A method for sparse feature extraction of wireless channel based on energy entropy

Abstract

The invention relates to a wireless channel sparse feature extraction method based on energy entropy. The method includes the following sequence of steps: (1) transmitting a pilot signal, collecting received signals, and determining a measurement matrix; (2) performing approximate decomposition; (3) selecting optimal atoms to constitute an optimal atom set; and (4) extracting sparse features by using the optimal atom set. According to the scheme of the invention, the optimal atoms are selected, wireless received signals are expressed by using the optimal atoms, and thus the extraction of the sparse features of a wireless channel can be facilitated; the optimal atoms are selected by using energy information entropy, and the problem of sparse feature extraction of the wireless channel can besolved; and by considering the structural features of decomposition coefficients, the atoms corresponding to the most ordered decomposition coefficients are selected to constitute the optimal atom set, sparse coefficients can be obtained by using the optimal atom set, and thus the accurate estimation of the sparse features of the wireless channel can be achieved.

Description

technical field [0001] The invention relates to the technical field of wireless channels, in particular to a method for extracting sparse features of wireless channels based on energy entropy. Background technique [0002] The wireless channel is an important part of the modern communication system. The characteristics of the wireless channel are very important for analyzing the performance of the channel and improving the reliability and stability of the communication. Wireless channel feature extraction is the core step in the estimation and analysis of wireless channel transmission characteristics. Due to the multi-dimensional complexity and time-varying characteristics of wireless signals in the channel transmission process, there are technical bottlenecks in the feature extraction of wireless channels, which restricts the reliability of wireless communication transmission. Therefore, the issues related to wireless channel feature extraction are one of the major topics i...

AI Technical Summary

Problems solved by technology

Commonly used methods for sparse feature extraction include compressed sensing residual matching method, principal component analysis method, etc. These methods are more effective for well-structured signals, but for signals with poor order, because they do not consider the structural characteristics of signal decomposition coefficients and distribution characteristics, it is difficult to guarantee that the representation of the extracted signal is the most sparse feature representation

Images