Re: [802.21] terminology clean up needed for link parameters and QoS parameters
Hello Qiaobing,
Qiaobing Xie wrote:
> Hi, Nada,
>
> ...
>> I agree. Some housekeeping is needed with respect to the parameters
>> that seem to be causing confusion.
>
> Great!
>
> ...
>>> 1) Network QoS attributes (as used in Network QoS IE) - they are
>>> *static* values assigned to a network (not a link) by its owner.
>> I agree that this a type of parameter, but I will observe here that
>> Network QOS IEs do not have to be restricted to static parameters,
>> although these are likely to be prevalent. If measurements are
>> available and can be used, I don't see the problem with supporting this.
>
> Right now, anything we put in the IS server is static. But it is
> entirely possible our IS service can be extended in the future to
> cover more dynamic information.
>
>
>> a) -
>> - define for each link specific parameter list, a subtype consisting
>> of: 1) network static/qos, 2) measurements, 3) state/configuration
>> - use subtype appropriately in IE, link primitives and MIH primitives
>>
>> OR
>>
>> b)-
>> - define a link specific list for each IE, link and MIH primitive.
>> This will consist in all link specific parameters related to that
>> primitive.
> Another approach could be that we support:
>
> 1) A generic set of Network QoS Attributes for MIH IS - I don't see a
> huge value for us supporting network specific QoS attributes.
>
> 2) For eash link type, a set of link specific measurements that we can
> set thresholds for - I don't see much value for supporting a set of
> generic link measurements (you have to have a real link that take the
> measurement before you can get it)
>
We do have MIH generic parameters related to QOS that we can pass to the
link layer. These can be considered link generic. The mapping from
generic to specific can either occur at the MIH or link layer. Although
today's links may not be able to understand the MIH generic QOS
parameters, the motivation is that future generation links or extensions
will include them. This is a compromise that we agreed to and hope we
don't have to undo it.
> 3) A generic set of link states that we can control (set and get) -
> many link states are generic already by their nature (OPERATION_MODE,
> BATTERY_LEVEL, etc.).
>
> regards,
> -Qiaobing
With the exception of item (2) where we'd like to keep both generic and
specific parameters, the
approach you have works as well.
-Nada
--
Nada Golmie, Ph.D.
Manager, High Speed Network Technologies Group
National Institute of Standards and Technology
100 Bureau Dr. Stop 8920
Gaithersburg, MD 20899
Email: nada@nist.gov
Phone: (301) 975-4190
Fax: (301) 590-0932
Web: http://w3.antd.nist.gov