I’ve noticed that when you’re on a fast network and the reception gets bad, you get downgraded to a slower network (e.g. 2G). I’ve always thought that the slower networks are just a result of poor reception on the faster networks; maybe less channels are used or something. But today I realised that this may not be the case.
Are the 2G, 3G and 4G networks independently maintained? Is it possible to build a 4G network without 3G or 2G? And is there any reason (other than compatibility and upgrade costs) why the older networks always have better coverage? I mean, do 2G cell towers and frequencies have better range or something? If I were to build a new network, could I go straight to 4G and have as good coverage as a 2G network?
The networks are independently maintained, and operate on different frequencies. In most of the world, we progressed from 2G to 3G to 4G. There are a few corners of Asia that skipped the 2G step and started at 3G.
The older networks tend to have better coverage because they are older, and have had more time for carriers to build them out.
There is no technical reason that you couldn’t build a kick-ass 4G network with awesome coverage. But it would cost a lot of money…so carriers maintained their legacy networks to handle the gaps as they rolled out the higher speeds in the most populated areas.
US carriers are phasing out their 2G networks. We use embedded 2G GPRS modems, and this phase-out is really going to hurt us. AT&T plans to completely shut down their 2G network by 2017, but they have already started the “turn-down” in parts of the US.
Wow thanks, I always thought 2G was just lossy 3G, like how a poor WiFi signal decreased speed. Do the different technologies have similar ranges? Is the decreased 4G coverage solely due to lack of upgrades?
Most of it is upgrades and frequency. The right to use a specific frequency in a specific region is bought from the US government at auction, usually for billions of dollars.
So not only do you have to deploy new equipment everywhere that’s 4G capable, you have to have the frequencies available to deploy it. So companies don’t just want to be able to get rid of the old equipment and such, they want to be able to move the frequencies to 4G and make it more capable.
In theory, range is going to be largely dependent on the specific frequency, not so much the technology (kind of like how you can get AM radio signals much further away than FM), however, in practice, I believe 4G technologies are more efficient in their use of the spectrum. If I’m planning out a cell network and I can only get 10 simultaneous calls on a tower, that’s going to be a much more limiting factor than any theoretical signal range.
Finally, backhaul can limit 4G deployment as well. Backhaul being the connection from the cell tower back into the provider’s network. 2G networks pretty much only need 1 or 2 T1 lines to handle the traffic, 4G sites need 20/50/100meg connections for good customer experiences. Needless to say, the further out you get, the more expensive that tends to get.
Certainly some networks have been known to stick a 4G site at the end of a T1 line just to meet marketing numbers, but it’s generally not a good long-term plan.
4G or LTE as it is in Australia is a data only network that is used to move data off 3G networks. This is needed as more and more data hungry devices are on the network. All 3 carriers in Australia have some degree of 4G augmentation.
4G uses different bandwidths and as a pure data play was built for higher speeds from day one.
The issue of coverage is that they haven’t rolled out the infrastructure as yet and yes the cell range is not as large.
GPRS is 2.5G. EDGE is supposedly 2.75G, but in practice I find that EDGE is slower than GPRS, to the extent that having the “E” show up on my phone is basically equivalent to “no network connection”. Although I think that is an iPhone problem. Certainly it’s a common complaint if you Google it.
To answer one of the OP’s questions, yes, there are fundamental differences in the 2G, 3G, and 4G ways of encoding the information on radio signals.
You can think of it like the differences between an older analog TV broadcast and a newer digital broadcast - if there’s not a digital channel, your TV could still work with an analog broadcast (back when they were still broadcast). With cell phones, it’s been digital since 2G, but the specific way that the info is encoded on the radio signal is different in each generation.
You are not being “downgraded” from 4G to 2G, as in a lower speed is negotiated and that equals a change from 4G to 2G. It doesn’t work that way as there are fundamental differences between 2G, 3G, and 4G. What has happened is that you can no longer find a 4G network, nor a 3G network, and the only available happens to be a 2G network. Remote areas or areas with low population counts are going to be on the tail-end of any service upgrades.
If you are in an area that has a great many users, you would still be on the 4G network, just at a lower data rate. It could be so bad that the 4G network has no availability for you and then your phone will look for a 3G, and so forth. Or, the recieve signal strength is too low on the 4G and a 3G is at a higher signal strength.
The 2g/3g/4g status is not dependent on frequency - there’s no rule of physics that prevents 2g working on the higher carrier, and 4g can run on all the same frequencies (so as to allow all frequencies to be used in city areas.), but as it happens when they added 3g,4g to a tower, its generally true that they leave 2g on the lower frequency…
Following up on comments like CurtC’s; The newer/higher G networks typically support higher data rate, but at a slight loss in signal sensitivity.
Most phones are made to be backwards compatible with older networks, and support all older G’s.
So, if you don’t have enough signal for a good 4G communication, there may still be enough signal available to/from an older G transmitter, and your phone switches to that, resulting in the lower data rate.
Of course another practical case is to supporter older G’s for the times when you’re in an area where the newer G simply isn’t available yet.
Another relevant case is the USB port on computers. There have been a similar series of versions, usually each faster than the last. But you can plug a USB 3.0 device into an older USB 2.1 computer and it will go back to the slower speed.
And Ethernet is like that too, now that I think about it. A Gigabit-capable device can be connected to an older 10/100 switch and it will drop back to that speed. That can also happen with a Gigabit-to-Gigabit connection if the cable length is too long.
Hey, if they’re different networks, can phones swap calls from 4G to 2G without dropping, like how they change base stations?
I thought 4G might have a shorter range because of less redundancy and error-correction to increase speed. Like how increased data density in disks and discs make it easier to lose data.
At this time, voice calls aren’t done on the 4G networks anyway. Your 4G phone uses that radio for data, but for voice calls, it uses its 3G radio. That’s because 2G and 3G had dedicated voice handling in its design, but 4G is completely Internet packets. Companies are working on doing Voice Over LTE (VoLTE), which is similar to VoIP, but complicated by radio transmission difficulties, so it’s not really out yet (I’m sure someone will come along and point out that it’s in use somewhere, but it’s uncommon).
Handover from 3G to 2G does happen without dropping, and should with 4G as well.
Not only are they different frequencies, but they’re different protocols – they’re very different in terms of how they work. 2G and 3G were specific technologies. 4G just means “better than 3G” and there are at least 4 different technologies with different advantages and disadvantages. 5G, when it comes, will be even more confusing, I bet.
I work for a large network equipment vendor. A few years ago, the head network designer for one of the major service providers was telling us about his vision for the future. He wanted someone to be able to watch HD video on a phone, starting say in a bar on its wifi, then walk outside and lose the wifi but and pick up the mobile network, without dropping video frames. That was just one example. He said that before long we’d be getting more bandwidth on our mobile devices than we did then on our wired networks. When challenged about what applications would need such bandwidth, he said, “Oh, don’t worry about that. If you give everyone the bandwidth, people will come up with great ways to use it!”
The first phone I worked on did a lot of stuff like that. It was great. I could place a call over WiFi in the building…walk out the door and seamlessly hand-over to cellular. Walk back in and pop back to WiFi.
This was back around 2005, when streaming data on phones wasn’t nearly as big as it is now. The data handovers are harder, but I can imagine a few ways to make it work…probably playing with a tunneling strategy of some sort.
And I completely agree with the quote about people coming up with great ways to use bandwidth.
The problem, of course, is that one of the main points of offloading to WiFi is that you’re not going to your telco for the traffic. Tunnel to the provider, and that means you’ll still hit data caps and whatnot.
Something on the app level may be better. I know right now if I’m logged into YouTube, watch part of a video, and then stop, and later log in from a different location/system it will remember where I stopped.
Or at least it’s close, it may be going back to the latest loaded chunk or something right now (and sometimes it doesn’t work at all), but I presume you can get more granular if you wish.