A Method of Improving Robust Estimation Using Noise

Abstract

The invention relates to the field of signal and parameter estimation, in particular to a method for improving the performance of a robust estimator by utilizing noise. The method is capable of reducing the mean square error (MSE) of the robust estimator on the basis of correlation principle of stochastic resonance. The noise mainly has two effects. Firstly, when the background noise is in thick tail symmetric distribution, L robust estimators are connected in parallel, independent and identically distributed noise eta is added into each estimator unit, the output of the L estimators is calculated and averaged to serve as an estimator of a system, the asymptotic efficiency of the estimator relative to the maximum likelihood estimator is calculated, and the asymptotic efficiency of the estimator of the system can be improved after the specific level of noise eta is added; secondly, under an asymmetric pollution distribution noise model, the maximum asymptotic deviation of the robust estimator can be reduced by adding the noise to observation data. The method is capable of further improving the estimation precision of the robust estimator when abnormal data exists, and can be used as an important means for improving robust estimation.

Description

technical field [0001] The invention relates to the field of signal and parameter estimation, in particular to a method for improving the performance of a robust estimator by using noise. Background technique [0002] In signal estimation and detection problems, we often assume that the population obeys a certain type or type of distribution, and on this basis, select the optimal criterion to design the corresponding estimators and test statistics. But in fact, due to the impact of impulse noise in the measurement activities, abnormal values ​​will be generated in the received or observed data. For example, in the outdoor wireless mobile communication channel, it will be interfered by the transient noise of the switch in the power line or the pulse noise of the car ignition; in the radar and sonar system, it will be interfered by the pulse of natural or artificial electromagnetic and sound waves; During position estimation and tracking, obstacles such as buildings and trees...

AI Technical Summary

Problems solved by technology

In signal estimation, such impulsive noise causes a large amount of observation data to have some "abnormal data", which makes the assumed ideal distribution model deviate. In this case, the estimator under the optimal criterion is difficult to achieve, and the traditional statistical The method is greatly affected by individual abnormal data, and it is against this background that the robust estimation technique emerges.

Images