Improving Indoor Positioning Accuracy through a Wi-Fi Handover Algorithm

The increasing numbers of applications on mobile device application stores are fuelling the need for an indoor positioning solution that addresses the user's needs for enhanced experience with his/her mobile. Most current solutions depend on a Wi-Fi receiver to scan beacons broadcast from well mapped Access Points. Independent of the technology/algorithm used to map these access points, the end user will need to know his/her estimated position as accurately as possible. In order to get an accurate position, two different processes need to be launched accurately. The first process will aim to map and build an accurate data-base for as many as possible of the access points that could be found in the testing area. The second process will involve using and exploiting, these previously stored data to estimate a user's position. In this paper we introduce a new algorithm that will improve the accuracy of an estimated position by improving the above second process's accuracy. So if we have an accurate map of access points for a test area, the user will use his/her Wi-Fi receiver to scan surrounding access points and calculate his/her position through triangulating between known access point positions. The proposed "Wi-Fi hand over" positioning algorithm is a multi step triangulation scheme that depends on movement between mapped access points. The algorithm is based on continuous position requests from moving users. While the user movines through a test area, Wi-Fi signal strength will vary from point to point. In our algorithm we have used a new measurement parameter we term HoR (Hands over Ratio). This parameter will determine the corner of the Wi-Fi range a user is currently based. In addition to pure signal strength parameters, and/or any other parameters used to estimate distance between the user and the access point, HoR will improve accuracy in situations where less than 4 known access points are found. Even in cases where there is no distinct improvement in accuracy when more than 4 known access points are found (dense Wi-Fi population areas), HoR will still be useful in avoiding the duplication of overlapped access points.