A significant portion of the tasks related to object-oriented identification of stochastic noise signals is associated with the analysis and investigation of narrowband signals. This chapter considers the envelope and phase method as the theoretical basis for solving the identification problems of such signals. It is demonstrated that applying the Hilbert transform to narrowband noise processes and signals allows for the unambiguous determination and investigation of their envelope and phase. The main properties of the Hilbert transform are presented, and the peculiarities of its implementation on finite time intervals are highlighted. Correlation functions of narrowband stochastic processes are discussed. A model of a random vector with independent Gaussian components in polar coordinates is analyzed. This model is used for the analysis of pre-envelope (analytic signal) at fixed moments in time. The probability densities of the envelope and phase are investigated for both narrowband noise stochastic processes and the additive mixture of amplitude-phase-modulated signal and Gaussian noise. Considerable attention is devoted to the use of phase characteristics in the identification tasks of narrowband signals. The specifics of applying discrete phase characteristics of narrowband noisy signals for determining circular statistics are discussed, which can be used as identifiers of narrowband processes. A methodology for using phase characteristics in the identification of narrowband noisy processes is proposed.