The invention relates to an identification method. In order to provide a better gait identification result, reduce work load during a human movement detection process and lower the requirement on hardware facility, the technical scheme of the invention comprises the steps of: S1. acquiring human body pyroelectricity signals; S2.outputting and amplifying the acquired human body pyroelectricity signals and carrying out filtering processing at the same time; and S3. carrying out data processing on the processed human body pyroelectricity signals, wherein the data processing includes wavelet transform, independent element analysis, K-Means clustering algorithm processing and identification. The invention is mainly applied to linear FM (Frequency Modulation) signal parameter estimation in digital signal processing.