Re: [16NG] Re: traffic classification

Riegel, Maximilian wrote:
> The host is not aware of the CIDs. CIDs belong to the MAC layer, and 
> it is the task of the convergence sublayer as upper part of the MAC 
> layer to feed packets into CIDs according to filter rules. IEEE802.16
> provides its own syntax for setting up filter rules. This syntax
> treats all header field the same way.

Ok, I understand the principle.  There are MAC filtering rules that
match an IP packet to a CID.  My question was about a specific value in
Traffic Class field.

> IEEE802.16 is done,

Ah?  This is something I'm not sure of.  For example, a new corrigendum
IEEE Std 802.16, Corrigendum 2 [Draft 2]) has been circulated among
802.16 members just a few days ago.

(In my reading, there are some modifications that are relevant to the
  discussion around entry procedure, for example it now says that only
  for managed SSs the IP connectivity shall be performed over Secondary
  Management Connection; previously (802.16e-2005) it said that for any
  SS, not necessarily managed.)

> but it may be worth to consider in the IETF the definition of a 
> generic way to specify filter rules for IP packets. The WiMAX Forum 
> NWG has adopted the filter rule syntax of DIAMETER. So far it looks 
> pretty sufficient.

I think some monami6 WG drafts look at filtering rules, although not
specific to particular link-layer (they're for filtering at HA and
mobility).

Alex

_______________________________________________
16NG mailing list
16NG@ietf.org
https://www1.ietf.org/mailman/listinfo/16ng