Method of an automatic recognition of sleep events is described. The method is based on analysis of amplitude histograms that are built on each of registered channels. The method is designed for sleep events, distinctive feature of which is expressed amplitude change that appears simultaneously on several channels. For example, "General body movement" and "Limbs movement" events. But this is not limited; method can be applied to other events, for example "Snore". Algorithm can be divided into three stages: histograms building and analysis, vertical analysis, results processing.

Keywords: PSG, sleep events, histogram, detection algorithm