Measurement Of Impulse Signals In The Frequency Domain Based On Heterodyne Architecture

Waveform measurement of electromagnetic impulse signals is the key to the application of time-reversal(TR)technique in the area of electromagnetics and electronic engineering.Due to its simultaneous temporal-spatial focusing characteristic,TR of electromagnetic signals is promising in applications such as target detection,imaging and ultra-wideband wireless communications.This work studies the measurement of electromagnetic impulse signals in the frequency domain based on heterodyne circuit architecture,which enables the measurement of broadband pulse with low sampling rate and low cost.The research content includes the following four parts:In the first part,a theoretical analysis of the measurement of electromagnetic impulse signals in the frequency domain based on heterodyne circuit architecture is studied.The spectrum characteristics of periodic pulse signals are studied first,based on which the idea of the frequency domain measurement of a wideband signal based on spectrum segmentation and stitching is presented.With the spectrum of a wideband signal divided into several subbands,a heterodyne circuit architecture is adopted to down-convert the subband to baseband,which can be measured with low sampling rates directly.Based on this idea,a mathematical model of the frequency-domain measurement of time-domain waveforms is established.The algorithm of reconstructing periodic impulse signals by stitching measured spectra of subbands is developed.In the second part,the feasibility of the proposed measurement approach based on spectrum segmentation and stitching is verified by an ideal circuit system with high sampling speed.Then,the performance of the proposed approach with reduced sampling rates is studied experimentally.The experiment is designed to measure a pulse signal with 400 MHz bandwidth divided into seven segments,each segment recorded with a sampling rate of 250 MHz and stitched to reconstruct the original pulse waveform.The experimental results are as expected,which verifies the effectiveness of the method for reduce the sampling rate of waveform measurement.In the third part,an error analysis of the proposed measurement system is provided.The effects of the non-ideal amplitude and phase response of low pass filters,the nonlinear characteristics of mixer,and the sampling error of the analog-to-digital converter are analyzed.Particularly,compensation of the error introduced by low pass filters is applied,which effectively decreases the measurement error as expected.In the fourth part,a systematic compensation method is proposed to minimize the measurement error.Errors due to non-ideal hardware performance as well as random noise are calibrated statistically by multiple measurements of a known sample pulse signal.Experimental results show that this method can effectively reduce the overall error of pulse waveform measurement.Typical test results show that the overall measurement error can be reduced from 20% to 12%,leading to accurate waveform measurement results in the time domain.