Laser radar echo blind denoising method based on high-dimensional eigenvalue analysis

The invention discloses a laser radar echo blind denoising method based on high-dimensional eigenvalue analysis. The method comprises the following steps of arranging echo data into a matrix after a group of echo data is obtained, and solving a covariance matrix; eigenvalues of the covariance matrix being calculated and arranged in a descending order, and an eigenvalue array contributed by noise being extracted according to a threshold boundary; when the number of the data points is very large relative to the number of the samples, correcting a high-dimensional overestimation error, and calculating a noise level parameter through the eigenvalue array to obtain a Gaussian filtering variance corresponding to the noise level; and obtaining a self-adaptive Gaussian filtering function by considering the self-adaptive filtering variance parameter and the window width weight correction function, filtering the input waveform, and outputting the filtered waveform. According to the method, the noise level can be estima