The question is from the exercise questions from the EE4595 course being taught in TUDelft. The question is as follows...
So far my attempted solution is like the following.
The discretised data equation is given by $d=G_S\cdot q$ where $G_S$ is M-by-N data matrix and $q$ is N-by-1 source vector. To arrive at $q$ we need to invert $G_S$ and since $G_S$ is non-square it is non-invertible therefore, I thought the closest we get to the solution by using pseudo inverse resulting in $q=G_S^{\dagger}d$. However, the question apparently asks us to use singular value decomposition to show this process and this is where I got stuck. Any help to solve this question would be very much appreciated. Thanks in advance!