| Because of China’s coal mine poor conditions and the weak safety awareness of managers,the lack of underground personnel vehicle positioning information will bring a lot of inconvenience and hidden dangers.Therefore,the need for coal mine personnel positioning is becoming increasingly significant.The company Ⅰ worked in as an intern focus on downhole positioning system(which belongs to the scope of indoor positioning).This system aims at positioning people,vehicles and other objects in mine high-precisely and real-timely.It provides a basis for daily mine management,people/vehicle scheduling and emergency/disaster management.The wireless positioning technology TDOA(Time Difference of Arrival)is used in this system.There are some existing problems in this system to be solved.Firstly,the base station group used by TDOA needs to select a main base station whose local time is the global time of base station group,but the algorithm for automatically determination of the main base station needs to be improved.Secondly,linear interpolation is the only method used for time deviation estimation.Other methods also need to be tested.Thirdly,the various limitations of terrain and measuring tools make it difficult to estimate the error of time offset.In order to solve the above problems,using matlab to implement the simulation system for positioning in mine and comparing different algorithms for time offset estimation becomes important.The simulation system can also improve the algorithm for automatically determination of the main base station.The thesis describes the system from the needs of the analysis to the test process.First,analyze the functional requirements of the simulation system,basing on the use cases from the perspective of the users.Second,the master base station selection module,the master and slave time synchronization module,the time offset estimation module and the positioning calculation module are designed and implemented.Finally,the four modules were tested.The simulation system functions well.It has passed the test and it can analyze the error of the time offset estimation.The time offset estimation module adds Kalman filter and closed-loop feedback control estimation method to solve the problem that the original system estimates the time offset in a single way.The master base station selection module improves the algorithm that naturally selects the main base station without manual assignment by combining PageRank. |