Font Size: a A A

Research Of Indoor Localization Technology Based On CSI

Posted on:2017-08-24Degree:MasterType:Thesis
Country:ChinaCandidate:K W XueFull Text:PDF
GTID:2348330518495644Subject:Software engineering
Abstract/Summary:PDF Full Text Request
In recent years,with the rapid development of wireless hardware and intelligent terminals,growing demand for location-based services,indoor localization technology has become a research hotspot.Indoor localization technology based on Wi-Fi stand Oout because of its deployment convenience and low cost advantages.Currently existing indoor localization technology based on Wi-Fi mostly generate fingerprints by measuiring the intensity of received signal at the receiving end to complete localization.Due to the RSSI signal degeneration and the influence of the multipath effect,there is some error in final location results.In this paper,we refer to the relevant literature and finish multiple experiments,take channel state information as a new unit of measure.We use a CSI indoor localization algorithm based on difference between sub-channels.Compared with the traditional location method based on RSSI,the location accuracy has great improevement.Firstly,we analysis the signal characteristics of CSI from the time domain stability,the spatial difference and the influence of human on CSI these three aspects.The feasibility and difficulty of CSI based localization is explained.Secondly,according to the signal characteristics of CSI,the sub-channel difference method is used to generate the location fingerprint.Based on traditional fingerprint matching algorithm,we obtained posterior probability through the Bayes rule and KMWNN algorithm to determine the final location result.NWe compare two kinds of difference algorithm,one is based on adjacent sub-channel,another is based on specific distance sub-channel,compare following three online matching algorithm:one sample matching,diversity of the samples and get the mean after the match,first match then get mean of diversity of samples.At last,we evaluate the algorithm from the number of transmit antennas,the value of K in KWNN algorithm,time-variability and algorithm complexity.Finally,according to the algorithm described in this paper,we verify it in the actual scene,the experiment result show that:the CSI localization algorithm base on adjacent sub-channel is better than RSSI localization algorithm,imporving the exactness rate of localization.
Keywords/Search Tags:indoor localization, fingerprint matching, channel difference, CSI
PDF Full Text Request
Related items