Signals are information-bearing entities which are usually represented by one or more functions of one or more independent variables. For instance, a voltage signal u may have representations with respect to time and temporal frequency correspondingly, however, carrying the same information of this signal. Signal analysis focuses on the investigation into the basic properties and essential characteristics of signals. We attach more concentrations to the variations of signals according to the time domain, likewise, the idea can be applied to other spaces and variables. Discussions on the variations of signals in the time domain are natural, because time is the most basic variable. Nevertheless, it is far from enough to disclose the essence of the signals. One way out is to find other proper representations of a signal, which can characterize it from different points of view leading us better understand the signal. Theoretically, a signal can be rewritten by a combination of functions in a complete set. Actually, there are infinite ways, among which frequency is the most well-known and meaningful physical quantity. The mathematical foundation of the frequency representations of signals is invented by the famous French engineer and mathematician Jean Baptiste Joseph Fourier. On the basis of Fourier's work, the modern spectrum analysis, one of the most powerful tools in scientific world, is established attributed to Bunsen and Kirchhoff's great contribution.