Font Size: a A A

The Optimization Technology Research On Location Fingerprint Positioning Based On WLAN

Posted on:2019-04-10Degree:MasterType:Thesis
Country:ChinaCandidate:F X HouFull Text:PDF
GTID:2428330548967861Subject:Electronic and communication engineering
Abstract/Summary:PDF Full Text Request
With the advent of the information age,location-based services have played an increasingly important role in people's daily lives.GPS positioning technology has become the world's most influential,the most widely used positioning technology because of covering a wide area,high accuracy,high efficiency and other unique advantages.It has been maturely used in many positioning and navigation services of areas,and achieved a good economy benefit and social benefit.However,in the indoor environment,the GPS signals received by the indoor devices are extremely weak due to the obstructions of the buildings,so GPS can not position accurately.The demand on location monitoring and tracking of the indoor objects becomes more and more intense because of people's daily production and living are mostly carried out indoors.Therefore,the research of indoor positioning is very important.Compared with other typical indoor positioning technologies,the location fingerprint positioning technology based on wireless local area network(WLAN)does not require additional hardware supporting and has the advantages of low positioning cost and high positioning accuracy,it has become a hot spots in the indoor positioning research field.Especially in recent years,the popularization of wireless networks and smart phones make WLAN location fingerprinting indoor positioning technology has a broader market prospects.Although WLAN location fingerprinting technology has obvious advantages,there are still two key issues: there is more fingerprint data in the offline phase,which results in a large amount of computation in the online phase,it reduces the positioning speed.Second,the WLAN signal is unstable,leading to large differences between the real-time fingerprint data and the fingerprint data in the database,it reduces the positioning accuracy.Focusing on the two issues,this paper optimizes the traditional WLAN location fingerprinting positioning technology,the specific work is as follows:Firstly,The WLAN location fingerprinting positioning optimization technology based on improved K-means algorithm is proposed.Firstly,the K-means algorithm is improved by RSS euclidean distance and standard deviation of reference points.Then the original K-means algorithm is used to cluster the original position fingerprint database,and a huge original database is reasonably divided into K sub-databases,it narrows search space for fingerprint database in online phase.It can improve the positioning speed significantly while ensuring the positioning accuracy.Secondly,The WLAN location fingerprinting positioning optimization technology based on the combination of access point(AP)selection and improved K-means algorithm is proposed.The APs involved in positioning are properly filtered according to the RSS threshold and stability,it can remove the unstable APs and improve the positioning accuracy.Considering both the positioning accuracy and the positioning speed into consideration,the new AP selection method and the improved K-means algorithm are combined to optimize the the traditional WLAN location fingerprinting positioning technology.It can make up the insufficient that K-means algorithm can not improve positioning accuracy obviously and new AP selection method can not improve positioning speed obviously.It achieves the full optimization of traditional positioning,this optimization technology can improve positioning accuracy and positioning speed of WLAN location fingerprinting technology effectively.Finally,the positioning performance of the traditional positioning technology and several optimization techniques proposed in this paper are analyzed and compared through relevant experiments.It can prove the effectiveness of positioning optimization technology.
Keywords/Search Tags:Indoor positioning, WLAN, Location fingerprint, K-means algorithm, AP selection
PDF Full Text Request
Related items