Help with Learning Curves in Matlab

[candela]

Hi everybody. I'm finishing my final project about echo cancellation in MATLAB with three kind of adaptive algorithms (LMS,NLMS,RLS y FDAF). I have done all the main code and I'd got suppress a part of acoustic echo. Now I have to presents the Learnig Curves to study convergence of each algorithm and I do not know where to start. I have been working with the Matlab function "msesim", but I don't understand it. So I appreciate any help. Thanks

 

[etmabreu]

If your algorithms converge that means that the RMS goes near zero. See the help of that function, it has graphs for the convergence.

 

[candela]

Thanks very much. I have been reading properties in Matlab Help, but I don't know how to use RMS function.
The Matlab help says:

H = dsp.RMS returns a System object, H, that computes the root mean square (RMS) of an input or a sequence of inputs over the specified Dimension.

In my code, I have an input signal (x), an Impulse Response (H) and a desired signal (dn). The inputs to the adaptive algorithms are x and dn (where dn is x filtered with H). Then, how could I use RMS function?
Thanks again.