The invention discloses a traction stretching motion state monitoring method based on
signal processing, and relates to the technical field of
signal processing, and the method comprises the following steps: collecting traction force, electrocardio, displacement and
conductivity signals, carrying out normalization and filtering
processing, extracting phase and amplitude components, and constructing a
signal spectrum containing track change and
time sequence characteristics; establishing a structure response mapping relation to generate a self-evolution feature map, and aligning the self-evolution feature map with the amplitude component to form a channel-level
feature matrix and a multi-channel feature representation map; structural path
perception convolution mapping is adopted, a path
dissipation factor is introduced to extract deep features, direction enhancement is achieved in combination with an
energy constraint factor and a scheduling matrix, and an energy adjustment feature map is obtained; finally, the features are spliced according to time steps to form a high-resolution sequence,
signal optimization is achieved through iterative training, and the
state recognition accuracy and stability are improved.