Analysis And Optimization Of Pseudolite Location Accuracy Under Indoor Multipath Environment

In recent years, people put forward more and more demands to location service. Outdoor navigation and location system embodies limitations gradually, and people focus their attentions on the booming indoor location technology. Indoor location involves extensive scenarios, and its research results are closely related to people’s life and work. The adverse factor indoor location needs to overcome is the complex transmission environment of wireless channel. Multipath effect, non-line of sight and shadow fading all severely restrict the reliability and stability of indoor location system. Improving system’s hardware structure and optimizing system’s software algorithms become a research orientation of indoor location technology.Pseudo satellite indoor location system is used as research platform in this paper, and location scenario is indoor non-line of sight. Firstly of all, system working parameters are set up, then indoor scenario is built and expectation is proposed. Secondly, ranging and location algorithm under indoor NLOS are designed, and algorithm performance is optimized by measured channel data. Finally, through experiment simulation, system’s reliability and stability are tested, and system’s improvement scheme is put forward.Channel measurement under typical indoor LOS and NLOS scenario is launched. Path loss index, shadow attenuation factor and root-mean-squared delay spread are calculated by measured data. Characteristic parameters of NLOS scenario are compared with LOS scenario. As a result, channel propagation characteristics of indoor environment is comprehensively grasped. Groundwork of designing and optimizing ranging and location algorithms is well done by channel measurement.A TOA estimation method under indoor NLOS scenario is proposed. Firstly sliding correlation functions of received signals in NLOS are acquired by spread spectrum-correlation technology. Then first path searching area is determined by setting threshold. At last peak first path belongs to is found and TOA is estimated. TOA-based ranging error is modeled, and then Laplace and Gauss model are proposed. After that, distribution of ranging errors is fitted and mathematical expectation and variance are calculated through statistics. Using cumulative distribution function, fitting performance of the two models is evaluated.Based on maximum likelihood estimate, single and multiple antenna location algorithms are designed. Performance of location system is simulated and analyzed. Single antenna location analyzes the influences on accuracy caused by pseudo satellite’s number and layout, and multiple antenna location analyzes the influences caused by antenna’s number in each pseudo satellite and correlation. Location area is a 200 meter by 200 meter square. Target’s location in the area is determined by location system with different parameters. Simulation results show that, in order to achieve 3 meter location accuracy under typical indoor NLOS scenario, single antenna location system requires at least 12 pseudo satellites, and multiple antenna location system just requires 3 pseudo satellites with 5 antenna in each pseudo satellite.