Wasn’t 3G not supposed to be about voice and SMS but whizzy new services: video calling, football highlights, interactive games. If 3G is as good as the marketing hype then why do we need 3.5G at all?
The first thing that turned out to be more hype than performance from 3G was the data rate. Coverage of 3G networks is still patchy – meaning some of the time data users are back. on GPRS rates.
The other side to throughput is latency – the end to end delay from a packet being sent from a game server (say) to it actually reaching your 3G terminal. Again there is quite a lot of variation in3Gsystems but values typically lie in the range 200–500 ms. This is a very high latency by IP and Internet standards – meaning that interactive games and real-time voice and video applications are impossible to use on most existing 3G networks.
The next technical issue with 3G is related to the nature of W-CDMA. In W-CDMA systems the interference caused by any transmission is ‘‘smeared’’ over the entire carrier – with the result that the capacity of the system is essentially limited by the background interference level. SinceUMTS employs a frequency reuse factor of 1 (i.e. all the cells/sectors use the same frequency or chunk of spectrum) this interference also comes from all neighbouring cells. However, it is estimated that 70% of CDMA interference is generated within the same cell – one of the reasons a frequency re-use of 1 is possible. In a lightly loaded cell the range of the cell can be much further than whenthe cell is heavily loaded and the interference level rises – this is known as cell breathing as shown in the illustration above.
In addition to causing cell breathing the fundamentals of W-CDMA means that to achieve a decent capacity the system has to support soft handover: a terminal is receiving the same radio frame from a number of base-stations and combining them to achieve a much higher signal to interference ratio than would be the case for any one of them. This allows the system to run at higher interference levels (i.e. higher capacity) but demands that voice and data streams are delivered to and from base-stations with very precise timing. In 3G this is solved by using ATM switching – ATM being able to control the delivery of frames very tightly in time. In hindsight, however, ATM has been declining (replaced with Ethernet or MPLS) and the ATM legacy of 3G is seen as inflexible and costly going forward.
In addition to network issues there have also been terminal problems – with the industry taking five years or so to come up with handsets that are able to match GSM for weight and talk/standby performance. IPR (Intellectual Property Rights) have also been a major cost of 3G – with an estimated 8–28% of each handset going in licensing fees.
3.5G and LTE are the mobile industry’s answers to fixing these issues – WiMAX is the computer industry’s answer.
In addition to network issues there have also been terminal problems – with the industry taking five years or so to come up with handsets that are able to match GSM for weight and talk/standby performance. IPR (Intellectual Property Rights) have also been a major cost of 3G – with an estimated 8–28% of each handset going in licensing fees.
3.5G and LTE are the mobile industry’s answers to fixing these issues – WiMAX is the computer industry’s answer.
