Research Of LTE-signal-based Channel Modeling And Positioning Algorithm In Post-disaster Environment

Natural disasters always affect people's life safety,how to quickly complete the search and rescue work in the post-disaster is the first priority.The positioning technology based on mobile signal has a good development prospect in the field of post-disaster search and rescue because of its well-developed infrastructure and the ability to quickly deploy base stations.Mobile phone carry-on rate and LTE mobile phone usage rate increase year by year.Compared with GSM and CDMA,LTE has a larger bandwidth and uses MIMO technology to improve the accuracy of time delay estimation.Using OFDM technology to improve the ability of anti-multipath and inter-symbol interference,so it can be used as a post-disaster search and rescue resource.There are few studies on the channel modeling and positioning technology of LTE signals in the post-disaster environment,while the channel model of LTE frequency band in the post-disaster environment needs to be measured on the spot,and the positioning accuracy is seriously affected by the non-line-of-sight environment after the disaster.Therefore,this paper modeled and measured the channel model of LTE frequency band and the research results are as follows:1)The channel modeling of the post-disaster environment is studied and a general channel fading model is proposed.The large-scale fading model and the multi-path channel impulse response model in small-scale fading are modeled,and the channel parameters are estimated and fitted using spatial-smoothing based ESPRIT algorithm.Draw the conclusion that the normalized multi-path delay conforms to the exponential distribution and its amplitude conforms to the gaussian distribution.It is concluded that spatial-smoothing based ESPRIT algorithm has higher evaluation on model order and normalized root mean square error than AR model method and ESPRIT parameter estimation method.The multi-path impulse response model for LTE signal frequency band was established based on two kinds of actual post-disaster environments.In addition,the obtained mean excess delay,the root-mean-square delay spread and kurtosis can be used for non-line-of-sight identification,and reduce the influence of non-line-of-sight on the measured value and play an important role in accurate location calculation.2)A TDOA accurate estimation algorithm is proposed to reduce the influence of non-line-of-sight,and its performance is verified by simulation.A positioning algorithm based on non line-of-sight error identification and mitigation is proposed,which uses the likelihood ratio test principle.Kurtosis,mean excess delay and root-mean-square delay spread parameter are used to identify the non-line-of-sight.Then the TDOA accurate estimation algorithm is used to obtain the TDOA measurements for locating in EFIR filtering algorithm.Compared with the weighted least squares method and the EKF algorithm,proposed algorithm achieves better positioning accuracy at cost of calculation time.In this paper,the channel of two kinds of actual post-disaster collapse environments is modeled based on LTE signal frequency bands.The error of the large-scale path loss model is within 2dB,and the evaluation criterions of small-scale multi-path impulse response model,including the mean excess delay,the root-mean-square delay spread and kurtosis,are less than 9%between estimation model and measured data.So it can be used to simulate the channel under two collapse situations.The proposed algorithm is simulated using data from two collapse environments.Compared with other algorithms,the influence of non-line-of-sight error is reduced at the cost of N times calculation time,and the location error is reduced by about 20%-30%in two post-disaster scenarios.