Re: on the ieprep charter

Fred,

> In the general case, I would expect push-to-talk to be handled by  
> provisioning. I'm not certain that the capability is of an  
> "emergency" nature. In civilian networks, it is commonly used on  
> virtual trading floors and as a walkie-talkie replacement. I can't  
> say I know for certain how it is used in military networks (they  
> haven't told me), but I would expect that it permits communication  
> among a physically distributed group of people that operate as a  
> team, such as command/control operations. In civilian networks, the  
> way it generally works is by ensuring (either by provisioning of  
> bandwidth or by placing traffic into an appropriate class) that  
> there is bandwidth available to the application at the time it  
> needs it, and is generally a low capacity channel.

a good read on things, though the headache arises in the details.   
Today, provisioning in LMR systems used by first responders (police,  
firemen, EMS, etc.) is accomplished by man-in-the-loop dispatchers  
that assign channels per talkgroup (analogous to per multicast  
group).  and what is known as "radio discipline" (by end users) is  
primarily used to control contention on the channel.  failing that,  
the dispatcher manages the group and can assign more talkgroups/ 
channels, though these are limited.

and as Janet mentioned, prioritization in LMR (layer 2) packets has  
recently been added tot he mix.  Project 16a of APCO defined a set of  
5 priorities, and Project 25 defines 7 levels.  one moral in that  
story is not to be fixated on the number of levels....they may change.

my understanding is that the number of towers involved in a talkgroup  
generally range from 0 (direct communication between end users) to 4,  
with a gausian decay after that.  today, these towers don't have IP  
embedded within them, but that will come in time.  outside of this,  
there are also remote observer participants that receive the  
communications from remote geographic regions.

-ken