Researching Of Mutiple Parameter Modifying In 3G Radio Propagation Model

In order to design and deploy WCDMA network we must get some accurate radio propagation models that can describe the loss characteristics of the radio wave on some spectrum in some geographical environment correctly. Generally the propagation models based on statistics are used widely. This paper uses what is named as Standard MacroCell Model that is applied to prediction of macro cell signal usually. And this paper aims at modifying some relative parameters of the propagation models according to local environment. Finally the modified model can be used to design and deploy 3G wireless network.Statistical model modification is a method based on a great deal of data from all kind of tests. According to theory of radio wave propagation signal will change slowly in random way that can be described by logarithm- normal distribution. Lee's theory states that if we get the average level of a signal on about 40 wave-length we can estimate its envelop value called as local mean value against the mean value on special area. We took advantages of a CW on a pre-assigned frequency band as test signal to acquire signal mean value of some area then get the local mean values that were used to modify the parameters of local radio propagation model. Accordingly,the local radio propagation model was determined correctly. The procedures as follow:1) Collect primary data: The accuracy of primary data has direct effect on modification of parameters. A series of measures were adopted to make sure the accuracy of data.2) Process testing data: The job of filtering some data not meeting needs and interpolating the value of longitude and latitude was done well. Some software was developed to analyze data, such as averaging data geographically and modifying map.3) Modify the parameter of model: Ericsson's tool named as TCP was made use to modify the parameters of model.4) Validate accuracy of model: The results that we had gotten from the above procedures were validated to watch if it had enough accuracy.Up to 9 models' parameters for different terrains were modified. The gross error is within 8 dB, which can meet the requirements of industry in fact. The models modified and validated are applied to design and deploy 3G wireless networks for GMCC.