Hi Will, The day may come when I need the precision and accuracy of rtk but today I mount sensor motes on vehicles, in & outside of buildings, on a fence on a beach in Florida - places where a location accuracy within 5 meters seems adequate. I usually have open sky and the Garmin seems to hold its own for now but I see your point. The sensor motes are small, 25 x 50 mm with a AA battery but not *that* small.
They operate in a mesh (ZigBee) with base station or star configuration (WiFi) with off-the-shelf access point(s). On board sensors include temperature, humidity, light (photo-detector); other analog inputs are used for corrosion, wetness, pH, potential for oxygen and aerosol particulate. Motes are connected to remote web portals through local cell phone-connected gateways with but each has onboard memory and is capable of store and forward. Current deployments are mostly static although vehicles come and go on a regular basis. The Fuller projection is a convenient way for me to describe an icosahedral-vicenary (base 20) framework but octahedral-octal and tetrahedral-quaternary work too. Octahedral-octal is especially helpful for those with a traditional Cartesian orientation. - Brian Will King wrote: > Brian > > I'd approach this from the bottom up. > > You asked about gps accuracy and - I don't think your garmin would cut > it. Personally I would use the most accurate gps you can get which is > rtk gps using a post processed base station (the data is processed > using your data plus data collected at fixed reference stations > throughout the local region). I'd also survey your fixed sensors > using a local projection for greatest field accuracy, converting them > to lat/long or fuller later. > > Please keep the comments coming because I'm fascinated as to why you > are using this particular method of storing data. > > > > On 7/5/08, Brian Grant <[EMAIL PROTECTED]> wrote: >> Hi Michael - I'm not sure I'm distinguishing theory from practice (or >> what's practical) - maybe this thought exercise is between the two >> overlapping spheres. >> >> What's pragmatic covers a lot of stuff; my use here is utilitarian and >> not necessarily deciding about what's "real" or "true" - I just need a >> tool for structuring data >> and points and polygons are not practical for me. >> >> I need a simple framework to associate geo-spatial location with >> measured attributes. Right now that framework is just a theory with a >> pragmatic need to test it among many minds who think in geo-spatial terms >> >> and aren't afraid to write them down and share it with each other. >> >> >> - Brian >> >> >> michael gould wrote: >>> Brian: I wonder what is pragmatic (or do you mean practical?) about the >>> below exercise. Pragmatics is another thing...proposed by a practicing >>> geodesist (CS Peirce) by the way...aimed at truth testing. >>> >>> And then I agree with Paul (see message 2 below), that categorization is, >>> uh, rather important to knowledge. I recommend George Lakoff's "Women, >>> fire >>> and dangerous things" on the subject (if bored, skip to chapter 16). By >>> the >>> ways categories do not need to be fixed, but can also be graded. The Pope >>> may not be a bachelor if a fixed definition is used, however his >>> bachelorness can be quite high all the same :-) >>> >>> -- >>> >>> Message: 1 >>> Date: Fri, 04 Jul 2008 10:17:23 -0400 >>> From: Brian Grant <[EMAIL PROTECTED]> >>> Subject: [Geowanking] pragmatic exercise >>> To: [email protected] >>> Message-ID: <[EMAIL PROTECTED]> >>> Content-Type: text/plain; charset=ISO-8859-1; format=flowed >>> >>> as a pragmatic exercise allow me to start with a map and end with a >>> database >>> schema. >>> >>> begin with a Fuller projection >>> (http://en.wikipedia.org/wiki/Fuller_projection) an icosahedral framework >>> of >>> 20 triangular areas (actually tetrahedra to the center - but we'll leave >>> that out for now). Each triangular area is designated a Major Triad. >>> >>> Each Major Triad is subdivided ** by the same base as the original sphere >>> ** >>> into 400 Minor Triads (20^2). If the edge of a Major Triad is 8,000 km the >>> edges of the Minor Triads are 400 km. Major Triads are designated with >>> characters A through T; Minor Triads are designated AA through TT. >>> >>> Minor Triads are further subdivided into Trixels (or whatever) again by >>> the >>> same base creating a recursive triangular mesh capable of defining unique >>> triangular regions 2.5 m on edge with 13 characters. >>> >>> Forgive me as it may be apparent by now that I'm not a geographer. I'm an >>> application engineer that builds wireless sensor networks and needs a >>> place >>> to store data based on sensor location - sometimes static - sometimes >>> mobile. I've got networks in the US, Pacific islands, India and Europe. >>> This >>> Recursive Triangular Mesh is what I'm using to do it. >>> >>> The database schema is nothing more than a wiki with a directory structure >>> that looks like "D/FH/KP/ET/SA/RO" - addressing a unique area of less than >>> 3 >>> square meters directly converted from a Lat/Long point. >>> >>> My other concern is the precision and accuracy of the location measuring >>> instrument - how many digits does my GPS provide? It defines the size of >>> the >>> final Trixel. >>> >>> - Brian >>> >>> >>> ------------------------------ >>> >>> Message: 2 >>> Date: Fri, 4 Jul 2008 08:30:48 -0700 >>> From: "Paul Ramsey" <[EMAIL PROTECTED]> >>> Subject: Re: [Geowanking] polarized light etc >>> To: [email protected] >>> Message-ID: >>> <[EMAIL PROTECTED]> >>> Content-Type: text/plain; charset=ISO-8859-1 >>> >>> Ah, you just want to discard all of Western thought back to Aristotle. >>> And here I thought you were a wanker. (categorization) >>> >>> P >>> >>> On Fri, Jul 4, 2008 at 2:07 AM, stephen white <[EMAIL PROTECTED]> wrote: >>>> On 04/07/2008, at 5:24 PM, Will King wrote: >>>>> So give us a sample problem, no waffle, Stephen and the pragmatists >>>>> on geowanking will probably come up with a pretty good solution. >>>> How can we organise all forms of captured information without >>>> categorising? Voxels? Turing? Red dots? Layers? >>>> >>>> That problem has the pre-requisite that you agree that categorisation >>>> damages information. >>>> >>>> -- >>>> [EMAIL PROTECTED] >>> [..cut] >>> ______________________________________________ >>> Michael Gould >>> Dept. Information Systems (LSI) >>> Universitat Jaume I, 12071 Castellón, Spain. >>> email: gould (at) lsi.uji.es >>> www.geoinfo.uji.es >>> >>> >>> >>> >>> _______________________________________________ >>> Geowanking mailing list >>> [email protected] >>> http://lists.burri.to/mailman/listinfo/geowanking >>> >>> >>> ------------------------------------------------------------------------ >>> >>> >>> No virus found in this incoming message. >>> Checked by AVG. >>> Version: 8.0.136 / Virus Database: 270.4.5/1536 - Release Date: 7/5/2008 >>> 10:15 AM >> _______________________________________________ >> Geowanking mailing list >> [email protected] >> http://lists.burri.to/mailman/listinfo/geowanking >> > > > ------------------------------------------------------------------------ > > > No virus found in this incoming message. > Checked by AVG. > Version: 8.0.136 / Virus Database: 270.4.5/1536 - Release Date: 7/5/2008 > 10:15 AM _______________________________________________ Geowanking mailing list [email protected] http://lists.burri.to/mailman/listinfo/geowanking
