This doctoral thesis presents a comprehensive investigation into the ultra-sonic acoustic emissions originating from lubricated tribological contacts. The study encompasses a variety of machine elements, simulating specific operational scenarios. The primary objective is to explore quantitative relationships between AE signals, their associated tribological mechanisms, and their implications for macroscopic effects such as wear, friction, or other relevant properties. A consistent AE measurement technique was employed throughout the study, ensuring data comparability across numerous tribological setups, geometries, materials, and lubricants. A range of modern and useful post-processing methods for AE signals in time, frequency, and time-frequency domains were established. These include the classical root-mean-square (RMS) value, various forms of spectral techniques, and derived parameters combining AE data with classical tribological values. A robust database of AE connected to basic friction phenomena and mechanisms was developed using a wide variety of test strategies, tribo-materials, lubricants, and lubrication methods. An in-depth understanding of how specific AE abbreviated values are connected with tribological phenomena was established, and the previously described quantitative correlations were derived. Specific tools for monitoring or predicting certain tribological parameters, including machine learning derived models from AE, were developed. One predicting the friction power of a tribological system from its AE alone, independently of lubricant, sliding material and running conditions. The second system detects the aforementioned particle friction modes and levels of surface roughness. In conclusion, this research presents a comprehensive investigation of AE in the field of tribology, especially concerning lubricated contacts. Major accomplishments include the development of a usable, calibrated and refined measurement system for AE of various tribological systems, the refinement of well-established post-processing methods, the mapping of specific characteristics of AE to certain tribological phenomena and effects, and the derivation of two machine learning assisted systems. This work lays a solid foundation for future research in this field.