Re: [netext] Access Network Information option for Proxy Mobile IPv6

[Sri Gundavelli <[e-mail]>]

Hi Alex,

Thanks for the review. Please see inline.



On 7/14/11 1:42 PM, "Alexandru Petrescu" <[e-mail]>
wrote:

> Le 14/07/2011 22:10, Sri Gundavelli a écrit :
>> Hi Ryuji:
>> 
>> 
>> 
>> On 7/14/11 12:08 PM, "Ryuji Wakikawa"<[e-mail]>
>> wrote:
>> 
>>> Hi Sri,
>>> 
>>> 
>>> On 2011/07/13, at 23:37, Sri Gundavelli wrote:
>>> 
>>>> Hi Ryuji,
>>>> 
>>>> Thanks for the quick review. Appreciate it.
>>>> 
>>>> Yes, static keying of the information is always an option. But,
>>>> that will be a provisioning nightmare and secondly the
>>>> information may be multi-valued for a given MAG ID and may not be
>>>> 1:1. We fundamentally need a container for carrying access
>>>> network information, that may be specific to the SSID,
>>> 
>>> ok
>>> 
>>>> Operator Id ... of the AP ..., having this generic container will
>>>> greatly help deployments. In one of our deployments, we have a
>>>> requirement to carry the Geo-location of one of the AP's out of
>>>> the several attached to the MAG. Hence, we need this simple
>>>> option.
>>> 
>>> For geo-location, do you assume lat-long? Don't you need to specify
>>> the format of  information?
>>> 
>> 
>> Yes, we need to carry that information element. We need to define a
>> sub-type for this.
> 
> Defining geo-localization in a more universal manner (IETF) may or may
> not be straightforward.  It is true that "GPS" is in widespread use in
> certain regions, and simple lat-long sound acceptable by most.  But.
> 
> Do other IETF documents define such a representation for lat-long?
> Maybe that could be re-used here.
> 

This is a good question. I assumed the SIP guys must have defined the
semantics for the Geo-location info, as they carry in the SIP headers. I
will check with the SIP guys.




> Does one consider altitude as well?  This may prove useful when two APs
> of same MAG serve clients at different height.
> 

I do not know the semantics of a GPS information element. I assume that is
very standard and I'm not sure what all is included in that. Even in the
digital pictures, the camera embeds the Geo info in the meta-deta, we need
to see that meta data. But, there must surely be very standard option
format.



> Does one consider encoding the precision of measurement (GPS talks HDOP,
> PDOP, etc.)  This is important when making decisions about where objects
> really are.
> 
> Does one assume WGS-84?  Others exist.
> 
> Would this work with devices not equipped with GPS, but with inertial
> platforms (i.e. wrist watches).
> 

Ok. For defining the sub-type, we need to use the data format that has
already been defined. Will check with some experts.

Thanks for the review.

Regards
Sri