This monograph explains my research in teletraffic modeling by using two-parameter correlation function from the point of view of abstract analysis in Hilbert spaces. Methodologically, the book utilizes the extensions of fractional Gaussian noise (fGn) to study teletraffic modeling such that the extensions with two parameters are more flexible and accurate for traffic modeling than fGn, independent of the generalized Cauchy process, which was recently noticed in stochastic processes. The monograph is in the style of combining abstract analysis of traffic with processing real-traffic data. It focuses on the correlation form of traffic. For some fractal properties of traffic, such as multi-fractal, readers may refer to other references or some references in this book. This monograph may be a reference for postgraduates, scholars, and engineers in computer science, as well as those who are interested in traffic time series in applied statistics.