Frank:
...
...I take the distance claims with a grain of salt....
I've been involved with systems that boast (indeed they performed at) OC-3 155 Mbps over distances of 200 to 400 meters. 10 Mbps at a more lofty 1,000 meters, and so on. Have you come across any such sliding graph from LU in this context? Curious.
No I haven't, which gains makes me think I need to do a little research. I too take the distance calculations to be largely PR puff, but I'm willing to suspend my disbelief momentarily. I went over here and found some research paper abstracts that talk about the attenuation caused by atmospherics but shopping on-line via abstract is a little too speculative to me. <g>
spie.org
e.g.,
192.149.147.102
I had assumed that there would be a large body of empirical evidence and hence formulations for quantifying attenuation by precipitation, et al, similar to that available for RF systems. If those calculations are available, my next step would be to investigate how much gain one recaptures by throttling down.
In terms of practical applications, the ability to use IR for interconnecting hub sites in a broadband wireless network would be a tremendous boon. For example, even at a modest 1 bit per hz, 2 times reuse within the hub, an LMDS hubsite, in theory, could support a load of 2.3 gbits. With multiple hubsites, the bill to a local backbone provider adds up rapidly (unless of course one wants to trench their own fiber which ain't cheap either). I would want to capture the cost efficiencies of larger pipes, only paying for interconnect at the PSTN or national backbone level ( also not having to pay for transport of traffic which was internal to my network.)
Costs for hub site interconnect can run anywhere from 5-10% of revenues in a mature system, depending on how one mixes leased circuits with using one's owns spectrum for backhaul. Using ones own spectrum of course is not free, since you have the capital costs associated with the backhaul radios in addition to the fact that you're reducing the ceiling of deliverable capacity to end-users from that hub site: an interesting mathematical and financial modeling exercise.
I'm sure there are all kinds of other neat potential applications: its hot though and my crystal ball is a little hazy - pun intended.
Best wishes,
ww
p.s. I am in full agreement on issues of privacy, security and authentication. |
