Real-Time Algorithm for Dimensionality Reduction and Deconvolution Recognition of English Complex Signal Features based on Openstack Architecture

Abstract

On the basis of the OpenStack computing architecture, the problem of low computational efficiency of feature extraction for English complex signals in a single computing node is deployed and solved. Principal component analysis (PCA) dimension reduction processing and random projection based on the recognition contribution rate of the two features (RP) The processing results are weighted and fused, and then the results are provided to the convolutional neural network for processing to extract the high-level features of image classification. A convolutional code recognition algorithm. The experimental results show that the dimensionality reduction effect of the parity check matrix recognition algorithm is the best.