A Phase Correction Method for Passive Direction Finding Channel

Abstract

The invention discloses a phase correction method for a passive direction finding channel, which uses a DNN network to extract the phase characteristics of a signal received by a receiver, converts the correction problem into a DNN network feature extraction problem, and finds out the difference between the phase difference of the channel output signal and the original phase difference of the antenna signal mapping, and correct channel phase errors. The phase error model is established for the received signal, and the product of the antenna signal and the channel frequency response function is used as input, and the full phase correction is achieved by correcting the initial phase difference of the antenna signal from 0 to 180° at sparse points. Compared with the traditional method, the present invention can have better flexibility, and by presenting the phase of the corrected signal in the form of standard deviation, it can better explain the stability of the network, thereby bringing better correction effect.

Description

technical field [0001] The invention relates to a passive direction finding channel phase correction method, in particular to a DNN-based passive direction finding channel phase correction method, which belongs to the field of radar signal processing. Background technique [0002] The inconsistency of the channel amplitude and phase will seriously degrade the performance of a series of array signal processing algorithms. The channel mismatch will degrade the DOA estimation performance, which will seriously degrade the performance of the beamforming algorithm. The channel mismatch causes the sharpness of the MUSIC angle estimation spectral peak to decrease and the spectral peak to be biased, and when the channel mismatch is serious, it will cause the inability to distinguish two signals with similar angles, and the angle estimation accuracy is greatly reduced. The inconsistency of amplitude and phase between channels is manifested in different frequency response functions be...

AI Technical Summary

Problems solved by technology

[0005] In view of the above-mentioned prior art, the technical problem to be solved by the present invention is to provide a passive direction finding channel phase correction method based on a deep neural network, so as to solve the problem that the accuracy of the channel correction method for designing an equalizing filter is not enough and the effect is not good And problems such as insufficient flexibility, establish channel mismatch model data, and correct the sparse phase point data of the received signal on the frequency band, so as to achieve full phase correction

Images