What is in a Cloud?

Thu, 01/19/2012 - 7:20am
Debbie Greenstreet, Texas Instruments
Debbie GreenstreetAs I look out of the airplane window upon my return from the two day, Linley Carrier Conference in San Jose, with clouds in view, above to the side and below us, I am reminded of my own presentation at this conference: Cloud base stations, Revolution or Evolution? I received feedback from several of the attendees that the high level concept I shared of the Cloud RAN concept was helpful for them. After all, many of the attendees weren’t immersed in the evolving architecture and trends of wireless networks, and they pointed out that “cloud” is such a loose term and has so many meanings already.

C-RAN towerSo, for clarfication, no, we are not talking about ‘cloud computing’, receiving attention elsewhere in the high tech market, nor are we referring to the often generic reference of the internet as simply, ‘the cloud’. Rather, for Cloud Radio Access Network (RAN), ponder this. The Cloud RAN is based on the premise (backed with measured data), that at any given time in a wireless network, there are radio base station processing resources that are powered up, yet sitting idle, and at the very same time, there are other resources in the network that are operating at capacity, even perhaps dropping calls due to high user activity in such wireless cells. The concept of the Cloud RAN architecture is that then if these various base station processing resources are moved from the current remote field location at the base of the antenna, and replaced by a centralized or clustered set of processing resources services the multiple remote end points, that fewer resources are required to adequately service all wireless users, and that the operators will then save on CAPEX, OPEX and potentially even reduce the overall power footprint, offering “green advantages.” Furthermore, it is argued that this wireless network architecture can provide new noise interference and user scheduling techniques, improving the use of the costly and coveted spectrum.

Our conclusion, as was with others at the conference that this concept is feasible, with some challenges to overcome. Essentially the technology exists to support such architecture, however the solution will most likely be a combination of this and the evolving heterogeneous wireless network topology. The cloud topology does pose some challenges with significant requirements for fiber backhaul, rendering some regions untenable. In addition, if the cloud base station is not properly architected and outfitted with performance and power efficient multicore processing elements, the power reduction benefits may not be achieved and may even require more overall power than the current topology.

One attendee commented that he recalled such a wireless network topology being batted around the market about ten years ago, and asked, why then did it not receive traction then, why is it being revisited now?

See original post:

Share this Story

You may login with either your assigned username or your e-mail address.
The password field is case sensitive.