Fingerprint location method based on CSI (Channel State Information) and time domain fusion algorithm

Abstract

The invention discloses a fingerprint location method based on CSI (Channel State Information) and a time domain fusion algorithm. In an offline phase, two-step training is carried out on a combined network according to amplitude information of the CSI obtained by each reference point, wherein the combined network is composed of a slot fingerprint-based location network (SLN) and a time domain position fusion network (FN); and in an online testing phase, real-time collected CSI is input into a trained network, and predicted positions are obtained. According to the method, the CSI of an LTE network is taken as a fingerprint with relatively fine granularity; time fluctuation and correlation of the CSI fingerprint are captured through utilization of a two-phase processing time domain fusion algorithm based on deep learning. The time correlation of the CSI fingerprint is taken into consideration in a location system. The method has relatively high precision, relatively good system robustness and relatively rapid online testing speed.

Description

technical field [0001] The present invention relates to a technology in the field of wireless communication, in particular to a fingerprint positioning method based on channel state information (Channel State Information, CSI) and time domain fusion algorithm under LTE cellular network. Background technique [0002] Existing global satellite navigation systems (GNSS) cannot provide accurate positioning when satellite signals are blocked. Therefore, the existing technologies for precise positioning of indoor and outdoor mobile devices through the high coverage of mobile base stations, such as: location fingerprint positioning method, through the offline "map" generation stage and the online prediction stage, use the multipath effect to overcome the non- The disadvantage of poor positioning accuracy under line-of-sight conditions. [0003] For the online prediction stage in the position fingerprint positioning technology, the existing technology uses RSSI as the fingerprint, ...

AI Technical Summary

Problems solved by technology

However, the defect of the above-mentioned prior art is that the fingerprint information is all one-dimensional data, and it is difficult to characterize the specific changes of the channel, such as the fluctuation of the channel state fingerprint, etc., resulting in a large difference between the maximum positioning error and the average positioning error, and the corresponding The online positioning algorithm does not have the ability to automatically extract features and cannot meet the growing demand for positioning accuracy

[0004] After searching the existing technology, it is found that the Chinese Patent Document No. CN108882151A discloses an indoor positioning method based on the regionalization of CSI information. By pre-collecting the CSI data of each area in the room as an offline fingerprint library for deep neural network training, The trained deep neural network is used to identify the CSI data to be tested, and the user location test method based on the probability vector is used to achieve accurate indoor positioning. However, the positioning result of this technology is based on the CSI value measured once, and the natural time-varying characteristics of the channel are not considered. The resulting CSI value fluctuations lead to poor system robustness and the maximum positioning error of this technology reaches 6m, which cannot meet certain scenarios that require high system robustness, such as underground parking garages, etc.

Images