[aprssig] NSR Algorithm Modification Proposal
AE5PL Lists
HamLists at ametx.com
Thu Jun 23 06:44:19 EDT 2005
As stated in the papers, there will always be room for improvement in
the algorithm. I can see two areas so far that IMO should be modified:
1. It was accidentally omitted from the algorithm that the SSID of the
destination field should _not_ be used in the dupe check algorithm.
This is due to some attempts at using the SSID for routing purposes.
2. I recommend that Item #3 should read:
The digipeater repeats any packets it sees digipeated by digipeaters on
its "ok" list _unless the packet has been digipeated by a digipeater on
its exclude list_.
The area between underscores is the portion added to the original text.
What this addition allows is the ability to have overlapping LAN's where
one or more digipeater areas are included in multiple LAN's. For
instance, lets take a 1 dimensional example: Town A, Town B, and Town C
all have digipeaters and Town B is in-between Town A and Town C. Both
Town A and Town C consider portions or all of Town B within their local
area and vice versa. But Town A does not consider Town C to be local
and vice versa. The above addition would allow the digipeaters in Town
A and Town C to digipeat everything from the Town B digipeater and vice
versa. However, a packet digipeated by the Town C digipeater which is
also digipeated by the Town B digipeater will not be digipeated by the
Town A digipeater.
user->A->B
user->C->B
user->B->A&C
Neither of these items break the NSR nor do the obsolete the previous
version. They simply provide the digipeater sysop more flexibility in
configuration and the digipeater author more accurate dupe checking.
73,
Pete Loveall AE5PL
mailto:pete at ae5pl.net
More information about the aprssig
mailing list