Some reflections from today’s Globecom 2012 “Industry forum” on Green Communication and Computing: Today Chih-Lin I, Chief Scientist at China Mobile repeated what I have seen in quite a few presentations: How the worlds (i.e. the wireless operators) problems can be solved by a) GHz of more licensed spectrum b) 8×8 Mimo c) centralized processing (“Cloud RAN) and d) some “small cells” (=lamp-post microcells) here and there. No one seems to note that a doubling of the capacity will buy them only one more year. Something more radical has to be done to meet the challenge of exponential traffic growth. The obvious “HetNet” concept, installing basestations more densely where capacity is needed is widely accepted, however the followup conclusion, that since maybe 80% of the traffic emanates from indoor these “small cells” should be indoor is carefully avoided. The fact that energy can be saved by not “blasting through” the walls from outdoor and that backhaul indoor may be almost for free (if standard IP access is sufficient) doesn’t come to the surface either.
The reason for this is not really technical, but a business problem. No one would come up with the idea of lighting the interior of my house with a huge floodlight in the street. But if my business is to build floodlights, that’s what I will do. The indoor (“offloading“) networks (the small “indoor lamps”) are not naturally owned by the public wireless operators .. they could be operated by “white label” operators or, god forbid, by premises owners providing access for free using unlicensed spectrum (i.e. not under the control of the incumbent wireless operators)! How can you make money in this case? A discussion is now rampant on who should pay whom for using the offload network. I think, in the days of the “capacity crunch”, public operators should be happy that there is someone that will offload their have traffic and keep their customers happy for free!
To the manufacturers high density, low-cost solutions seem like “herbal medicine” to the pharmaceutical industry – it may eventually work fine, but its cheap and it cannot be patented – so where is the incentive to invest money in developing it further? This is indeed a problem, since further developments of the indoor concepts ARE needed .. WiFi as is, is NOT the solution for ultra-dense networks and in addition we need seamless handover and SON-deployment features to keep the deployment cost low. Maybe providing more unlicensed spectrum for short-range indoor high-speed access is one way to incentivate this development.