Order Tray | Contact Us | Home | SIG Lists

[linux] Callsign coding (was Take a Stand)

Gerry Creager n5jxs gerry.creager at tamu.edu
Wed Oct 13 17:33:06 UTC 2004



gerry

Martin Ewing wrote:
> This might be good if the only consideration is to
> save address space (bits) for normal callsigns.  But
> 36 symbols are not enough. People use "/" and "-" and
> maybe others.
> 
> The trouble is that a "callsign" has no well defined
> syntax.  We find things like FG7/WA1ABC and even
> worse.  I suppose the roman (english) alphabet is
> standard, but which special characters and what
> maximum length to allow are not clear to me.  Is there
> an ITU standard? Probably not, except for prefixes.

The ITU/IARU standard is, in fact a non-standard, but for most cases, 7 
characters is sufficient to identify any station.  If we see some sort 
of localization marking for network geographic identification, there's 
then some implied localization of origin.  Not, however, that I'm in 
favor of trying to geo-reference IPv6 network addresses, really.

> For very best address efficiency, you'd want a unique
> binary identifier, and 24 to 26 bits should be enough
> for all the hams in the solar system.  But you need a
> registry database (like DNS) somewhere to convert to
> human readable callsigns.

Not really: With a standard encoding mechanism an application can 
readily read the callsign for you.  Personal preference: Let's go to 
7-bit ascii and encode all calls "normally".  There's sufficient 
additional network space in the addresses to supercede the desire to use 
AX.25-style SSIDs.

> Given that IPv6 has lots of bits, maybe we don't need
> to be so careful.  Or, we could even use a variable
> length field, ASCII-coded. (Takes more work to parse,
> but CPUs are cheap.)
> 
> So nothing is easy!

However, getting us started down this path is useful.

Gerry n5jxs

> --- "Bob Hansen, N2GDE" <n2gde at tapr.org> wrote:
> 
> 
>>Think of the callsign as a base-36 number.
>>For 7 places of 36 symbols (0-9, A-Z) you have:
>>36 to the 7th power or about 78 billion
>>combinations.
>>To encode this in binary, you need at least 37 bits:
>>2 to the 37th power is about 137 billion.
>>This is where Darryl got 37 bits.
>>
>>Of course, as N4HY suggests, most of those
>>combinations are not valid for
>>callsigns and an elegant hashing method may be
>>possible.
>>
>>-Bob H.
>>N2GDE
>>
>>
>>_______________________________________________
>>linux mailing list
>>linux at lists.tapr.org
>>
> 
> https://lists.tapr.org/cgi-bin/mailman/listinfo/linux
> 
> 
> 
> _______________________________________________
> linux mailing list
> linux at lists.tapr.org
> https://lists.tapr.org/cgi-bin/mailman/listinfo/linux


-- 
Gerry Creager -- gerry.creager at tamu.edu
Network Engineering -- AATLT, Texas A&M University	
Cell: 979.229.5301 Office: 979.458.4020
FAX:  979.847.8578 Pager:  979.228.0173
Office: 903A Eller Bldg, TAMU, College Station, TX 77843




More information about the linux mailing list