Now that we’re seeing major DOCSIS 3.0 deployments with downstream channel bonding, the question of upstream bonding is coming to the fore. Like its downstream counterpart, however, upstream channel bonding only addresses peak throughput, not total bandwidth available. For example, if someone is uploading a large video file to YouTube, channel bonding can increase the throughput for that user’s modem. However, that does not mean that there is more bandwidth available overall for an entire service group needing more upstream capacity. As more users need more upstream bandwidth, operators need solutions for both peak and total throughput.
On the downstream side, there are numerous options for cable operators looking to increase bandwidth in conjunction with applying downstream channel bonding: analog channel reclamation, switched digital video, and MPEG-4 compression, to name a few. The same cannot be said on the upstream side of the equation. Even as operators are starting to explore upstream channel bonding, very little is being said about how to increase overall upstream capacity. And guess what? The limited data we have suggests that average upstream usage by consumers is steadily increasing.
So what are the options? At the moment, the best option for cable operators is to use something that is, ironically, part of the DOCSIS 2.0 standard. S-CDMA (synchronous code-division multiple access) technology cleans up noise at the low end of the spectrum to increase upstream capacity by roughly 50%. This is not a replacement for DOCSIS 3.0. It is a very important complement. Upstream channel bonding? Good stuff. Upstream channel bonding with S-CDMA? Great stuff.