Research On WDAM-UWB System Ranging Localization And Error Mitigation Technology

With the development of Internet, networking and other technologies,the demand for accurate indoor localization is gradually increasing as the rising of the place of people’s activities happening in indoor environment. However, there are dense multipath and non-line of sight and other adverse factors in indoor environment, which bring challenge for accurate indoor positioning. The WDMA-UWB system is not only the ultrawideband advantages of very narrow pulse, high time resolution and low complexity, but also WDMA technology based on the orthogonal wavelet advantages of strong antiinterference ability, high transmission rate, which is very suitable for indoor positioning technology.Firstly, we use a estimation algorithm based on entropy for the TOA estimation algorithm of the ultra-wideband systems with low complexity. By calculating the information entropy of the received signal, the jump-change of entropy is used to distinguish the signal region and the noise region in order to determine the TOA of the first of trails. Simulation results show that the ranging performance is greatly improved compared to the traditional method.Secondly, a mitigation method based on Gaussian process regression with the error mitigation technology is proposed in this dissertation. By receiving signals extracted features as inputs, then send into the Gaussian Process model which has trained the hyperparameters. The ranging error is predicted, and the performance is improved obviously after mitigation.Finally, the main factors affecting the positioning accuracy is analyzed to focus on the indoor localization technology in this dissertation, and the co-location algorithm based on LSE-Taylor and PSO-Taylor is proposed. Based on the simulation of different ranging errors, the effectiveness and practicability of the cooperative localization algorithm are verified.In the WDMA-UWB communication system, the simulation results can achieve the needs of indoor positioning of centimeter level by the indoor residential and office channel model simulation analysis in this dissertation.