If Femtocells were to operate on a frequency that is shared with the macro network in a restricted mode (i.e. only specific users can access a particular Femtocell), there is a concern that interference between the two layers may create a doomsday scenario.
For some reason, many think that the uplink interference from Femto users will jam the macro recievers and cause macro uplink coverage to deterioate. Some technologists and vendors have responded to this challange by developing extremely complex techniques. I have seen documents on both FemtoForum and 3GPP talking about things like:
- an interface between the macro layer and femtocells to exchange radio specific contexts.
- A "cognitive pilot channel" proposal that seems to be at least 15 years ahead of its time.
- Proposals to use uplink interference cancellation technology in femtocells.
- Suggestions that only LTE access points and release 8 (or 9) handsets will eliminate the interference problem.
All of the above is very interesting for a technologist. Ofcourse there is always a way to make things better, but there are also other considerations: value added, cost, market segment, technology application, market drivers ...etc, otherwise we would be all driving a Ferrari or a Porsche.
Looking at this interference issue, I can't help point out a number of things:
- I believe that more than half of femtocell deployments will end up using a dedicated carrier.
- In the absence of a femtocell, the user will go on the macro network and will create an amount of noise rise anyway. When you put a femtocell close-by the user will radiate less power and create less noise rise to the macro site. There is at least a 25-30dB reduction in path loss, so the femto users will radiate much less power than if they are using the nearest macro cell.
- A residential femtocell will only support a very limited number of users, so the interfernce limit ishardly ever reached and additional uplink interference from macro users operating in the vicinity will hardly block the femtocell uplink.
Is is possible that theso called interference problem is nothing but a big red herring?