5G vs VDSL

Silly question, how can a wireless connection (5G) be faster than a physical connection (VDSL) for the last mile?

How? It can be faster by using different technology. VDSL is capped at 80Mbps in the UK and will have a lot of loss on a 1-mile length of copper. Most people are 500metres or less from their local cabinet.

5G is radio, so it will depend how many users there are using the service at the same time.

Why?

The Openreach VDLS (actually VDSL2) has a maximum connection speed of 80Mbps. Even full 4G easily beats that with line of sight of the mast.

5G is far faster than 5G 4G. (Typo fixed, see later post by jchamier ).

Don't be confused by the term "last mile". It is nothing to do with an actual mile. Normally the last leg of the (to<>from) data trip from a concentrator of some kind where multiple signals are routed out as a single connection. Think DSLAM/MSAN/Aggregation point?/Mobile mast.

That's why for instance many more 5G "masts" will be needed than 4G ones to give local car<>car communication for driverless vehicles.

Edited by RobertoS (Fri 14-Feb-20 22:32:58)

For example, I'm ~400m from my cabinet, and get ~76Mbps down. That's on a wired connection. It there was a 5G mast there, I'd probably get gigabit speeds potentially. How is there more loss over a physical wire, than through the air?!

That's the opposite of the question in your OP . You are now asking why a landline is slower than a radio wave. (The Subject can of course be read either way).

They just cannot transmit signals extremely fast and reliably down copper where attenuation is much higher than through air or fibre-optic cable. The capacity is much lower in copper.

Edited by RobertoS (Fri 14-Feb-20 17:31:22)

In reply to a post by jchamier:

How? It can be faster by using different technology. VDSL is capped at 80Mbps in the UK and will have a lot of loss on a 1-mile length of copper. Most people are 500metres or less from their local cabinet.

5G is radio, so it will depend how many users there are using the service at the same time.

Why?

On a side note when Origin operated via Digital Region you could get up to 100mb connection via FTTC (although probably a small proportion).

In reply to a post by RobertoS:

The Openreach VDLS (actually VDSL2) has a maximum connection speed of 80Mbps. Even full 4G easily beats that with line of sight of the mast.

On an early configuration, 2x 20 MHz of bandwidth (FDD) should give 150 Mbps download, and 50 Mbps upload. You can play with this calculator: https://tools.pedroc.co.uk/4g-speed/

5G is far faster than 5G.

Sounds a bit Teresa May! 5G-NR is faster because it has new encoding techniques (e.g. OFDM), and shares the air time in a different way to LTE, which was dramatically different to UMTS of GSM, why the Long Term is the first two letters. Today all 5G NR also needs to be connected to LTE at the same time - this is the Non Stand Alone configuration. I've not yet read of any SA installations.

That's why for instance many more 5G "masts" will be needed than 4G ones to give local car<>car communication for driverless vehicles.

I doubt that use case will ever happen myself, unless cars themselves are transmitters and you're not talking over a traditional cellular network. Many more NR transmitters will be required once we deploy mmWave spectrum, e.g. 60 GHz. I predict there will be little to no indoor penetration.

The UK (ofcom) hasn't yet licensed these frequencies. The US has, and Verizon is building mmWave connectivity in some cities. Note the new Samsung S20 5G does not support mmWave - so buyer beware!

Edited by jchamier (Fri 14-Feb-20 18:22:00)

In reply to a post by AndyPandy:

For example, I'm ~400m from my cabinet, and get ~76Mbps down. That's on a wired connection. It there was a 5G mast there, I'd probably get gigabit speeds potentially. How is there more loss over a physical wire, than through the air?!

That 76Mbps is due to Ofcom/BT limiting to VDSL Profile 17. This was for various reasons including not breaking neighbours still on the old ADSL signal from a far away exchange. Also interference picked up on the wires from heavy industry, or even cars / motorbikes driving past. These wires were not planned for internet access. Other countries have higher profiles and for some people 100 Mbps or faster. G.Fast for example has managed 300 Mbps across the copper wire, but you need to have 100m or less distance. e.g. AT&T U-Verse in parts of America.

With an Ethernet network using Unshielded Twisted Pair most people get 1 Gbps around a home but we are talking 20m to 30m maximum distances between the PC and the switch/router. If you could replace the wire in the street with some shielded coax, (e.g. Virgin Media's cable) then you could achieve 1 Gbps or faster quite easily.

Radio is more complex, if you have a strong signal (from mast to you) and in reverse (you to mast) you can send quite a lot of data quite quickly. However to pay for this you have to share it with other users, this is where 2G(GSM), then 3G(UMTS), then 4G(LTE) and now 5G(NR) are all increasing the total amount of data that is then shared with multiple customers.

I dont see it as VDSL vs 5g I see it as FTTP vs 5g. I am sure FTTP will win out in the long run as the technology develops and 5g becomes over used.

Agreed. Unless there is a constant growth of 5G base stations, in cities a fixed line network should always win. If it�s built properly. A challenge to Virgin Media and OpenReach.

For me it isn't that wireless is faster than copper - it is that different designs are in play. 5G is a raft of different standards and frequencies - the speed and range that someone gets will depend on which particular standard and frequency is in use. Copper is much the same - VDSL is a specific standard designed for a particular task using existing copper, a different config using different frequencies would have different results.

The other unknown will be how many 5G masts there are. For the very high speed high frequency stuff there will need to be lots of masts as the signals won't go far, the lower speed lower frequency will require fewer masts, be better at penetrating buildings but won't deliver the blistering speeds that 5G seems to promise. It is all a compromise and will be designed around the specific requirements of an area/supplier.

In reply to a post by jchamier:

With an Ethernet network using Unshielded Twisted Pair most people get 1 Gbps around a home but we are talking 20m to 30m maximum distances between the PC and the switch/router.

If you don't know what you are talking about do not spread disinformation, otherwise known as complete and utter rubbish.

Ethernet is good on Cat5e cable for 100m at 1Gbps and with the newer 802.3bz standard at 2.5Gbps, on Cat6 cable for 100m at 5Gbps using 802.3bz and on Cat6a cable for ~100m at 10Gbps. You can get 5Gbps on Cat5e at shorter distances and the same with 10Gbps on Cat6, that is situation dependant.

Basically with twisted pair ethernet the distance is 100m. If you could only do 20-30m as you suggest it would be utterly useless in a commercial environment.

It is unlikely anyone has Cat3 or Cat5 left in 2020 (they have been obsolete for over 20 years now) so it's 2.5Gbps for up to 100m as at a minimum for Ethernet. Anything better than Cat6a cable is a complete waste of time, money and resources by the way.

As it was obviously unclear, I meant the distances in the home are, in the average, 20m to 30m. Not the technical distances of 100BaseT.

Edited by jchamier (Mon 17-Feb-20 14:14:20)

Post deleted by MrSaffron

Deleted due to swear word in original post

In reply to a post by jchamier:

Agreed. Unless there is a constant growth of 5G base stations, in cities a fixed line network should always win. If it�s built properly. A challenge to Virgin Media and OpenReach.

FTTP doesn't present any potential risk to health, it could be argued it's another stop gap and isn't really needed they could upgrade 4g

5g installation inside street lights can't be a good idea , then there is deforestation needed because it needs line of sight

Edited by tommy45 (Mon 17-Feb-20 14:53:16)

>5g installation inside street lights can't be a good idea

Explain why it is not a good idea.

>deforestation needed because it needs line of sight

Why would a forest need to be destroyed to deliver 5G? HINT Does not need to happen 5G in 700 MHz will get through a forest, existing frequencies will behave similar to 4G in that respect. The mmWave is intended for dense urban deployment e.g. inside buildings, stadiums or busy streets rather than delivering Gigabit to every corner of a forest.

In reply to a post by MrSaffron:

>5g installation inside street lights can't be a good idea

Explain why it is not a good idea.

>deforestation needed because it needs line of sight

Why would a forest need to be destroyed to deliver 5G? HINT Does not need to happen 5G in 700 MHz will get through a forest, existing frequencies will behave similar to 4G in that respect. The mmWave is intended for dense urban deployment e.g. inside buildings, stadiums or busy streets rather than delivering Gigabit to every corner of a forest.

prolonged exposure to the radiation these transceivers will give off, How do we know 100% that this tech is not a risk to our health? And driverless cars, No thanks i own a car because i actually enjoy driving

Edited by tommy45 (Mon 17-Feb-20 15:25:36)

How do we know?

1. 3G and 4G have used same frequencies for years
2. TV transmitters have used the 700 MHz band at much higher power levels for years
3. mmWave has been used for decades for satellite TV signals

Absolutely no idea what driverless cars has to do with this, beyond some marketing people see 5G and its low latency low frequency bands as being useful for inter vehicle comms.

Stop believing all that you read on facebook

This isn't conspiracy
As for the 3 & 4G variants These use cell site#s which are not installed like 5g which will swamp the area it will be more concentrated that 3& 4G signals are,

Edited by tommy45 (Mon 17-Feb-20 16:15:41)

Has anybody here suggested a conspiracy? A conspiracy by who?

The general thrust of that article, and it's conclusion, seems to be very much what Andrew said. Somebody or bodies in Switzerland causing panic that isn't supported by any evidence.

Nobody can prove that radiation in general isn't harmful. Over-exposure to sunlight kills. Under-exposure to it kills. Nobody really knows the limits either way. From the article:-

Bafu will examine exposure through adaptive [5G] antennas in depth, if possible, in real-world operational conditions. This work will take some time.

It'll take an infinite time if the real world conditions are not created in a mass market.

It all sounds like the Bafu top brass are covering their backsides, so that when the rollout is authorised they can say "Not our fault" if there is an occasional problem.

In reply to a post by RobertoS:

Has anybody here suggested a conspiracy? A conspiracy by who?

The general thrust of that article, and it's conclusion, seems to be very much what Andrew said. Somebody or bodies in Switzerland causing panic that isn't supported by any evidence.

Nobody can prove that radiation in general isn't harmful. Over-exposure to sunlight kills. Under-exposure to it kills. Nobody really knows the limits either way. From the article:-

Bafu will examine exposure through adaptive [5G] antennas in depth, if possible, in real-world operational conditions. This work will take some time.

It'll take an infinite time if the real world conditions are not created in a mass market.

It all sounds like the Bafu top brass are covering their backsides, so that when the rollout is authorised they can say "Not our fault" if there is an occasional problem.

No, one actually said conspiracy theories, but i am aware of them,i can form my own opinions , This 5G has not been scrutinised enough imo and i don't want to be a Guinea pig for them,

In reply to a post by tommy45:

FTTP doesn't present any potential risk to health, it could be argued it's another stop gap and isn't really needed they could upgrade 4g

FTTP is end game for networking. There is nothing better that will EVER come along. The likes of Gigaclear and B4RN which have done P2P installations could upgrade a customer tomorrow to 50Gbps symmetric with a swap out of optics either end of the link using 50G BiDi QSFP28's. Ok that's pricey £2000 if the link is under 10km and £4000 if the link is between 10km and 40km but no change of the fibre required. However those 50Gbps BiDi are brand new to the market, expect them to be much cheaper a decade from now.

It's much cheaper for 10Gbps, you are looking at £100 for 20km capable SFP+'s and the switches needed either end are a lot cheaper too. Even 25Gbps BiDi SFP28's are not that expensive at around £250 for the pair.

No amount of radio will ever achieve those speeds outside large point to point line of sight microwave links. I would note that bang in some CWDM and you could get 450Gbps out of a single fibre using off the shelf components (though this will be pricey).

Finally note some fibre links use lasers beyond Class 1 and do present a risk to human health (that is they can blind you). However all the GPON stuff is Class 1 as is at least 1Gbps BiDi optics for <20km. The further the high power the lasers in general.

What 5G?

The one thats at 700 MHz bands? Uses masts just like now, but a need fewer to cover UK than you do today.

The one thats in the 1800 to 5 GHz bands? Similar mast density to today and will usually be an upgrade to existing mast, or even re-use some frequencies already in use.

The one thats in the 8 GHz to 28 GHz bands? This is the new bit and will be no different in mast density to the Wi-Fi networks that cover a number of city centre areas e.g. Square Mile in City of London and similar frequency bands are in use for things like Sky TV already

So nothing much is particularly new in terms of radiation.

The highest bands such as 28 GHz are such that a coat blocks the signal, does not even need to be foil lined.

In reply to a post by MrSaffron:

What 5G?

The one thats at 700 MHz bands? Uses masts just like now, but a need fewer to cover UK than you do today.

The one thats in the 1800 to 5 GHz bands? Similar mast density to today and will usually be an upgrade to existing mast, or even re-use some frequencies already in use.

The one thats in the 8 GHz to 28 GHz bands? This is the new bit and will be no different in mast density to the Wi-Fi networks that cover a number of city centre areas e.g. Square Mile in City of London and similar frequency bands are in use for things like Sky TV already

So nothing much is particularly new in terms of radiation.

The highest bands such as 28 GHz are such that a coat blocks the signal, does not even need to be foil lined.

Then why is there any need for 5g being installed into street lighting ? https://www.whatdotheyknow.com/request/476049/respon... it's this implementation that is a cause of concern

Edited by tommy45 (Mon 17-Feb-20 17:45:57)

In reply to a post by tommy45:

In reply to a post by MrSaffron:

What 5G?

The one thats at 700 MHz bands? Uses masts just like now, but a need fewer to cover UK than you do today.

The one thats in the 1800 to 5 GHz bands? Similar mast density to today and will usually be an upgrade to existing mast, or even re-use some frequencies already in use.

The one thats in the 8 GHz to 28 GHz bands? This is the new bit and will be no different in mast density to the Wi-Fi networks that cover a number of city centre areas e.g. Square Mile in City of London and similar frequency bands are in use for things like Sky TV already

So nothing much is particularly new in terms of radiation.

The highest bands such as 28 GHz are such that a coat blocks the signal, does not even need to be foil lined.

Then why is there any need for 5g being installed into street lighting ? https://www.whatdotheyknow.com/request/476049/respon... it's this implementation that is a cause of concern

For the higher frequency, high speed, short distance links.

Its only remotely economically viable for high density areas such as city centres or for building wide WiFi/wired LAN replacement. (I kinda think that's bonkers personally, but it is one of the suggested uses for a business to outsource their whole LAN to the mobile networks)

Back in the 1990s there was similar concern in our village when the installation of a first-generation mobile mast was proposed. This led to a public meeting where there was a large vote against the installation due to the "undoubted" dangers to children's health.

The residents of the village have spent the last 20 years trying to persuade the mobile phone networks that they didn't really mean it and could we have a mast please. We still remain a mobile free zone. Be careful what you wish for and what you wish to block on the basis of speculation.

Perhaps you need to look at why lamp posts are a common wi-fi hotspot location in cities...

i.e. have power and some height and are existing furniture with small space inside.

The height means line of sight to a lot of people nearby when using the 8 GHz to 28 GHz bands and this also means while more masts that lower power levels than a bigger 4G mast on top of a block of flats or office building.

In reply to a post by tommy45:

they could upgrade 4g

Which is what 5G is, taking over from 4G+, which added on to 4G, which took over from DC-HSPA, HSDPA, HSUPA, UMTS, which took over from GRPS, which took over from HSCSD, took over from standard CSD, and that took over from acoustic coupler over mobile (sometimes worked on analogue).

In reply to a post by gary333:

In reply to a post by jchamier:

How? It can be faster by using different technology. VDSL is capped at 80Mbps in the UK and will have a lot of loss on a 1-mile length of copper. Most people are 500metres or less from their local cabinet.

5G is radio, so it will depend how many users there are using the service at the same time.

Why?

On a side note when Origin operated via Digital Region you could get up to 100mb connection via FTTC (although probably a small proportion).

Indeed.

In reply to a post by MrSaffron:

Perhaps you need to look at why lamp posts are a common wi-fi hotspot location in cities...

i.e. have power and some height and are existing furniture with small space inside.

The height means line of sight to a lot of people nearby when using the 8 GHz to 28 GHz bands and this also means while more masts that lower power levels than a bigger 4G mast on top of a block of flats or office building.

Not to mention that many LED lights already have their own mesh network anyway for control purposes.

Plus more antennas means better beam-forming.

If it was a swear word, the filter should have caught it. I don't even remember it, but if it's the one I suspect, I guess you never watched much Father Ted.