Sorry about that. I didn't make myself clear. Instead of maintaining an analog-like channel plan, or frequency plan, I was suggesting a digital baseband format with discreet end point addressing. (Hello, IP v6, anyone?)
Think about a 4 Gb/s "switched," not shared, Ethernet coaxial link going back to the neighborhood node. Upstream from the node to the head end there would be a series of 10 Gb/s links now, or a 40 Gb/s or 100 Gb/s link down the road.
How many homes passed would this model support? 4 Gb/s implies the use of a QAM modulation technique, or some such.
Even without using a QAM modem, say, a digital line driver employing Manchester were used. 2 Gb/s could still be achieved for the worst-case distance homes that are currently being supported, where the 750 MHz or 1GHz worst case criterion has been met.
Mind you, this only applies to what I perceive as a potential means of salvaging the black coax in HFC. I wouldn't recommend it from scratch, although I think we're on to something here if we're interested in preserving the investments in embedded plant. The biggest problem with this schema could very easily be, how do you transition live customers onto the new platform in a timely and agreeable manner, since entire block sections would need to be "thrown" at once? |
