Gsm Tdoa-based Positioning Technology

With the continuously construction of the mobile communication infrastructure, the pure coverage could not meet all kinds of increasing demands of the subscribers. Several value-added services especially the location services gradually become the heat point in the field of communication.Actually wireless location service aims to locate the mobile terminal (location information includes the related coordinates in general). The radio parameters related with the mobile location consist of the propagation time, signal field-strength, angle of arrival, etc, so the two-dimension location can be gained by detecting and analyzing these data. Of course, the different parameters and the different corresponding methods cause the dissimilar effect. Therefore, the topic of wireless location technique makes great sense. In the beginning of this article, it presents some theories of main wireless location technique currently, and their strength & weakness, and also the possible application in part. Besides, the function entities together with system interfaces are introduced, which is to interwork with the current network element.The majority of this article analyzes a few classic algorithms based on TDOA, including Fang, Taylor series iteration and Chen algorithms, etc, and also their simulating test. The improvement is proposed. It's a combination that Taylor iteration based on the initial from the result of SX algorithm (Combine for short ). In order to speculate and analyse the performance of all algorithms raised, the software of Matlab is deployed for the programming, then the simulation platform of GSM system is set up,which has taken into account the factors of multi-path and NLOS affection, some parameters of the channel model are also referred to the part of the specification of T1P1. The tests are conducted via ideal criteria of Gauss with zero mean and the simulation criteria that splits into the different area environment model. These models of urban areas, suburb and rural areas are decided by the cell radius and specific radio parameters. After that, the illustrations of cumulate distribution function are formulated based on that output from simulation platform with the corresponding analyses.The result from the test platform implies: Chan algorithm could be the best under the ideal criteria of Gauss with zero mean, Combine algorithm could be equal to it under the certain criteria; While it's under the practical criteria, the Combine algorithm which takes the SX output as the initial to be iterated by Taylor series can eliminate the disturbing from multi-path and NLOS so as to get a better performance, to some extent. Furthermore, the tests for comparison between 7 BTS and 3 BTS are also processed, which intend to get a conclusion that the making full use of the BSS network resource can do a great impact to the location precision.The content of wireless location technique covers too much, plus the frequently updated communication technology, hence this article only concerns the tip of the iceberg. In the future, the further study in therms of the optimized algorithm, improvement of simulator and hybrid algorithm are strongly expected.