Significantly speed up C++ which is used in Matlab

I have a function in C++ wrapped up for usage in Matlab. I need to speed it up significantly. I have the target code to compare with. The goal is to achieve similar performance as with the target code. The tasks would be:
1. Identifying weak parts of C++ function (you will need to show me exactly which parts of the code slow down the whole process)
2. Suggest good ways to improve it