I would think the dominant noise in a computer is the multiple clocks, their harmonics, and their mixing products - very nasty. It is the reason that building WLAN and Bluetooth into computers has been so difficult. So AWGN probably pales in comparison. In other words, the dominant noise in a computer chassis is not white noise (random) at all, but highly predictable.
This might be of use. Also, search on the NIST website for specifically what you're looking for. Our tax dollars have funded a lot ofresearch on the topic.
http://w3.antd.nist.gov/pubs/mswim01.pdfJimbo
