"How to rename fields named 'data' in all the .mat files with respective filenames... file1.mat is a struct with field 'file1', file2.mat is a struct with field 'file2' ... filen.mat is a struct with field 'filen' "
I strongly recommend not doing that.
Processing data in multiple .MAT files is simpler and much more robust when the variable names in each .MAT file are exactly the same (just as your current files are). In contrast, adding meta-data into variable names (such as file numbers) is one way that beginners force themselves into writing slow, complex, inefficient, obfuscated code that is buggy and hard to debug:
Even if you LOAD into an output variable, then your new variable names will be more complex and slower to process than if you simply used the existing consistent fieldname "data".
If you need to store the meta-data (i.e. filename) in the .MAT files, then the best approach is to simply add it as a new variable into each .MAT file (which you can do very easily using the -APPEND option).
It is possible that you incorrectly believe that you need to give all of those variables different names in order to LOAD them into the workspace and prevent them from overwriting in a loop... but as you did not give much context in your question, this is just a guess.
