Unwedge does not contain a back analysis option. What you have to do is run a number of deterministic models varying c and phi independently (assuming you want to incorporate cohesion). There will be a number of combinations of c and phi that give a factor of safety less than 1. There will not be one unique answer for one particular wedge. It will be up to you, using wedges from multiple locations in the tunnel, to come up with a c and phi value that best fit all the wedges in the tunnel that are at a state of limit equilibrium.