NavList:

   

Reply

I work on "inverse problems", things like tomography where the forward
problem is easy (and generally solved by nature) while the inverse
problem is hard, typically as it is ill-conditioned. An archetypal
example of this type of ill-conditioning is finding your position from
a set of position lines that are close to the same direction.

I think Tony's idea was that the computational complexity is hard the
way we want to go. Actually this is true of most non-linear systems of
equations unless you are lucky enough to find an analytical solution.

Bill

On 19 July 2017 at 21:23, Tony Oz  wrote:
> Hello!
>
> :)
>
> Suddenly it occurred to me that certain CN problems are similar to
> cryptographical: in one direction the solution is (relatively) easy and
> straightforward, in the other direction - quite difficult and laborious.
>
> Consider finding a GC distance between the two points of known coordinates -
> easy, just use one formula. The reversed problem of finding the initial
> coordinates from a known distance is impossible, just like the cryptography
> on elliptic curve(s).
>
> (I was thinking about a way to obtain AP coordinates from a set of
> Ho|GPcoordinates without any plotting - but directly by a formula solving)
>
> Regards,
>
> Tony
>
> 



--
Professor of Applied Mathematics
http://www.maths.manchester.ac.uk/bl

   

Reply