« Traffic Shaping When Using more than 60 Gbytes a day | Main | "Internet Protection" On A New Prepaid SIM »



Regarding O2 and U900, they have now gone nationwide with their roll-out. They have even modified their coverage checker on their website and ask the user to pick whether they have a U900 device or not so they can display the 3G appropriate coverage. See here.. http://www.o2.co.uk/coveragechecker

Dave B.

AT&T Mobility is only just deploying UMTS in Alaska, and even that is only in the big 3 cities and on the North Slope. I'm skeptical we'll ever see 2G go away up here.


I think AT&T and some other operators can meet their target of switching off their GSM systems due the following thoughts.
First, a mobile operator should make sure that there are no GSM-only mobiles in their network. I recently checked the new shared tariffs from Verizon Wireless and I couldn’t find a tariff to which I would be able to connect with my own cell phone. I need to buy from them a mobile even if I want a Pay as You Go tariff (if I am wrong about this – please correct me). It appears to me that the operator wants me to change my mobile every two years. If AT&T has the same strategy about mobiles, then it can make sure that in a couple of years there won’t be GSM-only mobiles by selling 3G capable phones. Then it would be useful to restrict users to switch off 3G mode on their mobiles. I’m not the specialist in American market and smartphones, but as far as I know when an operator sells a mobile, it installs some programs on it. In other words, the operator has access to the mobile and OS, so I think it can somehow lock the capability of a mobile to be manually switched to 2G-only mode by a user. This will give the operator the possibility to monitor 2G usage in their network. Of course you can derive the data about mobiles and their capabilities from the real network, but how do you know that a 3G-4G capable smartphone isn’t in the area with 2G coverage only? Knowing about 2G traffic in your network, you’ll be able to monitor the trend and decide about switching off or not your 2G network.
Secondly, there is a problem with coverage. It’s clear that 3G coverage should be at least as large as 2G coverage before switching off the latter. Providing 3G coverage as vast as 2G could be a real challenge for an operator as the above comment about Alaska proves. However Alaska hasn’t seen UMTS not because it’s technically impossible, but because the company hasn’t aimed this so far. Usually a company has a budget to spend on capacity, coverage and features, and that money as always not enough to cover all needs. However, if a company aims to deliver 3G coverage as large as 2G and is ready to spend money on this, than everything is possible, at least from technical point of view. One can argue that it’s a lot of money and it’s not economically reasonable. Yes, it’s true, to some extent. There is a solution, however, and it has already been mention by Martin and sounds “Replace your aging 2G and 3G equipment with a new base station that can do both plus LTE on top!" As operators are gearing up for LTE they can choose to use SDR modules (software defined radio). With such module it’s possible so switch from GSM to UMTS or LTE by means of software. Not only an operator replaces their aging 2G equipment, but is also capable to switch from 2G to 3G, thus providing the same coverage (don’t forget about licenses, though, they represent a chunk of money). And modern SDRs are capable of operating in two modes concurrently - GSM+LTE, for instance, which is great.
Thirdly, there is a spectrum issue. SDRs are nice, but they have an issue called instantaneous window, which is 20 – 35 MHz in modern modules (varies from manufacture to manufacture). For instance, a GSM1800 module can operate in the whole 1800 band (1805 to 1880 MHz), but the frequencies assigned shouldn’t vary for more then 20-35 MHz. The same with concurrent mode, both technologies (GSM+LTE, for instance) should fit into that window. This restriction has to deal with Power Amplifier characteristics and can be mitigated in the future. I’ve heard that the US operators are planning to make some swaps of their spectrum among each other, and if they think through such issues carefully, they can derive from SDR as much advantage as possible.
So, supposing we’ve done the aforementioned and our network has been equipped with SDR, we’ve thought through spectrum issues and in our network there aren’t 2G-only phones. Now we monitor the network, observe how much 2G traffic we have and where, how many 2G roamers visit our network and how much money they leave here, and one day, judging all pros and cons, we’ll dare to switch GSM off.
The above represents my opinion, so please feel free to criticize. And I apologize for the long comment.


While normal end customer devices might not be a problem of being replaced by UMTS/LTE capable ones over some years, I believe that the whole M2M market may have a much stronger influence on a potential decision to phase out GSM.
I guess that still up to now most of the M2M business uses GSM/GPRS/EDGE devices because most scenarios don't rewuire high data rates and also these are optimised for low power consumption. A GSM module build in any kind of black box with special developed services on it can't be replaced very easily.
So my personal feeling is more that the operators will keep a minimum of resources/spectrum available for these legacy devices (at least as long the spectrum licence plans of the regulatory authorities does not go into a different direction).

The comments to this entry are closed.

My Photo

The Books to this Blog

Secure Hotel Wi-Fi Sharing

My Pictures on Flickr

  • www.flickr.com
    martin.sauter's photos More of martin.sauter's photos


  • Clicky
    Clicky Web Analytics