[TML] Jump variation

[shadow at shadowgard.com]

On 5 Sep 2007 at 3:08, Jerry W Barrington wrote:

> I would say at least 4 sensors, well separated, combining signal strength &
> timing with their own positions.  Hmm.  If the events are rare enough, they
> could be omni-directional and still give you accurate fixes.
> 
> The reason I say 4 instead of 3 like Greg is their is always the possibility
> of up or down entry.  3 sensors can only fix to a point in their mutual
> plane, with height above or below that plane unknown.  4 sensors not in the
> same plane fixes that.

3 sensors not on the same line can generate a "point" for a source in 
their plane. And a "line" for a source not in their plane. 

That not in the same line is important. 

And in the real world, you have to worry about blind spots for each 
sensor cause by various bodies. So you'll actually need more than 4.

--
Leonard Erickson (aka shadow)
shadow at shadowgard dot com