RE: Re: IETF IPCDN Signaling MIB - Draft 3 - Last Call for Comments

> Issue #7 - definition of pktcSigDevToneDbLevel
...
> In the current draft-03, this object is specified as follows: 
>  
> pktcSigDevToneDbLevel    OBJECT-TYPE 
>     SYNTAX       Integer32 (-60..4) 
>     UNITS        "dbm" 
>     MAX-ACCESS   read-write 

> (1) The analog reference point seems the logical reference point to use,
>    since this table is about signals that will be generated on the
>    analog line. When the reference point is changed to the digital
>    reference point, will this not be confusing for someone who wants to
>    fill out the values in the table? Maybe that person will have to do
>    a quick calculation to calculate what digital dB level he must
>    provide to obtain the analog dB level he wants, but if this is the
>    case, can the MTA not do this calculation?

This calculation would only need to be done once by the operator so that
the correct defaults could be determined. After that, these defaults would be 
used in the configuration of all MTAs. While we understand the concern, we don't
think that is too big of a hurdle.

>    Second, just as a tought, would the problem also be solved by
>    leaving the description of the object as it is, but changing the
>    range limits?

This would also resolve the issue, however, it might be difficult to
specify (in the MIB) the limited range since it will be dependent on the digital
loss plan deployed in the country. For example, a country that uses -10dB D/A loss
would need to have the tone range restricted to -10...-60. Meanwhile, a country with
very little digital loss (i.e. -3dB) could afford to have a larger range (-3...-60).
MIB definition cannot accomodote multiple ranges for the same object.

> (2) Does this discussion also apply to the pktcSigPulseSignalDbLevel
>    object? In other words, I think the interpretation chosen for the
>    tone dB level should also be the one for the pulse signal dB level,
>    and clarified in the defintion of the latter as well.

Yes, agreed. When the issue with "pktcSigDevToneDbLevel" MIB Object is agreed upon, 
the "pktcSigPulseSignalDbLevel" object should be treated the same way.
Independent of this, it might worth noting that the current range of the 
"pktcSigPulseSignalDbLevel" MIB Object (-250..152) is very large considering only 
one of the modes (meterPulse) uses it and the country specifications do not require 
it to be so high. We would recommend reducing this to a more reasonable range 
of (-250..0).

>    Also, is it intentional that the SYNTAX of the
>    pktcSigPulseSignalDbLevel object is TenthdBm, and that it is
>    Integer32 for pktcSigDevToneDbLevel?

We agree. By using "TenthdBm" SYNTAX in all "dB" related MIB objects we would make 
the SIG MIB more consistent.

Eugene.