Research On Noise Reduction Method Of Load Ground Tester Based On Power Integrity And Joint Filtering

It is very difficult for the conventional test system to detect such special signals,such as high frequency,weak and instantaneous signals,which are known but difficult to distinguish,in the satellite payload.This puts forward more demanding requirements of high speed and precision for load ground tester.In this paper,a noise reduction method based on power integrity and joint filtering is proposed,and the feasibility of this method is verified by experimental data.Firstly,aiming at the problems of insufficient power supply environment stability and high noise level in high-precision dynamic testing system,this paper proposes a power supply integrity parameter optimization method based on decoupling capacitor.The effect of equivalent series resistance and equivalent series inductance on impedance curve is determined by simulating the equivalent model parameters of decoupling capacitor.A discrete model is established to decouple the capacitor array,and the capacitance selection and quantity estimation of the parallel matrix model of the decoupled capacitor are realized by equalizing the impedance peak value,so as to optimize the power supply environment in the system and greatly reduce its impact on the performance of high-precision devices and the overall system test accuracy.Secondly,due to the impact of the precision of domestic components,when the hardware performance cannot be further improved,a joint filtering algorithm based on ensemble empirical mode decomposition/Wiener filter/particle filter is proposed to effectively solve the residual noise of the system.After mixed with white Gaussian noise,the measured signal was decomposed by empirical mode decomposition to obtain the inherent mode function set,which was divided into confidence interval,hybrid-mixing interval and noise interval according to the sample entropy.At the same time,the discrete entropy factor α is interpolated and fitted by double harmonic spline interpolation to predict the trend of previous empirical values of key parameters of sample entropy.Then wiener filter and particle filter were used to extract the effective components of the uncertain noise in the mixture interval and noise interval.Finally,the original real signal was reconstructed by using the reserved components to reduce the dynamic testing system of the loading ground tester to test the noise level.This article obtains from the power integrity,using empirical mode decomposition based on sets/wiener filtering/joint filtering particle filter algorithm for the introduction of the hard pieces of circuits in the complex environment uncertainty component,the influence of noise meet the test requirements of ground load tester,realize to known is difficult to distinguish the high frequency,narrow,and other special moment signal detection.