As some of you may know, I've worked with LEOsats for email
I am now investigating wireless mail and instant message networks in
regions without power or telecommunications lines or availability of mobile
What I would really like now is any comparative analysis of the different
technologies available and their actual costs when implemented in remote
regions, and any comments on their reliability based on experience.
Ideally this would be freely available from a development organisation, but
I would also like to know of any reports from commercial research analysts.
The data volumes are low - even old fashioned amateur packet radio would
probably do for the 9600 - 56000 bps range of speeds. I think we'll need
always-on (which is why I am looking outside my pacsat experience), but
latency of a few seconds should not be a problem.
Distances are 1-10 Km. I *think* line-of-sight may be always available, but
that may be simply imposed by the capability of available solutions.
Power consumption has to be low enough to be supportable by renewable
power; this obviously has cost implications in that solar panels may be
more expensive than the comms equipment!
I have no bias or preference for OS - DOS, Windows, Linux, or unusual ones
like Psion, etc. I am expecting most real-life case studies to be networks
of remote PCs or laptops or handhelds networked to a central server, but
our interest is in message alerts rather than bandwidth-heavy software
applications. For that reason, I would also like to hear from anyone who
has used wireless for remote data collection, eg from water meters, weather
observatories, or similar.
Cost is a parameter rather than a cut off point. Having said that, I think
we will probably exclude the most expensive broadband satellite solutions,
but circumstances alter cases and some locations may have the budget for that.
I'd like to hear from others who have implemented working networks in this
area. I know of some from previous contacts, but I'd like to put out this
call to refresh my awareness.
Patrick O'Beirne, Systems Modelling Ltd.