Compressive sensing and sub-Nyquist sampling

The Nyquist-Shannon sampling theorem states that if all that you know about a signal is that its highest frequency is no greater than B, then if you want to sample the signal without any loss of information, you must sample it at a rate that is greater than 2B. If we call the rate at which we sample Fs, then we find that B < Fs/2, and Fs/2 is called the Nyquist frequency. For a simple proof of the necessity of the condition B < Fs/2, please see Sidebar: The Necessity of the Condition B < Fs/2.