The invention relates to a physiological parameter monitoring-combined human body tumble warning and detecting method. According to the physiological parameter monitoring-combined human body tumble warning and detecting method, human body physiological parameters are monitored and analyzed in real time through devices such as a mobile phone and a tablet personal computer to establish an easy-to-tumble model to perform tumble warning; a human body physiological parameter and movement data synchronous collection platform is established to achieve synchronous collection of the human body physiological parameters (such as oxyhemoglobin saturation, the pulse rate and the pulse signal intensity) and the movement data (accelerated speed at a waist hip bone position and the trunk instant angle); a tumble warning and detection model is established in combination with the human body physiological parameters and the movement data, tumble warning is performed by establishing the human body easy-to-tumble model, and meanwhile an SVM tumble detection algorithm is provided to perform tumble detection; finally characteristics of collected data are extracted and scaled, related characteristic attributes are obtained, and are in combination with type information to form a training set sample; a classification forecasting model is obtained by training the training set to be used for predicating categories of samples in unknown types, characteristic vectors in the tumble activity are added to a training data set to be used for a next round of training.