WIFI FINGERPRINTING INDOOR POSITIONING WITH MULTIPLE ACCESS-POINTS IN A SINGLE BASE STATION USING PROBABILISTIC METHOD
WiFi fingerprinting for indoor positioning is well known as one of the most efficient techniques to estimate indoor target location instead of other existing methods which are triangulation and proximity. This method requires a training phase to collect Receiving Signal Strength (RSS) samples that w...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2017
|
| Subjects: | |
| Online Access: | http://utpedia.utp.edu.my/22073/1/UTP%20MSc%20in%20EE%20Thesis%20JULY%202017%20-%20G03140%20-%20Muhammad%20Al%20Amin%20Amali%20Bin%20Mazlan.pdf http://utpedia.utp.edu.my/22073/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | WiFi fingerprinting for indoor positioning is well known as one of the most efficient techniques to estimate indoor target location instead of other existing methods which are triangulation and proximity. This method requires a training phase to collect Receiving Signal Strength (RSS) samples that will be utilized in location estimation phase using matching algorithms. Recently, most commercially indoor positioning solutions used on smartphone utilizes the current building infrastructure to estimate personnel location where wireless router is used as an Access Point (AP) in each base station to extract the RSS values. However, for buildings with inadequate infrastructure setup, implementing multiple base stations using a single AP in each base station would require an exhaustive resources of manpower and time especially for a small scale positioning setup. There are also not enough distinct RSS values at each location covered by a single base station. Thus, WiFi fingerprinting using multiple APs with omnidirectional and directional antennas in a single base station employing a probabilistic approach has been proposed to minimize the infrastructure setup. Based on experimental results, the proposed multiple APs in a single base station was found to reduce the number of base stations required to achieve the same or better accuracy as existing approach using the same number of APs. The feasibility of multiple APs in a single base station was demonstrated using kernel estimation with the results indicating the ability of the proposed work with an accuracy of 1.82 m to outperform the existing work by minimizing 22% of Root Mean Square Error (RMSE). |
|---|