4G upstream and downstream

[DTMark]

Hi

This forum is a wealth of resource on the net for mobile broadband - you've all been very helpful before, so I wonder if I can impose and ask a question..

We're in the UK and using a carrier called EE for our home broadband. This is an LTE service which AFAIK is @ 1800Mhz.

With a dongle, we see 22 Meg down and 20 Meg up with the dongle just outside the window upstairs.

We bought an Huawei B593. That can't go outside the window. Downstream unchanged, upstream a bit slower.

Plugged into the left hand socket on the back of that is an antenna I used to use for 3G with another carrier. It's directional, on the roof, and gave about a 15db improvement with that service. The carrier we use now, however, is a different carrier and LTE not 3G.

The B593 elected not to use that - not surprising since there's only one antenna when there should be two for 4G/LTE.

Up until recently. Now, it is using it. I think we have a new transmitter. Moreover I think it's on the same tower as the other carrier - can't be certain, but we've gone from a 2 bar signal to a 5 bar one.

At the same time, the upstream has skyrocketed. This is what we see now. Contention must be very low, as it's incredibly consistent in both directions.

As far as I knew, the maximum upstream on that service was/is 50Mbps. Maximum down is 100Mbps.

The router says this:

Antenna1 Status: External
Antenna2 Status: External

1 PLMN: 23430
2 Service status: Valid service
3 RSSI (dBm): -59
4 RSRP (dBm): -84
5 RSRQ (dB): -6
6 Roaming: No

I believe that the transmitter is 2.7km away if it is where I think it is.

Antenna is this one:
http://www.solwise.co.uk/3g-antenna-lpda-0092.html

Question is this:

Am I seeing an improvement on one leg only (upstream) because of the single antenna, and if that were either supplemented by a second identical one, or, replaced with a dual antenna array for 4G, I'd see an improvement on the downstream too, contention permitting?

From what I have read, what I have now "shouldn't work". Yet, it seems to.

Any of you knowledgeable experts care to venture an opinion - I'd love to see 50 Meg + downstream too

Thanks.

 

[Nerfherder]

1st world problems.

 

[SteveO]

its LTE FDD, not 4G

The high UL is probably due to high utilisation of DL resources and frequency and low utilisation on the UL frequency

 

[DTMark]

1st world problems.

There is a good cable network but it doesn't reach here. If you can't get that, you're probably stuck with ADSL over a really ancient phone network which here manages 1.5 Meg down and 0.2 Meg up. VDSL is being rolled out, but that performs worse than 3G once the knackered old cables are more than about 1km long - like they are here.

its LTE FDD, not 4G

The high UL is probably due to high utilisation of DL resources and frequency and low utilisation on the UL frequency

I thought about that. However if I run a speed test which does single and multi-threaded tests, they're about the same result. I was led to believe that significantly higher multi-threaded results indicate congestion/utilisation but that doesn't appear to be the case - there seems to be a limit of about 24Mbps down which I'd hoped (!) was imposed by signal strength and not contention.

 

[SteveO]

There is a good cable network but it doesn't reach here. If you can't get that, you're probably stuck with ADSL over a really ancient phone network which here manages 1.5 Meg down and 0.2 Meg up. VDSL is being rolled out, but that performs worse than 3G once the knackered old cables are more than about 1km long - like they are here.



I thought about that. However if I run a speed test which does single and multi-threaded tests, they're about the same result. I was led to believe that significantly higher multi-threaded results indicate congestion/utilisation but that doesn't appear to be the case - there seems to be a limit of about 24Mbps down which I'd hoped (!) was imposed by signal strength and not contention.

Could be a DL policy to limit to 24mbps

 

[DTMark]

Could be a DL policy to limit to 24mbps

That makes a lot of sense.

However according to the provider, this is not so.

They have a two tier system which they call "normal" and "double" speed (which is going to get them into some difficulty with the regulator at some point)

"Double speed" means that a cap of 30Mbps down is removed. Of course, you and I know that doesn't necessarily mean that it will go any faster or even hit 30Mbps which is why they'll be in trouble at some point, it is misleading and somewhat meaningless.

I have seen 37Mbps down before but only on a couple of random occasions including leaning out of the window with the dongle, and others on the same service genuinely do see 60Mbps+ downstream, best I've seen anyone get is 88Mbps down. But then those people aren't here, as in: in this area, and their transmitter may have more bandwidth on tap.

The downstream returning ~23.5Mbps is incredibly consistent and no different with or without the antenna. Given LTE is meant to have two antenna, and with a single antenna the upstream is solid at just under 50Mbps, this led to my theory about it being related to that. It's all a bit vague, though.

 

[Rickster]

LTE is a 4G wireless communications standard developed by the 3rd Generation Partnership Project

So LTE is technically 4G.

 

[DTMark]

So LTE is technically 4G.

My understanding is that what we have here in the UK isn't really 4G, but LTE. But, as is the case with so many things here, the marketing and the reality can be somewhat different.

I did start by describing it as LTE in the first post and then slipped into calling it 4G later and in the topic title.

That said, I'm not sufficiently up on the tech to know precisely what the difference is.

But, compared with what can be achieved down telephone lines, it's remarkably performant.

And, it's very new here and speeds are bound to fall away over time as the services become more contended.

 

[access]

i would try with 2 antennas.

maybe the combined speeds of the two open antenna ports is the same as when using 1 antenna, that's why you see the same downstream speed.

does the MIMO tech use the other antenna port when there is such big signal/speed differences between the 2 channels? or does it maybe slow down to the slowest antenna speed? to match channel speeds. I think the 2 antenna speeds have to be pretty similar for the MIMO tech to work sufficiently..

my 2 pence, im no 4g/lte fundi