Inpracticalparameterestimation,wehavealwayschoseneitherLeastSquaresEstimation(LSE)orRobustEstimation.Sincethedistributionofobservationsisunknown,toselectacorrectestimationmethodisverydifficult.Itiswellknownthatifobservationsincludegrosserrors,theresultofLSEwillbebadlycontaininated.Ontheotherhand,ifobservationsdonotincludeanygrosserrors,theresultofrobustestimationisnotasgoodasthatofLSE.Tosolvethisproblem,Wang(1999)developedanestimationmethodcalledInformationSpreadEstimation(ISE)basedontheinformationspreadprinciple.TheISEisaverygoodmethodforestimatingoneparameterwhichisveryrobust.However,mostofinstancesinsurveyingdataprocessingaremulti-parameters'estimation,owingtotheinherentrestrictionsofISE,itcannotbeappliedtothesurveyingdataprocessingdirectly.Toapplythegoodmethodtothefieldofsurveyingdataprocessingwidely,theauthorhasdonetheresearchdeeply.ThispaperappliesISEsuccessfullytotheadjustmentoflevelingnetworkbyusingthespecialtiesofleveling.
