Compressive sensing offers an alternative to Nyquist sampling in applications where taking a complete set of measurements is impractical due to either length of time required or the volume of data produced. Here we focus on applying compressive sensing theory to RF signal environments (which will be referred to as the "signal"). For multi-gigahertz bandwidth signals, using Nyquist sampling requires billions of samples per second resulting in data rates of tens of billions of bits per second. This is the cost of being able to capture any arbitrary signal. Compressive sensing exploits the fact that man-made signals are structured and non-arbitrary to greatly reduce the number of samples required for signal capture. The compressibility of a signal is measured by its sparsity, which can be thought of as the number of parameters needed to describe the signal. Compressive sensing can reduce sample rate by more than an order of magnitude.
Read full abstract