If you know of an article that sums up what this standard protocol does, how it does it, and why that wasn't possible until now, then could you please post a link?
> Is there any chance 5g will break the walls that prevent real unlimited data?
For providers to provide true unlimited data, you have to have true unlimited bandwidth.
Wireless connectivity is provided through a shared medium. For comparison you have wired networking where every customer has a dedicated medium, the wire, for themselves. This is what makes all the difference between actually unlimited and capped plans.
> Wireless connectivity is provided through a shared medium. For comparison you have wired networking where every customer has a dedicated medium, the wire, for themselves.
Isn’t wire also a shared medium? It’s not that there is an end-to-end circuit when you connect to another machine on the Net; your packets will be send along with others’.
> Isn’t wire also a shared medium? It’s not that there is an end-to-end circuit when you connect to another machine on the Net; your packets will be send along with others’.
It does only because the end user has slow speeds. If anyone could stream 4k, the bandwidth for a whole city during peak hours would probably not fit through that fiber anymore.
A strand of fiber can carry something like 20 Terabits/s. Streaming 4k takes less than 20 Megabits per second. That's still a million people streaming 4k simultaneously.
That's a really tricky question. At some point, the packets are going to share a link with those from other customers, possibly oversubscribed...
Cable internet is clearly a shared medium at both the physical and link layers.
Fiber like GPON is shared physically-ish, but DWDM isolates individual waves so thoroughly that it's hard to think of that as a shared physical medium unless you're thinking about backhoe fade and the like.
> Fiber like GPON is shared physically-ish, but DWDM isolates individual waves so thoroughly that it's hard to think of that as a shared physical medium unless you're thinking about backhoe fade and the like.
This is incorrect. GPON does not use DWDM, or any individual waves per subscriber. GPON is a shared medium with one wavelenght used in the upstream and one wavelenght used in the downstream direction. Each subscriber is allocated bandwidth on the shared medium.
Not even next generation PON uses DWDM. Next generation PON still uses on upstream and one downstream wavelenght per PON tree, but there can be multiple PON trees on the same backhaul fiber. It's more like a CWDM overlay of PON trees.
The only PON that uses DWDM is WDM-PON, but it isn't deployed commercially at scale yet.
The difference is that if the cable gets full you can put a new cable next to the other one, and double your available bandwidth. If the ether is full, it's full. It's also mixed with all kinds of signals, obstacles, power losses, etc.
But practically, Massive MIMO, small cells and 30+ GHz frequencies effectively mean that it's possible can provide more bandwidth than near-term consumers can use with 5G in most situations, if you just throw enough $$ at it.
30+ GHz require line of sight. It's not feasible to deploy stations, except maybe indoors like shopping mauls.
More realistic is the 700 MHz range. 2100 TDD is also pretty much unused, as well as 2600 TDD. Then again you can reuse existing frequencies that were previously used for GSM/UMTS/LTE.
> Wireless connectivity is provided through a shared medium. For comparison you have wired networking where every customer has a dedicated medium, the wire, for themselves. This is what makes all the difference between actually unlimited and capped plans.
if you are willing to sacrifice on mobility aka fixed-wireless-access, there are companies that are trying to do this e.g. via beamforming, which essentially tries to make wireless a 'switched-media' rather than 'shared'.
Networks are usually capable of this. (See examples in Europe). It is a deliberate choice from the ISPs to not offer unlimited data packages. You can make much more money with selling overprized "1 GB packages". It is a business issue, not a technical one.
> true unlimited data, you have to have true unlimited bandwidth.
Semantics: no, you just need to throttle each connection such that ever user can have continuous data flow given the available back-haul.
Can your ISPs infrastructure handle everyone saturating their connection? It’s all about QOS vs Contention Ratio. It’s unlikely everyone will want to stream ten 4k videos to their home or smartphone all the time, but so long as telephony and streaming video work most of the time for most people we can probably consider that unlimited in the common parlance.
Netflix bandwidth usage is pretty predictable and limited, as it's a realtime service that requires a human at the other end to consume the stream. Probably limiting-happy ISPs will sell you a Netflix worth of monthly traffic allowance at a reasonable price.
Same in Austria (but 20 EUR instead of 10). LTE is a competitor for ADSL and Cable here and LTE providers sell WiFi routers that use LTE instead of ADSL/Cable as uplink.
I recently replaced my home ADSL connection with LTE (for 20 EUR a month: https://www.yesss.at/tarifoptionen/unlimited) and get approx 2x the download bandwidth and 10x the upload bandwidth for the same price as my previous ADSL connection. Even when there's congestion (which doesn't happen a lot) it's still significantly faster than ADSL.
Latency in LTE is, I think 5 ms. The rest from the delay is normal network delay and not an issue of LTE. So with 5G (with everything else being the same) will reduce the delay by about 4 ms.
Not really a network engineer so cant really help with that. The 20ms is just my experience and the 1 to 2ms is the theorical number given by the people designing 5G. Found this writeup but still not really sure beyond improvements to the slot-based framework and higher frequency as being the reasons
All these latency numbers vary a lot based on connection quality. As I understand you basically need line of sight for 5G to work at its maximum frequency and will fallback to what is basically 4G when such a connection is not possible. But then again the antennas/stations for 5G should be much smaller and thus easier to deploy closer to the end users.
Latency will go down with 5G if you pay for it. How much is the big question.
There are 3 sides to 5G:
- eMBB: massive boradband --- high throughput, mmWaves;
- URLLC: ultra reliable and low latency, for industrial applications typically;
- mMTC: massive IoT. Cheap and low power.
The network can handle all 3, but as a device or user that doesn't mean you'll get all the benefits at the same time, some choice will be required. The reduced latency always has some cost impact.
Today the LTE air latency is 4-5 ms, the rest is in the core network. 5G has some air interface changes to reduce the air contribution to 1 ms, in the context of URLLC. For basic users, it'll likely be like LTE (already pretty good). There may be some general improvement if they optimize the core latency for all, but TBC.
How is latency for you?
When watching a stream or downloading latency which playing an online game jumps up to ~1s (1000ms) in the game. Croatian TELE2 provider, indoors at -92 dBm.
Usual latency is about 15-20ms doing ookla speedtest but "real life" it's about 40-50ms.
This issue is likely to be bufferbloat [1] and not directly related to LTE. If you can put a properly configured router in front of the modem that may help. I haven't looked into it but it seems recent OpenWRT have the necessary support to mitigate bufferbloat.
Depends on the server you want to reach. Same country you can get latency ~20 ms. If you get lucky also lower. But haven't seen anything below 10 ms. Also "-92 dBm" doesn't say much. There are 4 relevant values for LTE: Rssi (probably your -92, the most irrelevant one), Rsrp, Rsrq and Sinr.
France here, for 20€ I got 100GB domestic + 15GB EU roaming, which is very much asymptotic to unlimited for me. Considering moving to 50GB (domestic+EU roaming) for 25€ and stop worrying when I cross a border.
50 or 100GB isn't really a whole lot.
I couldn't live with those limits.
OUr 2 person household usually generates/consumes 600GB-1TB traffic / month. IPTV, Games, Downloads, Uploads.
Who said you need to funnel everything through your mobile plan? Of course downloading a single Xbox One game would burn through the whole allowance but why on earth would I do that when I'm at home? Maybe completely unlimited plans will come some day, but in the meantime I don't feel limited at all just by being pragmatic and sensible.
"Funneling everything through your mobile plan" was the basis of this comment thread (asking about when we can stop worrying about WiFi and just watch Netflix on our phones)
> asking about when we can stop worrying about WiFi and just watch Netflix on our phones
Streaming Netflix is a blip on the radar and effectively consumes not that much, as witnessed while I was on vacation for two weeks and binge-watched on the train (three times 7 hours didn't even cross the 5GB mark).
Not entirely, no. From what I understand, a provider is allowed to limit you in roaming to the volume you'd get at a set rate (I believe right now at around 7 €/GB, with set steps reducing it the next years) for the price of your plan.
So if you pay 50 € per month, your provider is allowed to cap you at 50€/(7€/GB) = ~7 GB while roaming.
Deutsche Telekom is currently offering 24 hours of real unlimited data for free in Germany at every World Cup game of the German soccer team. (Their usual plans offer single-digit GB data per month.)
I would expect that to be a near worst case scenario from a bandwidth perspective. Everyone all at once suddenly has unlimited data during roughly the same 24 hour window. Apparently their existing infrastructure is good enough to handle such a spike.
This is limited somewhat in that this is a manual process (you have to click a link on a website to get it and then click a link in the confirmation SMS), but still. I think they could offer unlimited data at a price well below what they are currently offering (€80 per month for unlimited data – actually that’s not so bad, only a couple months ago the situation was much worse).
I would hope that this is really their goal - But I'm not too sure about it. I expect them to aim to only slightly surpass their competitors. Maybe this is actually to signal to them their capabilities?
Real unlimited data is, well real. At least in Austria. Prices range from 20-50 €. It is only usable in Austria though, but that's not an issue. This has been this way even before LTE.
Define Unlimited. Because there is no such thing as unlimited.
The average household uses 250GB per month with broadband.
An HEVC 1Mbps stream should offer decent enough quality on a mobile phone. That is only 30GB a month if you watch 2 hours everyday. And if you live in EU that is hardly a problem even with 4G network. It is a problem with high population density such as Tokyo, Hong Kong or Philippines. But 5G will fix all those capacity problem.
Most of the problem you read / have are likely from US. In 2016 the data show US has 3 times less cell tower per area when comparing to UK or other EU countries, but US is also much less populated. I have no idea how the business maths works out or if there are any regulation that requires them to cover most part of US.
From the first link: "Comparing OFDM to LTE today we find a better scalability to a much lower latency (an order of magnitude lower round-trip time [RTT] than LTE today) in OFDM."
Doesn't LTE already have quite good latency properties?
One data point: LTE gives me 65-90ms ping RTT in reasonable non-crowded indoor conditions. Subjectively, working interactively via LTE is totally fine for me. Compare that to UMTS (3G T-Mobile DE) where I think RTT was more like 350ms and working interactively is a pain.
As I understand it, not good enough for "vehicular communication, industrial control, factory automation, remote surgery, smart grids and public safety applications" [1]. A lot of the improvement is down to the frame structure (i.e., how user and training data are distributed in time and frequency). See "Frame structure" in [1].
I've heard serious concerns about lack of safety testing and regulatory oversight. The problem is that it's difficult to find credible sources pro / con. Would be interested in reading if you have any to share.
The first article feels confused, as if the author copy-pasted from various press releases without understanding the topic. OFDM is not new at all for example.
The PPT is from 2015, when 5G was still just a bunch of ideas.
Maybe someone could help me with a LTE issue here...
I'm in Croatia, about 1km from the tower, -70 dBm.
The (ZTE MF286) router has a MTU 1500 and MSS 1300 setting.
I can change this setting.
In Germany I have a VDSL (Telekom DSL) 100/40 MBit/s home network.
A Fritz!Box 7590 is on the front and it's providing VPN access over old age cisco type xauth1+psk (aka user pass and group pass).
When I try to connect in Archlinux using Networkmanager-vpnc I can connect to the German home network.
But I can't ssh into any boxes in the German network.
I am able to do that over Croatian DSL without issues.
The adapter's MTU is 1500 and the tun0 MTU is 1412.
So this works over DSL but not over TLE.
When I change the adapter's MTU down to 1420 (reading libreswan's site I stumbled upon this value) I can login via ssh and seemingly also use it as it's supposed to.
But then other issues arise...
TLS Handshakes timing out. As if a dog is trying to chase its tail.
I've canceled the TLE service subscription in the 14 day period because of that and I'm going to return the router today in 4 hours tops.
This this is kind of a last straw kind of help request.
Otherwise I'm stuck on a 14/1 Mbit/s VDSL connection which is painful to say the least, but at least I'm able to ssh into my German home network without problems.
Do you remember when 4G has become the new standard for mobile communication? Almost 10 years later, many networks have just reached the connection speeds required by the 4G standard. Given the time it took to reach true 4G, it may seem premature to think of building a network ready for 5G.
66 comments
[ 3.7 ms ] story [ 128 ms ] thread[1] https://www.ericsson.com/en/ericsson-technology-review/archi...
Another somewhat high-level PPT http://www.5gsummit.org/docs/slides/John-Smee-5GSummit-Silic...
I've worked on LTE; the 5G NR physical layer look to me to be very similar to LTE. See http://www.3gpp.org/DynaReport/38-series.htm and https://www.keysight.com/upload/cmc_upload/All/Understanding... for gory details
If not, what must happen in order to allow everyone use Netflix on their phones without wifi?
No. For a while, perhaps. But demand will scale to meet supply as new uses for the network are found.
For providers to provide true unlimited data, you have to have true unlimited bandwidth.
Wireless connectivity is provided through a shared medium. For comparison you have wired networking where every customer has a dedicated medium, the wire, for themselves. This is what makes all the difference between actually unlimited and capped plans.
So, no. Not today. Not anytime soon.
Isn’t wire also a shared medium? It’s not that there is an end-to-end circuit when you connect to another machine on the Net; your packets will be send along with others’.
not since you have had ethernet switches.
That's a really tricky question. At some point, the packets are going to share a link with those from other customers, possibly oversubscribed...
Cable internet is clearly a shared medium at both the physical and link layers.
Fiber like GPON is shared physically-ish, but DWDM isolates individual waves so thoroughly that it's hard to think of that as a shared physical medium unless you're thinking about backhoe fade and the like.
With DSL, do you consider the DSLAM?
Etc
This is incorrect. GPON does not use DWDM, or any individual waves per subscriber. GPON is a shared medium with one wavelenght used in the upstream and one wavelenght used in the downstream direction. Each subscriber is allocated bandwidth on the shared medium.
Not even next generation PON uses DWDM. Next generation PON still uses on upstream and one downstream wavelenght per PON tree, but there can be multiple PON trees on the same backhaul fiber. It's more like a CWDM overlay of PON trees.
The only PON that uses DWDM is WDM-PON, but it isn't deployed commercially at scale yet.
More realistic is the 700 MHz range. 2100 TDD is also pretty much unused, as well as 2600 TDD. Then again you can reuse existing frequencies that were previously used for GSM/UMTS/LTE.
if you are willing to sacrifice on mobility aka fixed-wireless-access, there are companies that are trying to do this e.g. via beamforming, which essentially tries to make wireless a 'switched-media' rather than 'shared'.
Semantics: no, you just need to throttle each connection such that ever user can have continuous data flow given the available back-haul.
Can your ISPs infrastructure handle everyone saturating their connection? It’s all about QOS vs Contention Ratio. It’s unlikely everyone will want to stream ten 4k videos to their home or smartphone all the time, but so long as telephony and streaming video work most of the time for most people we can probably consider that unlimited in the common parlance.
(And to nitpick, nothing really prevents unlimited traffic today with LTE - some places have multiple operators offering it: https://www.linkedin.com/pulse/finnish-like-unlimited-mobile...)
I recently replaced my home ADSL connection with LTE (for 20 EUR a month: https://www.yesss.at/tarifoptionen/unlimited) and get approx 2x the download bandwidth and 10x the upload bandwidth for the same price as my previous ADSL connection. Even when there's congestion (which doesn't happen a lot) it's still significantly faster than ADSL.
What about latency?
https://www.edn.com/electronics-blogs/5g-waves/4460346/5G--T...
All these latency numbers vary a lot based on connection quality. As I understand you basically need line of sight for 5G to work at its maximum frequency and will fallback to what is basically 4G when such a connection is not possible. But then again the antennas/stations for 5G should be much smaller and thus easier to deploy closer to the end users.
There are 3 sides to 5G: - eMBB: massive boradband --- high throughput, mmWaves; - URLLC: ultra reliable and low latency, for industrial applications typically; - mMTC: massive IoT. Cheap and low power.
The network can handle all 3, but as a device or user that doesn't mean you'll get all the benefits at the same time, some choice will be required. The reduced latency always has some cost impact.
Today the LTE air latency is 4-5 ms, the rest is in the core network. 5G has some air interface changes to reduce the air contribution to 1 ms, in the context of URLLC. For basic users, it'll likely be like LTE (already pretty good). There may be some general improvement if they optimize the core latency for all, but TBC.
[1] https://www.bufferbloat.net/projects/bloat/wiki/Introduction...
Streaming Netflix is a blip on the radar and effectively consumes not that much, as witnessed while I was on vacation for two weeks and binge-watched on the train (three times 7 hours didn't even cross the 5GB mark).
So if you pay 50 € per month, your provider is allowed to cap you at 50€/(7€/GB) = ~7 GB while roaming.
When greed is not the problem (due to competition, etc) then those markets already have unlimited data.
I would expect that to be a near worst case scenario from a bandwidth perspective. Everyone all at once suddenly has unlimited data during roughly the same 24 hour window. Apparently their existing infrastructure is good enough to handle such a spike.
This is limited somewhat in that this is a manual process (you have to click a link on a website to get it and then click a link in the confirmation SMS), but still. I think they could offer unlimited data at a price well below what they are currently offering (€80 per month for unlimited data – actually that’s not so bad, only a couple months ago the situation was much worse).
The average household uses 250GB per month with broadband.
An HEVC 1Mbps stream should offer decent enough quality on a mobile phone. That is only 30GB a month if you watch 2 hours everyday. And if you live in EU that is hardly a problem even with 4G network. It is a problem with high population density such as Tokyo, Hong Kong or Philippines. But 5G will fix all those capacity problem.
Most of the problem you read / have are likely from US. In 2016 the data show US has 3 times less cell tower per area when comparing to UK or other EU countries, but US is also much less populated. I have no idea how the business maths works out or if there are any regulation that requires them to cover most part of US.
Doesn't LTE already have quite good latency properties?
But I find reliability the far bigger problem with LTE than performance if everything goes right. Not sure how 5G will perform there.
[1] https://www.ericsson.com/en/ericsson-technology-review/archi...
https://www.edn.com/electronics-blogs/5g-waves/4460346/5G--T...
The PPT is from 2015, when 5G was still just a bunch of ideas.
patience you must have my young padawan, patience ! give it a few years...
When I change the adapter's MTU down to 1420 (reading libreswan's site I stumbled upon this value) I can login via ssh and seemingly also use it as it's supposed to. But then other issues arise... TLS Handshakes timing out. As if a dog is trying to chase its tail. I've canceled the TLE service subscription in the 14 day period because of that and I'm going to return the router today in 4 hours tops. This this is kind of a last straw kind of help request. Otherwise I'm stuck on a 14/1 Mbit/s VDSL connection which is painful to say the least, but at least I'm able to ssh into my German home network without problems.
____________________________________________________ https://dltutuapp.com/ https://kodi.software/ https://luckypatcher.pro/