Op 19 jan. 2012, om 15:25 heeft <[email protected]> het volgende geschreven:

> We have a related use case in the document:
> 
> 4.7.  Rapid deployed network for emergency scenario
> 
>   Organizations involved in handling emergency operations often have a
>   fully owned and controlled infrastructure, with dedicated spectrum,
>   for day to day operation.  However, lessons learned from recent
>   disasters show such infrastructures are often highly affected by the
>   disaster itself.  To set up a replacement quickly, there is a need
>   for fast reallocation of spectrum, where in certain cases spectrum
>   can be freed for disaster relief.  To utilize free or freed spectrum
>   quickly and reliable, automation of allocation, assignment and
>   configuration is needed.  A preferred option is make use of a robust
>   protocol, already adopted by radio manufacturers.  This approach does
>   in no way imply such organizations for disaster relief must compete
>   on spectrum allocation with other white spaces users, but they can.
>   A typical network topology would include wireless access links to the
>   public Internet or private network, wireless ad hoc network radios
>   working independent of a fixed infrastructure and satellite links for
>   backup where lack of coverage, overload or outage of wireless access
>   links occur.
> 
> So it seems we'll need a requirement to free spectrum on a short notice. From 
> the above use case my reading is that 'short' is close to instantaneous.

Related to timescales for current frequency allocation (in years), yes, it is 
instantaneous. I recommend we leave out-of-scope how & why spectrum is 
reallocated. A question is: if spectrum is freed, how to inform PAWS master 
nodes, to get it. In this use case, there is a dedicated channel and 
applications would make use of it to receive triggers. Triggers could be SNMP. 
Maybe we work on MIBs some day, and simply define a trigger. 

Teco

> 
> - Gabor
> 
> -----Original Message-----
> From: [email protected] [mailto:[email protected]] On Behalf Of ext 
> Rosen, Brian
> Sent: Thursday, January 19, 2012 6:09 AM
> To: Peter Stanforth
> Cc: [email protected]
> Subject: Re: [paws] next steps for the wg
> 
> At this stage, we should say that we have a requirement to change 
> availability of spectrum on a short notice, we need a use case that would 
> motivate some decision on what "short" means, and we should not worry about 
> which mechanism we should choose to achieve it.
> 
> Push, fast poll, and notice broadcast (i.e. a single bit sent to all clients 
> telling them to "phone home" soon) are all reasonable mechanisms to meet the 
> requirements.
> 
> Brian
> 
> On Jan 19, 2012, at 8:40 AM, Peter Stanforth wrote:
> 
>> I would like to suggest that we back up and ask the question. How do 
>> we pre-empt a device if the primary spectrum user needs it back? The 
>> use case I can think of is public safety, where they don't generally 
>> need the spectrum until there is an emergency. They do not know when 
>> or where that will happen so they would need a mechanism to clear the 
>> spectrum in a specific location quickly.
>> A push mechanisms only one way to do this. Another, off the top of my 
>> head is to only give channel permissions for time durations less than 
>> the time they need to clear the channels. There may be other options 
>> and there are clearly advantages and disadvantages to all.
>> Peter S.
>> 
>> On WedJan/18/12 Wed Jan 18, 11:38 AM, "[email protected]"
>> <[email protected]> wrote:
>> 
>>> 
>>> Hi Joel,
>>> 
>>> The proposal to include unsolicited Push notifications from the white 
>>> space database to a master device is different from the 
>>> Request/Response mechanism itself.
>>> A master device making a request for available channels expects a 
>>> response in some time window. Not proposing we change that.
>>> However the white space database knows of devices which have 
>>> registered with it. And hence can send push notifications at will 
>>> without necessarily having to react to a request.
>>> 
>>> -Raj
>>> 
>>> On 1/17/12 8:03 PM, "ext Joel M. Halpern" <[email protected]> wrote:
>>> 
>>>> While responses have time windows, as far as I know, requests do not 
>>>> specify when the response will be acted upon, if ever, or for how long.
>>>> 
>>>> As such, this seems to imply either that we add significantly more 
>>>> information to requests, or that any change in anything that has 
>>>> ever been asked for gets pushed?
>>>> That does not sound like a good design.
>>>> 
>>>> Yours,
>>>> Joel
>>>> 
>>>> On 1/17/2012 6:08 PM, [email protected] wrote:
>>>>> 
>>>>> Hi Gabor,
>>>>> 
>>>>> On 1/12/12 8:26 PM, "ext 
>>>>> [email protected]"<[email protected]>
>>>>> wrote:
>>>>> 
>>>>>> P.3 currently says:  The protocol between the master device and 
>>>>>> the WS Database  MUST support pushing updates in channel 
>>>>>> availability changes to subjects.
>>>>>> There were comments that this requirement involves a mechanism, we 
>>>>>> should reformulate to be mechanism agnostic.
>>>>>> There was a suggestion to "make the requirement "quick way to 
>>>>>> change availability" rather than imply a mechanism.".
>>>>>> The use case is that if the channel availability changes in the 
>>>>>> DB, the client has to be able to detect it and get the new 
>>>>>> availability list within a time period set by the regulator.
>>>>>> Can someone send suggested text on how to reformulate this 
>>>>>> requirement?
>>>>> 
>>>>> The requirement to enable Push notifications to be sent to a white 
>>>>> space device which has registered with a database is important 
>>>>> especially in the context of Ofcom requirements (I believe). The 
>>>>> reasons for such push notifications could be for purposes that go 
>>>>> beyond just channel availability updates. A proposal for the 
>>>>> requirement is as follows:
>>>>> 
>>>>> Requirement: A white space database should be able to send 
>>>>> unsolicited messages to a master device which has registered with 
>>>>> it. The protocol between the WS database and master device MUST 
>>>>> allow for push notifications to be sent from the database to the master 
>>>>> device.
>>>>> 
>>>>> -Raj
>>>>> 
>>>>> _______________________________________________
>>>>> paws mailing list
>>>>> [email protected]
>>>>> https://www.ietf.org/mailman/listinfo/paws
>>>>> 
>>> 
>>> _______________________________________________
>>> paws mailing list
>>> [email protected]
>>> https://www.ietf.org/mailman/listinfo/paws
>> 
>> _______________________________________________
>> paws mailing list
>> [email protected]
>> https://www.ietf.org/mailman/listinfo/paws
> 
> _______________________________________________
> paws mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/paws
> _______________________________________________
> paws mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/paws

_______________________________________________
paws mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/paws

Reply via email to