The system for determining the activity of a person lying down has at least one motion sensor (CM) having at least one measurement axis, which is provided with a fastener (MF) for firmly connecting the motion sensor (CM) to a user. The system includes a filter (FILT) for selecting, for at least one measurement axis of the motion sensor (CM), a high-frequency signal (HF) and a low-frequency signal (BF) a first calculation unit (CALC1) for calculating a first variable (x(n)) representing a temporal variation of the low-frequency signal (BF) a second calculation means (CALC2) for calculating a second variable (y(n)) comprising the Euclidean norm, along at least one measurement axis, of the high-frequency signal (HF) and an analysis unit (AN) that determines an activity of prone user as a function of time using a hidden Markov model having N states corresponding to N activities respectively. The analysis unit (AN) also combines conjoint probability density functions of the first and second variables, said probability density functions being defined for each state of the model in question and probabilities of transitions between two successive states.
