Consider an empty room with one point source that emits a stationary signal (constant sound, radioactive radiation, ...). The energy nor the position of the point source is known. We send someone in the room to do some intensity measurements on different positions in the room.
With this collected data I want to determine the position and the energy of the point source. Which mathematical theory can be used to do such calculations?
Assuming the data set is in the form of $$\{(E_i,x_i,y_i,z_i)\}_{i=1}^{i=n}$$ And Given the following notations:
We know (from Physics) that that the measured energy decreases in proportion to the distance from the source squared. $$E_*\propto \frac{E_i}{(x_i-x_*)^2+(y_i-y_*)^2+(z_i-z_*)^2}$$ Adding a proportion coefficient $K$ we get this equation $$E_i=KE_*((x_i-x_*)^2+(y_i-y_*)^2+(z_i-z_*)^2)$$ We just got ourselves a model with the 5 parameters: $$\theta=(E_*,x,_*,y_*,z_*,K)$$ The parameters can be approximated using Quadratic regression (Maximum Likelihood approach) or Bayesian inference with $K$ as a nuisance parameter.