A new approach to simulate GPS measurements

The Global Positioning System (GPS) is a satellite based precise positioning system available worldwide. The performance of the basic GPS system has been augmented by the technique of Differential GPS (DGPS) for military as well as civilian uses. Performance evaluation of a DGPS system requires the availability of truth data, GPS measurement data and DGPS corrections as functions of time. In many parts of the world, a lack of base stations and other infrastructure makes it impossible to have the desired quality and quantity of data. Thus, it is useful to develop a simulator which can generate GPS measurements for an arbitrary number of truth points. A common method to estimate errors associated with GPS is to identify various individual sources of S error and develop linear models to estimate these values. This paper presents an artificial neural network which is designed and trained to estimate measurements of a particular GPS receiver. The simulated data is required to satisfy some principal characteristics associated with actual measurements for the same system in a statistical sense. For example, the mean, standard deviation as well as minimum and maximum error values are closely correlated between the observed performance and simulated data. The goodness of fit is established by the use of appropriate methodology based on hypotheses testing. Thus, the approach presented can be viewed as the development of a highly nonlinear model to estimate GPS errors as opposed to the erstwhile linear models.