Prediction based RSS fingerprinting for positioning and optimization in cellular networks

The location of cellular terminals is of a major interest to both business and government. Some applications do not have access to measurements dedicated for location. Fingerprinting based on received signals strength (RSS), also known as RF pattern matching methods, may provide an appropriate solution. In general drive tests and measurement campaigns are needed to obtain the RF fingerprint. This paper discusses the problem of replacing the measured RF fingerprints with predicted ones, with the advantages of the ability to obtain the required fingerprint for each point in the area of interest with negligible cost. The accuracy of this method is discussed and applications for which this accuracy is sufficient are presented as well.