Power Spectrum Estimation

Power spectral density (PSD), or simply the power spectrum, of a signal is a representation of the spread of signal power as a function of frequency. This chapter reviews basic concepts and techniques of power spectrum estimation. Power spectrum estimation is the first step of signal processing a cognitive radio may perform in order to gain spectrum knowledge. Based on this sequence of discrete-time signal samples and assuming that the signal can be modelled as a second order (wss) stationary random process, the radio may compute the power spectrum corresponding to the sensed signal. All other nonparametric approaches to PSD estimation essentially trade off the spectral resolution to reduce the variance. The Blackman–Tukey PSD estimate is one way to achieve this. The chapter also discusses two other classical PSD estimators that achieve lower variance than the periodogram by trading off the spectral resolution.