Re: [802.3_NGEPON] 25G ONU Options to increase Network Capacity
Shawn, you say:
>More wavelengths and optics capability (i.e. wavelength-tunability) for 25G ONU may reduce CapEx and/or OpEx for operators to deploy additional wavelengths λ1, λ2, λ3. It may not.
The commercial availability of solutions implementing G.989.2 NG-PON2 may result lowering the costs of Tunable ONUs to a point fit for reducing the CapEx and/or OpEx of operators.
However, in your slides, you are calling for the following straw vote in Macau:
“ONUs with 25G capacity (25G ASIC) shall only be available on a single wavelength pair λ0”
This seems only to be aligned with the ‘it may not’ portion of your statement above.
A lot of functionality will be lost for 25G ONUs if the frequency plan for λ0 is not anticipated to be in a range compatible for tunability across λ0, λ1, λ2, & λ3.
The first one, is the lack of capability to move an offending ONU in the λ0 TDMA pool to another TDMA pool with more bandwidth, or worse, being unable to move all ONUs on λ0, to another channel if λ0 goes down. Those use cases were regarded as important in the process of standardizing on NG-PON2.
I think the frequency plan needs to be looked at more carefully before dismissing the benefits of tunability based only on the current costs of tunable receivers, as the situation is bound to change with NG-PON2 being deployed.
I believe it to be critical to ensure that λ0 is in a frequency plan that is compatible with the tuning range of 10 nanometers in the direction of the ONU to the OLT.
In the direction of the OLT to the ONU, it doesn’t matter as much if λ0', λ1', λ2', & λ3’ are in a contiguous tuning range of 10 nm or not.
I’m voting no ;-)
-=Francois=-
Francois Menard
CTO
AEPONYX inc.