“There is a perception that the only way broadband cable can compete with the likes of CenturyLink, Google Fiber, and others, is to adopt a fiber-to-the-home (FTTH) network architecture”

 Should operators deploy FTTH (fiber-to-the-home), DOCSIS (Data Over Cable Service Interface Specification) 3.1 or is there another answer? Very hot discussions about this topic in linkedin SNS website. FOT blog now sharing those very professional ideas and discussions as following:

DOCSIS 3.1 Cable Gigabit Broadband solution

DOCSIS 3.1 Cable Gigabit Broadband solution

 Zdenek: Agree with you, if MSO build HFC system RF tight there is lot of possibilities as DOCSIS 3.1 is. In combination with RFoG is it ideal way.

 Gordon: Interesting perspective but I’m wondering what the long term costs will be associated with the following embedded areas of HFC:

  • Continued degradation of the coax/copper facility
  • Continued powering and cost requirements
  • The necessity to sweep the 3.1 to higher standards and to maintain that level of performance on old plant
  • The continuous problems with lightning and fault line power issues

 Additionally, I’m not an expert on HFC but with the advance of IoT (The Internet of Things) and the on-slaught of the upstream bandwidth will this HFC network be able to handle the data loads of the future? I do think there is room for an RFOG/GPON solution that will allow migration to an all fiber network.
 This will be a great topic to discuss.

 Mark: This (broadband cable) is just delaying the inevitable (FTTH development).

 Miguel: you can squeeze the coax and increase bandwidth but your OPEX to clean and avoid ingress noise will also increase. I can understand that in brown field it is difficult to skip temptation to upgrade to docsis 3.1, but in green field. …no doubts! HFC will never be able to reach services that fiver does.

 Ian: DOCSIS 3,1 is great but P2P FTTH is better, in the short run repairing an old car is cheaper, but if you want to have it running for the next 20 years then a new one is better. So if you need Internet for the next few years then opt for DOCSIS 3,1 , if you want it for al longer time then look at fiber

 Thomas: Agree, while Docsis 3.1 has merits for existing infrastructure, it is merely a life support for said infrastructure. For new builds or upgrades to older Docsis installations, fiber, especially PON which mirrors HFC deployments (topology wise) is a more future proof and less OPEX intensive solution. G.PON and N.GPON 2 offer OSS supporting DOCSIS provisioning (Open Link Cable) and vendors like Calix offer sealed OLTs, powered via Coax for distributed outdoor deployments allowing replacements of micro nodes.

 Nick: Interesting perspective and one we debate internally as our company started from a Cable TV background, and now have customers on both sides of the house. One item that continually comes up in the discussion is the restoration aspect of these networks. When you have an event that cuts a cable in a HFC environment (backhoe, tree falling into lines, etc) the mean time to repair and get the network back up is as simple as either running a temporary, or placing a new cable and splicing new connectors, which takes a matter of minutes. In the same scenario in a FTTH environment, you are looking at hours to prep new splice enclosures and fiber, then a couple of minutes a burn for each fiber burn, and then end to end testing (this also assumes the restoration event doesn’t affect an existing splice enclosure, in which case how long would it take for you to get the fiber matrix records onsite?).

Tips: DOCSIS 3.1 is designed to allow cable modems to hit performance targets that were previously only possible with fiber networks through the use of 4096 QAM (quadrature amplitude modulation) and other steps that improve signal efficiency and transmission quality.

 Ervin: I think that this is mainly related also with Broadband strategy for each country. Lately I have been reading a lot of these strategies (most of them are most developed countries) where is simply written that by the end of 2020 – 50 or 75% of the houses or apartment are going to have 50 or 100mbps connectivity. Here it starts the discussion of the technology that can offer this such speed. Are we going to use the HFC or the FTTH? Both technology can deliver the above speed easily. On internet services both technologies at customer side look the same modem/onu meanwhile the HFC setbox with DVBC or hybrid one look more friendly. This advantage that the HFC has on customer side with DVBC channels, has its operational costs that for me in long term are a bit higher that a FTTH network. Yeah its true that you fix very fast a coax cable cut compere to a fiber cut but this is the worst that can happen on a FTTH network.

 Beyond the equipment cost, power consumption, coax degradation etc you need to have very skilled technician to keep to a certain parameters the HFC network in order to apply the latest technology like docsis 3.1. And these are getting older day by day. The FTTH network with all do respect (I am referring from the OLT-ONU part of field network) can be built or maintain by dummies, just explain the to keep in mind transmitting on +5 and signal it can be received on -25dBm. Try not to go under -28 or -29dB and it is done. Just seeing these data of fiber link a cable guy might struggle it self cause for a proper function you can’t go les than -6 or-7dBm where optimal is -2dBm for a fiber node. Than you have to tune the node for both side up an down. Than the amplifier and so on. In the end you have to terminate every port of your network. After all this is done all the area is ready for service. After some time that you have a customer that some how has damaged the RG6 cable.

 This customer is sending you back noises. Your area SNR has dropped below 28dB and you have changed the upstream qam from 64 to 16 in order to keep all modem connected or you start playing with the 3 state switch amplifier in case you do have a monitorise amplifier network in order to isolate the noise. And all this things need very skilled people. The Doscis (docsis 3.1 just has expanded for a better use the bounding channel, a modem can take 192Mhz bandwidth 24channel on eudocsis and 32 on docssis to be shared with 4096Qam on DS, up to 1.2GHz and on the US we have a bandwidth up to 200Mhz which from the node laser load can be very difficult to transmit fully load, DBF lase don’t know ) network is a real engineering network where knowledge is not enough if it is not mix with passion. It is very difficult now days to find these people. Everyone now is a blogger 🙂 and for this FTTH is the proper technology that has to be deployed on all cost. It is more human resource problem.