Abstract
This article contains a brief review of the contemporary empirical pattern of 1/f noise in conducting systems and of attempts of its theoretical interpretation. It is noted that a spectrum of the form ~ 1/f can be explained either as a consequence of the continuous hierarchy of macroscopically long relaxation times, or, on the contrary, as a result of absence of long-lived correlations and a characteristic time macroscale ("scaleless" 1/f noise). Theories of 1/f noise (fluctuation models of the number of charge carriers, temperature fluctuations, and "degradation" models) are traditionally directed toward the first type, and restrict the treatment to specific types of systems. The second type makes it possible to detect universal features of 1/f noise and, concealed in the formal fluctuation models of mobility carriers, it allows for a uniform and successful description of many situations. An explanation was recently suggested for the nature of mobility fluctuations, the diffusion coefficient, and other kinetic quantitites. It is noted that the current values of kinetic quantities are quite uncertain. Empirically this uncertainty is perceived in the form of "scaleless", flickering fluctuations of flowing kinetic quantitites. It also implies that real Brownian motion of carriers is characterized by non- Gaussian statistics. Analysis of the statistics leads to a spectrum of precisely 1/f type and to expressions for 1/f noise levels (depending only on microscopic space-time interaction scales of carriers with the medium) in satisfactory agreement with experiment.