RAvirani

Mine as well.

I spent a lot of time in the LA hills and drove the length of Mulholland Drive today, and I’ve got to say that Sprint is killing it up there. The network is PCS only and largely LTE-only in the hills, but coverage is FAR ahead of both T-Mobile and Verizon overall (my AT&T device was sadly dead for most of the day today). Sprint has tons of live L1900 small cells that we don’t have on our maps, and with with them all broadcasting 10x10+5x5 L1900, the network performance pretty awesome. T-Mobile seemed to be a 50/50 toss up between HSPA and midband LTE with a lot of patches of no service at all. Overall I wouldn’t call them very usable in the hills. Verizon had a good deal of small cells, but definitely not as many as Sprint. I’d say their overall performance was sub-par—their site spacing was not ideal, they have a much larger market share than Sprint, and most of their small cells were broadcasting L700 only, making for not-so-great RSRP levels and overall low throughput. These are exciting times for Sprint. I can’t wait to see more deployments like this across the country .

Hmm—I wonder where they are looking...

I can confirm that is true.

I can confirm that’s true. I was in Vegas last week and had no issues at all outdoors on the strip. Indoors was good about 80% of the time. And a good portion of indoors issues were a result of low upload speeds on Band 41. There are still a few non-NV sites that plague the strip though...

Don’t forget B12/17!

Few quick updates on the Seattle market: Verizon B5/B26 MFBI is now live in Seattle. Verizon is also replacing all of the antennas on many of their sites with new MIMO (and possibly NR-capable) antennas. It seems that in a large majority of these setups, only half of the ports are used—I think they will come back and hook up new RRUs (again maybe for NR) to these unused ports later. Verizon has historically had phenomenal site optimization and load balancing but not-so-great site density in my area, although recently, they have been closing that gap. Although they still seem to be avoiding using wooden utility poles as much as possible, they are colocating onto many existing macros and buildings as well as constructing their own towers. I think they will soon become a much stronger competitor to the other carriers in this market. All Sprint towers in the market that can support the weight are being upgraded to one hexadecaport antenna (4x4 800, 4x4 1900, 8x8 2500) and one 8T8R antenna (8x8 2500) per sector. Sites with weight limitations are also getting triband upgrades but with lower order MIMO. The common setups in order of weight are: A single hexadecaport antenna (4x4 800, 4x4 1900, 4x4 2500) per sector. A standard 6-port dual-band NV antenna (2x2 800, 4x4 1900) and a 10-port dual-band next-gen antenna (2x2 800, 8x8 2500) per sector. (This setup is quite rare). A single decaport triband antenna (2x2 800, 4x4 1900, 4x4 2500) per sector. I haven’t seen any L600 in the Seattle metro on T-Mobile yet. They largely seem to have stagnated over the last year or so. Many sites (probably 1/4 or 1/5) still don’t have L700, mostly sites on wooden utility poles or powerlines. I haven’t seen permits or anything to upgrade these sites... AT&T also seems to have largely stagnated in the Seattle market following their WCS rollout. That being said, their network is probably holding up the best in my immediate area. They have the best site density in my market and their 15x15 lowband, 20x20 midband and 10x10 highband LTE is working out really great for them.

Sprint does not run EVDO on 800 MHz anywhere, only 1x. And I never said 1x was going anywhere...just that EVDO was disappearing on some rural sites...

Crazy. Nice.

I don’t think the shutdown is widespread at all, but I’ve noticed it on several more rural sites.

I disabled LTE several times (probably 15+ times) and was unable to pick up EVDO at all—just 1x.

I agree. I think the spectrum is much better used for LTE or 1x than EVDO. Hopefully they will improve their handoff settings soon, though, because I drop from L1900 to CDMA (skipping L800) far too often for me to feel confident in VoLTE working reliably (at least on iPhones).

I think we are starting to see Sprint shutdown 3G in favor of a 1x/LTE only network. Today, I drove south in I-57 from Chicago to St. Louis and a large majority of the towers along my drive were only broadcasting 1x and LTE. I noticed this a few weeks ago in rural Indiana as well. Wider channels for VoLTE, I guess, and exciting times for Sprint .

The rescan timer shouldn’t matter because it’s usually between two and a half to three and a half minutes on iPhones (it varies with each update). ORNL should take care of the handoff much sooner than that... Maybe the ORNL was disabled or its transmit cycle was changed in your market or vendor region for some reason? In Seattle I’m still seeing my iPhone hop back onto LTE within 15-30 seconds or dropping to CDMA...

It stands for OtherRATNeighborList. ORNL in combination with RRC release messages (that contain redirection information) are the closest Sprint can realistically come to gaining the functionality of SIB13.

The amount of time it takes a modern phone to reconnect to LTE after dropping you CDMA is not related to the phone. Sprint has implemented ORNL network-wide and that should be taking care of CDMA-to-LTE handoffs.

I have not noticed a decrease as a result of the iOS upgrade...what you may be seeing is coverage loss with the new 16-port antennas. They have worse RF performance than Sprint’s previous antennas.

Tht seems pretty easily doable...what are the monitors connected to—a computer?

In low traffic areas, I personally don't see much benefit in deploying 8T8Rs over mini macros, if there are no capacity issues. 8T8Rs do provide better coverage and propagation than mini macros, but I think the difference is small enough that it can be overlooked. Ultimately, to deploy NR (which will be the next major wave of upgrades we see), both mini macro and 8T8R antenna/RRU sets will need to be swapped out for massive MIMO antennas. In the meantime, why waste money on 8T8Rs where they aren't needed?

It’s the same deal in Seattle minus 2 sites which probably aren’t ever getting upgraded. I’d go ahead and claim: 1x800 > WCDMA/HSPA800 ≈ 1x1900 > L700/800 as far as voice propagation goes. In open air, I’d give an edge to 1x1900 over WCDMA/HSPA800 athough when there are obstructions (in a forest, for example), I’d give the edge to WCDMA/HSPA800. VoLTE only requires a few mere kilobits per second to maintain a call, and today, Calling Plus calls will reliably go through at -118 to -120 3x3 L800 (in my market). With QoS on the network side, a call could probably be maintained to a much lower signal level, especially in markets where a 5x5 carrier is deployed.

Equipment and crews I believe. There was a recent shortage of 800/2500 antennas and there are never enough tower crews to go around.

Completely agreed. I think good times are ahead for AT&T.

Nice find! B14 is going to be huge for AT&T as it allows them 25 MHz of lowband FDD LTE in most markets while maintaining lowband WCDMA/HSPA. The network expansion requirements are going to be really good for their network as well. I’m excited to see how this plays out for them.

About time!

The first three things that come to mind are a faulty radio, loose jumpers or a site config error… I would tweet Marci Carris on twitter and include all of the details you know of—that’s probably the best way to quickly get in touch with a local engineer.