On Positioning Techniques Based On Wirelss Network

Recently,location based services bring a tremendous convenience to our daily life.The development of the services also makes new demands on wireless positioning technologies such as higher precision and ubiquitous support.At present,the major and popular wireless positioning methods are based on satellite constellation but facing degraded accuracy in urban areas and indoors because of poor coverage.Positioning methods based on wireless network(e.g.cellular and WLAN)are widely mentioned due to strong coverage in urban districts and easy deployment.There are two wireless-network-based methods in common use:time difference of arrival and location fingerprinting.This thesis focused on some problems in wireless network based positioning methods as follows:1.The thesis proposed a novel location fingerprint called TDOPA and a corresponding location algorithm.The fingerprint utilizes location characteristics in multipath propagation.On the foundation of a model with Gaussian noise,the CRB of the fingerprint is analyzed and a location algorithm is derived.The analysis of the performance of the algorithm is conducted using correct estimation probability.The thesis hopes the analysis results above are helpful in relative experiments and algorithm improvement.2.The thesis studied the problem of the estimation of flight delay for PRS(Positioning Reference Signal)dedicated for ranging in LTE downlink positioning method,a standardized time-difference-of-arrival method.The study handles the problem on the basis of channel matrix in frequency domain and needs to estimate the number and the delay of paths simultaneously.An MDL-MUSIC method was proposed,including two implements:a two-step method and a traversal method.An SIC-DFC method was proposed as well.According to link-level simulation results,the MDL-MUSIC method can provide a more accurate estimation under high SNR(Signal Noise Ratio)and the SIC-DFC method has a benefit of lower complexity.The thesis hopes the results can contribute to the accuracy improvement of final location estimation in the downlink method.