Jump to content

NV/LTE Breaking point(s)


Recommended Posts

Ok so, who knows the numbers and wants to do the math?

 

What is the maximal consistent/sustained throughput of a fully completed standard sprint NV tower (not considering any backhaul limits)?

 

Let's assume we start with the current deployment makeup, then consider the effect (if any) of adding smr800 then lte800

 

- There are ho.w many RRU's in a standard deployment?

- Each RRU's max aggregated sustained throughput is?

- Max single subscriber sustained throughput is?

- Max number of subscribers able to be served (not necessarily at max sustained throughput) simultaniously per RRU?

- known bottlenecks in the chain that would affect either/or single or aggregate user experience?

 

 

What I am looking for is the # of subscribers and/or the sustained bandwidth point at which an NVE tower becomes over-saturated and starts degrading users experience.

  • Like 2
Link to comment
Share on other sites

The breaking point in which overloading occurs could either be due to an overloaded carrier channel or overloaded backhaul. Cell splitting can take a single cell site broadcasting one carrier in all directions to a carrier that broadcasts several different carriers using the same spectrum. The addition of new channels effectively allows a cell site to become two sites operating in the same space adding further capacity. Backhaul has a set bandwidth according to the back haul provider which under ideal conditions is never exceeded.

 

cellsplitting.png

 

spectrume.png

 

Overloading occurs when a carrier channel is over burdened beyond it's capacity or the back haul link has to0 little bandwidth for the demand.

Link to comment
Share on other sites

For fiber-backhauled sites, backhaul can be taken out of the equation as a limitation. For microwave sites, you'd have to load things up pretty heavily to hit backhaul limits, but you could reach them one hop down the chain by putting three 5x5 carriers on each of three sectors and saturating those sectors (e.g. two PCS and one SMR). Still hard to do though, considering that we're talking about hundreds of megabits per second, more bandwidth than a residential cable provider (with DOCSIS 3) offers a comparable number of subscribers.

 

Realistically, to start with Sprint has one 5x5 LTE carrier per sector, and three sectors per site. In ideal real-world conditions, you get around 35 Mbps of bandwidth per sector on the download side, which in today's world is the side that you have to worry about (upstream capacity is 15 or so Mbps per 5MHz carrier). So that's 105M of capacity per cell, spread over whoever's in that cell (could be a mile radius, could be 2-3x that, could be one-third that).

 

If you add another 5MHz PCS carrier in, double those capacity numbers, spread over the same subscriber footprint. You may only need to do this on one sector, depending on load...35 Mbps (I've seen 32 Mbps down and 12 Mbps up on a speed test on Sprint LTE before) goes a long way in a smartphone-heavy mobile setting when no one is torture-testing their connection constantly.

 

If you add a 5MHz SMR carrier in, that's another 35 Mbps on that sector, though the additional range may give you more users on that carrier than on the others. But that can be balanced out with the proper network intelligence (people with better signals stay on PCS).

 

35 Mbps assumes two receive antennas and 64QAM modulation (six bits per Hz of frequency per second before error correction), which is reasonable for areas with good LTE service on any currently-available LTE phone. Higher data rates could happen if a phone and a base station had more antennas interfacing with each other (3x? or better MISO or MIMO rom the phone's perspective). But I doubt that will happen in the near future.

 

If you add in Clear TD-LTE, you have a ton of capacity per sector per channel, since there's 20 MHz of bandwidth that can be configured in a variety of download-upload ratios. If Clear configures it for 3:1, you've got the equivalent of a 15MHz FD-LTE carrier on the download side, plus a 5MHz carrier on the upload side. So similar upload speeds to the macro-cell network (using 16QAM modulation, which should still work at 2500/2600MHz for the distances we're talking about), but with download capacity in excess of 100 Mbps per channel, assuming two receive antennas and good signals on both (enough to do 64QAM).

 

Any more questions? :)

  • Like 1
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • large.unreadcontent.png.6ef00db54e758d06

  • gallery_1_23_9202.png

  • Posts

    • I've now seen how things work in Kobe, Hiroshima, and Osaka, as well as some areas south of Osaka (e.g. Wakayama, Kinokawa), and tried three more SIMs. The two physical SIMs (different branding for each) both use IIJ, which provides a Japanese IP address/routing on NTT, aleit LTE-only, so latency is ~45ms to Tokyo. The catch with NTT is that it uses two frequency bands (B42/3500 MHz LTE, n79/4900 MHz NR) that you're not going to get on an Android sold in the US, and I'm guessing that B42 would be helpful speed-wise on that network, as it doesn't have B41. I also found one place that doesn't have cell service: a vending machine in the back of the Osaka Castle tower. Or, rather, the B8/18/19 signal is weak enough there to be unusable. Going back to 5G for a moment, I saw a fair amount of Softbank n257 in Hiroshima, as well as in some train stations between Osaka and Kobe. 4x100 MHz bandwidth, anchored by B1/3/8, with speeds sometimes exceeding 400 Mbps on the US Mobile roaming eSIM. Not quite the speeds I've seen on mmW in the States, but I've probably been on mmW for more time over the past few days than I have in the US over the past year, so I'll take it. My fastest speed test was actually on SoftBank n77 though, with 100 MHz of that plus 10x10 B8 hitting ~700 Mbps down and ~80 Mbps up with ~100ms latency...on the roaming eSIM...on the 4th floor of the hotel near Shin-Kobe station. Guessing B8 was a DAS or small cell based on signal levels, and the n77 might have been (or was just a less-used sector of the site serving the train station). I'm now 99% sure that all three providers are running DSS on band 28, and I've seen 10x10 on similar frequencies from both NTT and SoftBank IIRC, on both LTE and 5G. I also picked up one more eSIM: my1010, which is different from 1010/csl used by US Mobile's eSIM unfortunately, as it's LTE-only. On the bright side, it's cheap (10GB/7 days is like $11, and 20GB for the same period would be around $15), and can use both KDDI and SoftBank LTE. It also egresses from Taiwan (Chunghwa Telecom), though latency isn't really any better than the Singapore based eSIMs. Tomorrow will include the most rural part of our journey, so we'll see how networks hold up there, and from tomorrow night on we'll be in Tokyo, so any further reports after that will be Tokyo-centric.
    • I think the push for them is adding US Mobile as a MVNO with a priority data plan.  Ultimately, making people more aware of priority would allow them (and other carriers) to differentiate themselves from MVNOs like Consumer Cellular that advertise the same coverage. n77 has dramatically reduced the need for priority service at Verizon where the mere functioning of your phone was in jeopardy a couple of years ago if you had a low priority plan like Red Pocket. Only have heard of problems with T-Mobile in parts of Los Angeles. AT&T fell in between. All had issues at large concerts and festivals, or sporting events if your carrier has no on-site rights. Edit: Dishes native 5g network has different issues: not enough sites, limited bandwidth. Higher priority would help a few. Truth is they can push phones to AT&T or T-Mobile.
    • Tracfone AT&T sims went from QCI 8 to 9 as well a couple years ago. I'm pretty neutral towards AT&T's turbo feature here, the only bad taste left was for those who had unadvertised QCI 7 a couple months ago moved down to 8. In my eyes it would have been a lot better for AT&T to include turbo in those Premium/Elite plans for free to keep them at QCI 7, while also introducing this turbo add on option for any other plans or devices. As it stands now only a handful of plans can add it, and only if you're using a device on a random list of devices AT&T considers to be 5G smartphones.
    • My Red Pocket AT&T GSMA account was dropped to QCI 9 about a year ago.  Most recently 8 for the last few years prior.  Voice remains at 5.
  • Recently Browsing

    • No registered users viewing this page.
×
×
  • Create New...