[RE] Requirement: Bridging to 1394
One requirement that also needs discussion and consensus is this:
---
Residential Ethernet must provide a backbone for IEEE 1394 clusters that
does not introduce significant loss of functionality. Indeed, it would be
best if Residential Ethernet was *higher* in "QoE" than 1394.
---
The implications are the following:
1) Sub-millisecond latency for isochronous packet delivery within a one-hop
network. Single digit millisecond latency for multihop networks.
2) Accurate clock reference for time stamping of IEC61883 packets (1394 uses
a well-known and synchronized 24.576MHz clock).
3) Management of bandwidth resources such that existing connections should
be lost or packets dropped in the normal case (or in normal "exceptional"
cases such as the addition or removal of a non-participating device
somewhere else in the network).
4) Transport of IEC61883 packets with minimal (preferably NO) changes.
5) Transport of 1394 uncompressed video packets with minimal (preferably NO)
changes.
6) Simple tunneling of 1394 asynchronous transactions.
Note that 4, 5, and 6 above can be done using the same techniques used in
the 802.11 and 802.15.3 "Protocol Adaption Layers" for 1394 ... Providing
that the basic isochronous transport and clock synchronization mechanisms
are in place.
--
-----------------------------------------------------------
Michael D. Johas Teener - Mike@Teener.com PGP ID 0x3179D202
23 Acacia Way, Santa Cruz, CA 95062-1313
+1-831-247-9666, fax +1-831-480-5845
------------------- www.teener.com ------------------------