The era of wireless broadband is upon us and is gaining momentum fast. I expect the launch of Apple’s 3G iPhone in July will accelerate the use of wireless broadband applications considerably and no doubt the arrival of this device in the marketplace will stimulate a slew of other competitive devices adding fuel to the fire. 

All this means that subscribers are going to expect to be able to use more and more applications with different bandwidth and quality of service requirements on a broader and broader scale.  

The trouble is that mobile service providers have not geared up for this demand and nowhere is this more true than in the ‘backhaul’ part of the network – Meaning the connections from the radio base stations back to the control equipment in the network such as Radio Network Controllers (RNC’s) 

The main problem for most service providers is that their backhaul infrastructure was not designed to support wireless broadband usage and does not provide adequate bandwidth – In fact the majority of existing connections to the base stations are made over (T1 and E1) copper circuits, many of which are leased from third parties.  

The service providers short term response to providing adequate bandwidth to base stations is to add more T1 or E1 circuits – But when you apply this over thousands of sites it means that leasing costs go through the roof and it is not a viable long term solution.  

Furthermore, different wireless broadband applications have different bandwidth and quality of service requirements that copper infrastructure simply wasn’t designed to support – Video for example requires more real time bandwidth than e-mails, which use less bandwidth and can be queued without degrading the quality of the service.  

The combination of these factors means that in order to meet customer expectations and of ‘carrier grade quality of service performance’ for wireless broadband – Service providers need to migrate their backhaul networks towards wholly IP infrastructure immediately! 

But how can they plan for this migration, when there is no historical demand information for wireless broadband from which to work – To plan you need to understand where and when demand is appearing. And to exacerbate this problem, the demand appearing at the base stations and on the backhaul network needs mapping into the core network infrastructure, to ensure there is adequate end to end capacity – Otherwise capacity bottlenecks will occur and again end user quality of service will suffer.

 To maintain a competitive advantage, keep costs under control and have the right bandwidth available at the right time, throughout the network – It is imperative that service providers move quickly to deploy solutions for optimal bandwidth to the base station – Short term solutions include deploying edge routers to compress traffic over copper. But there is no doubt that these will have a limited life before moving to wholly IP solutions such as Metro Ethernet, X-DSL and Cable for example. 

In tandem with this they must quickly evolve their traditional planning processes to leverage more real time demand information into smart tools that enable them to make rapid planning decisions on ‘what if’ design scenarios. 

If they do not do this, they run a serious risk of losing customers because of inadequate network performance and quality of service, backhaul and other costs skyrocketing out of control, or at worst a combination of both!