"Along the same line of thought, I'm beginning to sense that if wireless is to deliver equal (or near equal) capabilities to those of wireline offerings, especially in areas of relatively high penetration, then it may only be possible through the use of fiber to microcells, or to pico and then possibly to femto cells, allowing for maximum re-use of frequencies. All of which means smaller footprints per access point and a heckuva lot more of them."
Wasn't that always the concept? Increasing infill?
In 1999/2000, Wi-LAN proposed a network called 4GNT. It came to nothing, but the concept was very much like what you describe - i.e., frequency reuse, with progressively smaller footprints, from tower to picocell.
With enough infill, you can jack up RF throughput to equal some wired equivalents.
The problem that we saw (among others, at the time) was cost: when does the infill cost for an RF system approach the cost and difficulty of simply wiring the whole thing? When do you start to lose one of wireless's main advantages? When does it start to be mostly a network of cells, picocells and femtocells linked by fiber?
To Peter's discussion about QOS and ethernet-like best-efforts, circuit-switched vs. packet-switched thinking, his comment about a marginal increase in QOS with WiMax up to a quickly vanishing point is correct; but that brings us back to the question of the underlying spectrum, and how much traffic it can be made to carry, regardless of how data is switched.
Also, can future iterations of the WiMax standard address the weaknesses Peter has noted, just as WiFi was incrementally improved?
It appears that some mobile operators and some carriers are lining up for WiMax; the guess is that in future they intend (or hope) to grab complementary spectrum.
If this is a discussion about WiFi vs WiMax, then I still believe that before a clear winner will emerge, the rules have to change, and that change must be at spectrum.
If it's a discussion about any possible wireless technology (UWB, HSPA, LMDS, WiFi, WiMax, MMDS, LTE, etc) then we know we're going to see a patchwork quilt. Governments, OEMs, standards bodies, regulators, and the world + dog will have a say in changes, which will come slow.
Meanwhile, operators (and their circuit-switched thinking) will work behind the scenes to protect their turf, and WiMax is (apparently) their RF choice, beyond their commitments to LTE, HSPA, EVDO, and others.
Report of the ICN Working Group on Telecommunications Services (.pdf)
tinyurl.com
Have I understood the discussion correctly? There are so may factors that bear on global RF trends.
Not that it matters, but I believe Peter's right. His thinking will probably factor in whatever "4G" turns out to be, maybe 20 years from now.
Jim
|
