Has anyone successfully gotten this to work?
"Asking for a friend.... "
If so, what field components are you utilizing?
What CMTS settings?
Only thing RFOG specific on the casa I can see is rfog scheduler in the service group.
At this point in time, the 85mhz return with OFDMA has been replaced with 12 6.4 ATDMA upstreams and only allowing 8 channel upstream bonding modems out the door for those nodes.
For my understanding OFDMA is not compatible with RFoG as long the burst mode on the RFoG nodes are used. The rfog-scheduling will not really help because you will lost a lot of capacity. The rfog-scheduling will make sure, that only one CM can send at a time. So, in worst case, if you have a D3.1 CM with OFDMA upstream and the CM will only transmit low throughput, only one minislot of 0.4 MHz might be used. If you have a OFDMA channel with 40 MHz channel bandwidth you'll waste 39.6 MHz at the time the particular CM will only use 0.4 MHz during the timeslot. the available 39.6 MHz can't be used by any other CM because of the enabled rfog-scheduling. Very inefficient.
If your RFoG setup is using Multi-Diode-Receivers (MDR) and you're able to activate the CW mode on the RFoG nodes, then OFDMA can be used again.
A small backround on this. The typical RFoG node with SC-QAM/A-TDMA based upstream channels are using the practical circumstances: There is a certain ramp-up time until enough energy of the burst signal will trigger the transmit mechanism of the RFoG node (if Burst-Node is enabled). Then we have the relativ long preamble pattern in the beginning of the upstream burst in time domain. I mentioned the energy of the single upstream channel, which is e.g. 3.2 or 6.4 MHz. All this can work good together with the RFoG nodes in burst mode.
Now with OFDMA we have some problems: The ofdma burst, depending e.g. on the modulation (spectral efficiency) is very fast in compare to SC-QAM ==> very short ramp-up time. The OFDMA upstream burst doesn't have this preamble like A-TDMA/SC-QAM. Then the OFDMA channel is also devided in minislots at the frequency layer. One OFDMA minislot is 0.4 MHz wide and if the D3.1 CM has not much to transmit the 0.4 MHz minislot is all what the CM will transmit. In compare between SC-QAM/A-TDMA which is always using the entire channel bandwidth with e.g. 6.4 MHz and OFDMA we have a minimum of 0.4 MHz we have in a worst case scenario a difference of 12 dB which might be much to low to trigger the RFoG transmit mechanism (burst mode) in the same way like with SC-QAM/A-TDMA with 6.4 MHz. Maybe you can play around with the Guard-Intervall/Cyclix-Prefix but this will not really help.
Is a multi-diode receiver (MDR) an active optical splitter?
Do you have experience using them? I'm planning to switch to RFoG, but I can't find suitable equipment. I would like to be able to remotely disable a single port or change the attenuator.
You must have posted this seconds after I last checked and figured nobody knew. :D
So to make sure I understand correctly, rfog scheduler isn't advanced enough to handle OFDMA, allowing only one modem to talk at a time, basically. This is why nobody has managed to make it work?
Is there any research into an OFDMA scheduler anywhere?
Help me find the arris agilemax manual
thanks in advance
serykh (dot) mobile (etc) gmail (dot) com