A bidirectional DFT symmetrical compensation phase measurement method and device thereof

Abstract

The invention discloses a double-direction DFT symmetric compensation phase measuring method and device. The method comprises the steps of performing DFT on a sample respectively with the forward and backward counts with the number of N, extracting the phase measuring values Phif (k*) and Phib (N-k*), and performing superposition averaging on the two phase measuring values to obtain a phase measuring result of the {x(n)} central sample position. The device comprises an analog-digital converter, a DSP device and a drive and display module, a to-be-measured signal with the frequency f0 passes through the analog-digital converter for sampling to obtain a sample sequence x(n), the signal enters the DSP device in a parallel digital input mode, internal algorithm processing of the DSP device is performed on the signal, and parameter estimation of the signal is obtained. The drive and display module displays the phase measuring result, the phase measuring variance and the Cramer-Rao lower bound. According to the method and device, unbiased estimation is performed on the phase, the blank of effective and precise estimation of signal phase parameters is filled in, and the high-precise phase measuring of complex exponential signals can be achieved.

Description

technical field [0001] The invention relates to the field of digital signal processing, in particular to a bidirectional DFT symmetrical compensation phase measurement method and a device thereof. Background technique [0002] Classical complex exponential signal a·exp[j(ω 0 n+θ 0 )] The phase estimate is in communication [1] , GPS navigation [2] , Power harmonic analysis [3] , Optical Engineering [4] The important problems widely encountered in other fields have aroused the general attention of relevant scholars at home and abroad. [0003] Phase estimation can be divided into analog method and digital method. Digital phase estimation has become the mainstream of phase estimation methods because of its advantages of high accuracy, flexible measurement process and no need to introduce complex circuits. For the digital phase estimation method, the index to measure its measurement accuracy is the mean squared error (Mean Squared Error) calculated statistically from mult...

AI Technical Summary

Problems solved by technology

[0005] However, these phase measurement methods proposed in literature [7-19] need to estimate the signal frequency value first and then use the frequency value to correct the phase, so the frequency estimation error will be brought into the phase estimation error, resulting in poor estimation performance

As pointed out in literature [19]: "The bias of the frequency estimate will be brought into the phase estimate, thus increasing the uncertainty of the phase estimate." Undoubtedly, this uncertainty makes deriving the theoretical performance of the phase estimate very difficult. Therefore, the phase estimation methods in literature [7-19] have not derived the theoretical expression of the phase estimation variance

For the above DFT-based phase measurement methods, since frequency estimation cannot be avoided in the phase measurement process, the phase measurement involves estimating three parameters [a,ω 0 ,θ 0 ] T , so the phase estimation variance is also determined by the three-parameter CRLB (denoted as CRLB 3 ) limited by

Images