes ago, T-MoblieUser207 said:

eNB 49434 not only got the full setup, but they also added an AIR 3246 to the site.

Great pics! Nice to see continued sector-splits.

28 minutes ago, T-MoblieUser207 said:

New convert site: eNB 875360. I was excepting this once I first saw it, especially since Sprint eNB 73854 over in Canarsie is being kept. Not sure what the 4th panel is, and I don't think the pic I took is helpful.

I think this might be a leftover Sprint antenna? Hard to tell but it kinda looks like what was there the last time Streetview rolled through.

29 minutes ago, T-MoblieUser207 said:

Lol, what the hell. Love this density but this just looks busted.
 

32 minutes ago, T-MoblieUser207 said:

I took a trip up to the Bronx and added a few towers/splits to the map. I still haven't fully gone through all the data I added yet, but I know I found a few convert sites. Also, if anyone is looking for T-Mobile CBRS to speedtest/check out, the Morrisana neighborhood is the hotspot for it. In addition to the tower Chuck located, the surrounding eNBs 53408, 53747, and 41098 have the CBRS panel up.

Surprised to see T-Mobile decided to invest in deploying the spectrum! Pull any good speeds or are they all backhaul limited?

57 minutes ago, Paynefanbro said:

Neville confirmed that C-band deployment is gonna start in 2023 and they're looking to use a single antenna for both their Auction 107 and Auction 110 spectrum.

https://www.fiercewireless.com/5g/t-mobile-eyes-345-ghz-c-band-one-and-done-rollout

Yeah, I saw that.

Ericsson's official statement to the FCC was that implementation of a two-step OOBE limit on 3.45GHz (instead of the single -13 dBm/MHz OOBE limit applied to 3.7GHz) "would prevent development of a single radio in the U.S. across the 3.45 GHz and the 3.7 GHz band, requiring development of two U.S.-specific radios" (source). 

Nokia's statement was that designing hardware around a two-step OOBE limit was doable (without indicating if they could design a radio to support both 3.45GHz and 3.7GHz). 

So I'm not really sure how Neville is going to achieve that.

On 1/27/2022 at 3:32 AM, mikejeep said:

Can you send a diagnostic report next time you see that? Was it reporting n41 before this version? Thanks.

I don't believe I've recorded n41 SA or n41+n71 SA (with n41 as the PCC) with SCP before, so unfortunately I'm not sure. Can definitely send diagnostics the next time I'm over there, though!

Another keep site in downtown Brooklyn (DOB Job# B00464934):

96 Schermerhorn St

T-Mobile eNBs 875823/875821

Sprint eNB 5854

SA n41 is being recognized as SA n38 in the latest SCP beta. gNB 1327777 on the map.

https://www.pcmag.com/news/exclusive-we-found-t-mobiles-secret-weapon-against-c-band

Seems like he's seeing:

B2+B66 +n41+n41
n71+n41

Not that it really matters, I suppose:

Quote

Looking at how my traffic divided between frequencies, I found that some of my transmissions didn't send any data over the second carrier. Two speed tests sent a little bit of data over the second carrier.

That might be the usual configuration issues, but it might also just be the fact that my tests weren't even saturating the first 5G channel; I'm pretty sure my speed was being limited by the capacity of the tower, not of the airwaves.

1 hour ago, dkyeager said:

Great job on your analysis of small cells versus the 3.45GHZ and the population coverage requirements.

Crunching the numbers a bit more, covering the full MSA would result in slightly under 80% population coverage. But they'd have to deploy on essentially every micro/macro cell in the area.

1 hour ago, dkyeager said:

I wonder if the fcc is stalling because they may address the key limitation of of 2.5: the convoluted frequencies. Not sure how they would get there, but it would be better public policy if you could actually use a single license in current times, ie 5, 10 or 20MHz. Of course they could also go in more of a nonprofit or small business direction. But most likely they will keep it as planned given how messy the transition would be.

In many/most metro areas BRS/EBS is fully licensed. Would be nice if they put pressure on the squatters.

Yeah, I'm not gonna pretend like I know how either negotiations with squatters or Auction 108 will pan out. That said, I can't see 3.45GHz being *that* valuable to squatters considering the strict buildout requirements. 

Another interesting thought I had was a possible spectrum trade with Dish. Dish is still leasing 600MHz to T-Mobile in quite a few metro areas, and I'm sure T-Mobile is looking ahead to what their lowband spectrum situation will be once those leases expire. If T-Mobile is truly going to rely on n71+n41 CA as much as they say they will, it sure would be nice to be working with more than 5MHz-10MHz of n71 uplink capacity.

Don't go quoting me on that, though, total (pipedream) speculation haha.

21 hours ago, Paynefanbro said:

Edit: If you take a look at where they purchased the license, you can see that they’re mostly in urban areas where meeting the buildout requirements of this spectrum would be fairly easy. It lends credence to the idea that they may be using it primarily for small cells with deployments on macros for areas where small cells are less feasible. 

I didn't actually look at the buildout requirements before making that comment - they're definitely going to have to deploy on macro sites if they want to hit the buildout requirements.

PEA001 has a population of 25,237,061, of which they will have to hit 45% in 4 years (11,356,677) and 80% in 8 years (20,189,649). If they were to cover the entirely, all five boroughs, of NYC using only small cells (something I'd say is impossible considering their current small cell density), they'd only be covering 32% of the population in the PEA.

And this is with spectrum that only adds 300Mb/s per sector and will likely have only 50% the range of their current n41 equipment. Doesn't really seem worth it to me.

I'm of the opinion that they're looking to hedge their bets in further EBS/BRS acquisition. 

12 hours ago, Paynefanbro said:

I wonder if T-Mobile will begin deploying it this year or if they'll wait until the end of 2023 to do a single climb and deploy it alongside C-band.

Agreed, it'll be very interesting to see what their deployment strategy is with this. The fact that this will require separate radios from C-Band will definitely work to their detriment - I can still see T-Mobile shelling out for mMIMO gear on C-Band, but can't see it being worth the investment here. That means you'll be looking at a relatively small boost in capacity (374 Mb/s theoretical with standard 4x4 MIMO equipment) and limited range.