[Let me know whether my brand new PGP signature disturbs]
==> it's worth mentioning the unstated assumption here is that wireless links being considered here have very long roundtrips (e.g.,
0.2-1.0 seconds)
Some measurements I've done, if worth considering: the UMTS/GPRS wireless links that I consider have between 0.02s (that is about 20ms measured) and 7s roundtrip, from the MN to the first IP hop. I have no idea whether this link is exclusively wireless or includes some wired segments, but measurements are between two IP points without a router in between, and at least one segment is wireless.
Also those links are asymmetric and I believe the downstream is faster than the upstream but I don't know by what percentage (i.e. an Echo Request - Echo Reply roundtrip takes 20ms but I can't swear that the Echo Request was 8ms or 5ms, it's just that it wasn't 10ms).
Their MTU is also asymmetric, in that the MN's IPv6 stack does fragment the upstream but does not defragment the downstream.
I have some measurements available in a deliverable somewhere...
Attachment:
signature.asc
Description: OpenPGP digital signature