This topic has been discussed heavily on RG, and if you simply search for 'dislocation density xrd' in the search bar, you will find a lot of discussions covering many aspects of this.
In essence, you compare the peak shape, height and width of deformed specimen to pre-deformed (ideally strain free) specimen. The comparison is made using various models, such as that of Warren Averbach (widely used but now recognised to have limitations), whole pattern fitting (Rietveld refinement- bit of a top-down approach), or whole pattern modelling by MWP/CMWP methods (considered by some to give the most microstructural information). With a high symmetry material like yours all these methods will be applicable and hopefully not too complex (though experts warn of plenty of complexity when you dig deeper).