Quote:
Originally Posted by Sephiroth
But if they pinched it from another set, the misery continues. I'd say they were more likely to upgrade their optical systems to uniformly use WDM so that at least 16 channels could go down one fibre. (This is just the downstream we're talking about - corresponding upgrades for the upstream would be necessary, a different topic). Or they would lay more fibres.
Igni may well get stuck in to us both!
|
The fibre isn't a problem. The issues preventing additional downstream channels are either side. The capacity of line cards / Edge QAMs at the VM side and the RF capacity of node and amplifiers towards the customer.
WDM allows splitting of nodes without additional fibre, multiplexing nodes onto the same fibre pair, not delivery of more bandwidth to each node.
---------- Post added at 14:37 ---------- Previous post was at 14:30 ----------
Quote:
Originally Posted by vm_tech
I both agree and disagree with you. The bottle neck is the CMTS port, but you are correct in saying that the number of homes connected to the node affects this. The network I work on has a few different designs, the earlier stuff was 2100 home build, which a few years back had new fibres to break it down to 525 build. The rest of the stuff was 525 build originally. DWM has recently been used to push the node deeper into the network in my area (replacing an RF amp). So effectively the main node has another node running off of it, which has its own CMTS port
|
So this is resegmentation, however rather than running new fibre all the way from hubsite or headend to each new node the existing fibre goes to a WDM mux colocated with the previous node, and new fibre is run from the original node only.
Removes the need for new long fibre runs; new fibre is only installed on that last few hundred feet in the field and however long in the hubsite/headend between the WDM mux and the Edge QAMs / digital return modulators / media converters / whatever.
Doesn't change how many channels each node can use unless the node's capabilities were the only bottleneck, but reduces homes passed per node.
To allow more downstream and, indeed, upstream channels to be used is why lucky people like the good tech here are supervising replacement of 750MHz / 860MHz total capacity plant, upstream going no higher than 5-50 or 5-65MHz, with 1.2GHz plant split at 5-85MHz up, 108MHz-1.218GHz down, with diplexers field replaceable to move the split to 5-204MHz up, 258MHz-1.218GHz down.