On a related note (locating a mobile phone - not necessarily triangulation) 

The Hepkie system uses ordinary GSM cell phones as beacons, without modifications or any additional programs needed on the phone. The Base Unit triggers signaling from the cell phones in the coverage area so that the direction to them can be detected with the Search Unit(s). The Base Unit is fully automatic and requires no operator, and it presents a list of all cellular phones in the search area to the Search Unit.

The operator of the Search Unit can search for all phones present, but can also choose to search for one or several specific phones. The phones belonging to the rescue team or are identified as not belonging to a victim can easily be excluded from the search.

The system is independent of cellular network coverage and it is independent on the service providers. Field trials have showed that the possible range of detection is very large, over 10 km and that it is possible to locate buried cell phones in snow with an accuracy of around 30 cm at a victim depth of 180 cm. There are also very positive results from search in collapsed buildings.

http://www.mhbc.se/Classof2009/resqu/

Regards,
Tito

On 23 August 2011 10:04, Kinuthia Ngugi <kinuthia.ngugi@gmail.com> wrote:
Maybe there is something that needs to be clear;

1. Your major interest is to pinpoint the location of a mobile device, and to do this you have two definite references, that is the timing advance
obtained from the MRs sent by the mobile to the network, and the serving cell which you definitely know where it is from the co-ordinates. Now,
the 6 neighbors that you get in the periodic MRs cannot be used reliably for the location purposes,why? Because they are just the best signals
that a mobile is 'seeing' arranged in order of strength, so a cell could actually be seeing a very distant cell which happens to be overshooting into
that location due to favourable propagation conditions (better line of site, etc etc). Note that there is no TA info about those best 6 neighbours, see 2 next for the reason. So those 6 cells will not give you reliable location info, save for knowing there locations from co-ordinates. The attenuation angle
you were thinking of is admirable but very simple. Radio signals sometimes take a life of their own once launched into free space despite the best modelling.

2. A mobile only sends MRs when it's in dedicated mode, meaning that it's in an active connection. Also, a mobile can only be served by one cell at a time, so the TA information you get is only from the serving cell. IF, you want to get MRs from another cell, the mobile has to be handed over to that cell so that it becomes the current serving cell. The vectors theory mentioned above needs info from at least two cells, so if you can have a way of making the mobile handover between specific cells at your will, the triangulation effort becomes very easy. However, that is very difficult outside a lab
environment; GSM has very specific algorithms on how a mobile hands over, distinct parameters are set for this. You might be wanting the mobile to hand over but it won't to do that until the right conditions are met.

Note that i am taking the simplified case of a rural set-up with few known cells. In an urban set-up like Nairobi it would be confusion galore! Maybe I should ask what precision are you looking to attain, in meters? Just my thoughts.



On Tue, Aug 23, 2011 at 8:40 AM, hellen masita <hellenmasita@hotmail.com> wrote:
ok sot the drawing disappeared!!! see attached for it

From: hellenmasita@hotmail.comDate: Tue, 23 Aug 2011 08:37:37 +0300

Subject: Re: [Skunkworks] Discussion: Triangulation in GSM

@steve,

i agree, using just the timing advance and serving Cell may give to wide an area for estimation, but that is why i said we get some readings for 6 of the strongest neighboring cells that the mobile station can sense!!!!

option 1
so far, this is what i am considering;
 attenuation on the air interface can be estimated using engineering models, depending on whether it is a rural or urban site, given the attenuation, can we not estimate the distance from the other cells and use this to narrow the possible location.the  challenge with this idea is that most engineering models come inbuilt into specific expensive software and might not flexible to be used as stand alone

option 2,
About the vectors, even without the attenuation model, give that i can get the coordinates of all the sites and azimuths of the sectors(where they are pointing) of the serving cell, how about if we draw vectors and check the points of intersection and areas falling within this points.
option 3
use both of the above together





excuse my drawing, but see what i mean, if sites B to F are the six strongest neighbors, we could use them to narrow down the location.....


> Date: Thu, 13 Oct 2011 18:21:24 +0300
> From: steve@sobbayi.com
> To: skunkworks@lists.my.co.ke
> Subject: Re: [Skunkworks] Discussion: Triangulation in GSM
>
> Guys don't worry about about the complexities, Its easy once you get used to the laws and properties of numbers
> and shapes
>
>
>
> ----- "Timothy Mutugi" <timothymutugi@gmail.com> wrote:
>
> > ....ok @washi, it is just that I have been called a n00b before and
> > so
> > I am feeling kinda shy and embarrassed and "shame-on-me" to as what
> > this triangulation thing is all about...
> >
> > ....I did home economics and so perhaps the only thing I can
> > triangulate is a samosa because it looks triangular...
> >
> > On 8/22/11, Odhiambo Washington <odhiambo@gmail.com> wrote:
> > > LOL!!!!
> > >
> > > @Mutugi - if you have no idea, then it's time you shed off your
> > *skunk*
> > > skin<LOL>
> > >
> > > Anyway, the way @Obbayi, put it out, you surely need to have been
> > present in
> > > a classroom where they taught engineering mathematics.
> > >
> > > On Mon, Aug 22, 2011 at 15:40, Timothy Mutugi
> > > <timothymutugi@gmail.com>wrote:
> > >
> > >> ?
> > >>
> > >> I have no idea of what is being discussed here, but it sounds very
> > >> skunk-like and highly technical...good post, keep it up:-)
> > >>
> > >> Timo.
> > >>
> > >> On 10/13/11, Steve Obbayi <steve@sobbayi.com> wrote:
> > >> > There are two ways of doing this. Although I am not an expert on
> > GSM and
> > >> BTS
> > >> > etc, The solution here is common to
> > >> > that which I would commonly use in 2D/3D graphics programming.
> > Hope you
> > >> > understand a bit of linear algebra and geometric
> > transformations.
> > >> >
> > >> > Nways, here goes
> > >> >
> > >> > 1. The direction doesn't matter, If you can get readings of three
> > or
> > >> > more
> > >> > points as zero vectors then you can compute the resulting vectors
> > of the
> > >> > location of the device.
> > >> > Using two initial zero vectors wont work as your computed results
> > will
> > >> > intersect at two possible vectors.
> > >> >
> > >> > which brings me to the second method
> > >> >
> > >> > 2. Using your trigonometry skills, you can calculate the the
> > angle
> > >> between
> > >> > the two fixed known points of the BTS and the two possible third
> > points
> > >> of
> > >> > the resulting two triads based on the distance of at least two
> > BTS
> > >> stations.
> > >> > This should get you good reading.
> > >> >
> > >> > Steve
> > >> >
> > >> > ----- "Kinuthia Ngugi" <kinuthia.ngugi@gmail.com> wrote:
> > >> >> Am afraid in GSM, the TA and serving cell is as narrow as it can
> > get.
> > >> You
> > >> >> can not hope to have more than half a kilometer precision.
> > Normaly,
> > >> timing
> > >> >> advance goes in units of 550m, so TA 0 means 550m form the BTS,
> > TA 1
> > >> means
> > >> >> 1..1KM, so and on. Then there is the issue of how wide a cell's
> > >> >> coverage
> > >> >> is; you could be having a BTS on a hill with an antenna that has
> > a
> > >> >> 90deg
> > >> >> beamwidth in the far field....that could mean a span of like
> > 10km
> > >> >> across
> > >> >> that field, so even if the TA tells you are 3km from the BTS,
> > but in
> > >> which
> > >> >> direction? North or South or East? such are the challenges. All
> > the
> > >> >> best
> > >> >> though.
> > >> >>
> > >> >>
> > >> >> On Mon, Aug 22, 2011 at 10:11 AM, hellen masita <
> > >> hellenmasita@hotmail.com
> > >> >> > wrote:
> > >> >>
> > >> >
> > >> >
> > >> >
> > >> >> Hi skunks/skunkettes,
> > >> >>
> > >> >> in line with an earlier discussion on this being a technical
> > forum...
> > >> >>
> > >> >> May i call upon your help without monetary gains taking an
> > upper
> > >> hand????
> > >> >> please :)
> > >> >>
> > >> >>
> > >> >>
> > >> >> i want to try and create an application for triangulating a
> > mobile
> > >> phone.
> > >> >> i am at a loss on where to start....really!!! so i throw it to
> > the
> > >> floor.
> > >> >>
> > >> >> *******************
> > >> >> In a GSM network with than one vendor a proprietary software is
> > >> >> normally
> > >> >> supplied to handle their equipment including the BTS.
> > >> >>
> > >> >> Mobile phones however send Measurement reports which are
> > standardized.
> > >> >>
> > >> >> these reports contain the following,
> > >> >>
> > >> >>
> > >> >
> > >> > 1. signal level of 6 best neignbours
> > >> >> 2. signal quality\level of serving cell
> > >> >> 3. the BTS can determine the Timing Advance of the Mobile(The
> > distance
> > >> >> (usually a range) the mobile is from the bts) as well as serving
> > cell
> > >> >> (Sector)
> > >> >>
> > >> >> Now, with just the Timing Advance (TA) and Serving cell , i can
> > tell
> > >> >> the
> > >> >> general direction of a user. however i would like to use the
> > other
> > >> >> measurements to narrow down the location
> > >> >>
> > >> >> *************************
> > >> >>
> > >> >> The significance of this is that it a network can track a phone
> > that
> > >> does
> > >> >> not have a GPS. it would make phone thefts less bothersome
> > among
> > >> >> things.
> > >> >>
> > >> >> If guys discuss it for a bit, i'm sure i'll find my footing,
> > >> >>
> > >> >> regards,
> > >> >>
> > >> >> Hellen
> > >> >>
> > >> >>
> > >> >>
> > >> >>
> > >> >>
> > >> >> _______________________________________________
> > >> >> Skunkworks mailing list
> > >> >> Skunkworks@lists.my.co.ke
> > >> >> http://lists.my.co.ke/cgi-bin/mailman/listinfo/skunkworks
> > >> >> ------------
> > >> >> Skunkworks Rules
> > >> >> http://my.co.ke/phpbb/viewtopic.php?f=24&t=94
> > >> >> ------------
> > >> >> Other services @ http://my.co..ke
> > >> >>
> > >> >>
> > >> >> _______________________________________________ Skunkworks
> > mailing list
> > >> >> Skunkworks@lists.my.co.ke
> > >> >> http://lists.my.co.ke/cgi-bin/mailman/listinfo/skunkworks
> > ------------
> > >> >> Skunkworks Rules http://my.co.ke/phpbb/viewtopic.php?f=24&t=94
> > >> >> ------------ Other services @ http://my.co.ke
> > >> _______________________________________________
> > >> Skunkworks mailing list
> > >> Skunkworks@lists.my.co.ke
> > >> http://lists.my.co.ke/cgi-bin/mailman/listinfo/skunkworks
> > >> ------------
> > >> Skunkworks Rules
> > >> http://my.co.ke/phpbb/viewtopic.php?f=24&t=94
> > >> ------------
> > >> Other services @ http://my.co.ke
> > >>
> > >
> > >
> > >
> > > --
> > > Best regards,
> > > Odhiambo WASHINGTON,
> > > Nairobi,KE
> > > +254733744121/+254722743223
> > > _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
> > > I can't hear you -- I'm using the scrambler.
> > > Please consider the environment before printing this email.
> > >
> > _______________________________________________
> > Skunkworks mailing list
> > Skunkworks@lists.my.co.ke
> > http://lists.my.co.ke/cgi-bin/mailman/listinfo/skunkworks
> > ------------
> > Skunkworks Rules
> > http://my.co.ke/phpbb/viewtopic.php?f=24&t=94
> > ------------
> > Other services @ http://my.co.ke
> _______________________________________________
> Skunkworks mailing list
> Skunkworks@lists.my.co.ke
> http://lists.my.co.ke/cgi-bin/mailman/listinfo/skunkworks
> ------------
> Skunkworks Rules
> http://my.co.ke/phpbb/viewtopic.php?f=24&t=94
> ------------
> Other services @ http://my.co.ke

_______________________________________________ Skunkworks mailing list Skunkworks@lists.my.co.ke http://lists.my.co.ke/cgi-bin/mailman/listinfo/skunkworks ------------ Skunkworks Rules http://my.co.ke/phpbb/viewtopic.php?f=24&t=94 ------------ Other services @ http://my.co.ke

_______________________________________________
Skunkworks mailing list
Skunkworks@lists.my.co.ke
http://lists.my.co.ke/cgi-bin/mailman/listinfo/skunkworks
------------
Skunkworks Rules
http://my.co.ke/phpbb/viewtopic.php?f=24&t=94
------------
Other services @ http://my.co.ke


_______________________________________________
Skunkworks mailing list
Skunkworks@lists.my.co.ke
http://lists.my.co.ke/cgi-bin/mailman/listinfo/skunkworks
------------
Skunkworks Rules
http://my.co.ke/phpbb/viewtopic.php?f=24&t=94
------------
Other services @ http://my.co.ke