Tomorrow's mobile wireless [Under Telecom]
The way things have moved from GSM to GPRS to EDGE to UMTS to HSPA to LTE, I imagine that tomorrow's radio access and core backbone will be much more simplified, stronger (in terms of parameters like spectral efficiency, bit rate, edge bit rate, latency in both user and transmission planes etc.), scalable, and independent of each other.
Current
Current 2G/3G wireless networks (especially GSM/UMTS ones) look like as below. You may also refer GSM and UMTS articles).
Mobile terminal, radio access network elements, core network elements and their interfaces are all more or less specific to wireless systems. Compared to 2G, 3G core network interfaces are based on open or widely used standards like SIGTRAN, IP/IETF etc. (refer related UMTS/Iu interfaces). Though radio access is more independent of network protocols, differentiation is made between CS and PS services.
In 3G+ (Rel 5/6), radio access is more and more packet based, and spectral efficient. But yet more is desired for edge bit rate and latency to support consistent performance and high end services (like real time interactive sessions e.g. gaming, high quality video call etc.). In 3G+ networks, 2G/3G elements are still retained, but they are modified and number of new "open" elements are added like Media GateWay etc. But this in fact has made network architecture more complex.
Future step 1
I expect that future networks would have simpler architecture and interfaces would be based on open/widely used standards like SIGTRAN, IP/IETF as shown below. Instead of having elaborate radio access network elements, we will have fewer elements forming an interface between mobile terminal and core network. The interface between RAN and CN would be such that core network will have fewer functions to handle (specific to wireless like mobility ,authentication, ciphering etc.). This would make core network architecture also simplified. We may or may not have direct connection with PSTN, instead PSTN may have some connection with Internet or IP cloud.
Looking at current developments, future radio access is likely to be packet based running on OFDM/OFDMA. Network signaling stack would be thinner or may even be based on open standards. This would result in least possible latency in both user and transmission planes.
Future step 2
The next future step would be: change in the way mobile terminals will use services from operator. The users can register to services from more than one operator. Based on service and radio conditions, mobile terminal can dynamically select appropriate wireless system (e.g. LTE, WiMAX) and operator. This can be configured in mobile terminal or just be made automatic. Of course, mobile terminal will be required to have related wireless system stacks and hardwares.
For example, say user is currently on a VoIP call over LTE and it finds that there is a stronger 802.16e/WiMAX cell available from same or different operator. Mobile terminal should be able to handover or choose this new cell without affecting the continuity of service. As can be guessed, this require lot of work on interworking between various wireless systems and operators.
Users can always choose the appropriate access and operator as per pricing or perceptive quality for the individual services.
Needless to say field usage of LTE and WiMAX will determine the future direction.
| Copyright © Samir Amberkar 2010 | § |