Non-parametric inversion of strong lensing systems

We revisit the issue of non-parametric gravitational lens reconstruction and present a new method to obtain the cluster mass distribution using strong lensing data without using any prior information on the underlying mass. The method relies on the decomposition of the lens plane into individual cells. We show how the problem in this approximation can be expressed as a system of linear equations for which a solution can be found. Moreover, we propose to include information about the null space. That is, we make use of the pixels in which we know there are no arcs above the sky noise. The only prior information is an estimation of the physical size of the sources. No priors on the luminosity of the cluster or shape of the haloes are needed, making the results very robust. In order to test the accuracy and bias of the method we make use of simulated strong lensing data. We find that the method reproduces accurately both the lens mass and source positions and provides error estimates.