A Novel Indoor Localization Algorithm for Efficient Mobility Management in Wireless Networks

Algorithm 1

The algorithm of MFF.

Input: The RSS and CSI values of the test point. The coordinates, RSS and CSI values of the

calibration points.

Output: The coordinates of the test point

According to the coordinate, RSS and CSI values of the calibration points to build RSS, He

and Hp fingerprint database. Step 1

In the case of timeliness of positioning results, according to RSS and CSI values collected at

the test point and the fingerprint databases built in step 1, Fusion1 feeds out the RSS and He

to Horus and FIFS, then calculates the candidate position for each approach. The position of the

test point is obtained by weighted fusion of the two candidate results and the coordinates

of the test point is returned. Step 2

In the case of high positioning accuracy scenario, Fusion2 uses KL distance as the similarity

metric and calculate candidate calibration point set under D-CSI. Merge the candidate

reference points which are obtained by Horus, FIFS and D-CSI approaches and calculate the

degree of each candidate calibration point. Step 3

Depending on the size of each calibration point degree obtained in step 3, Fusion2 optimizes

the final location of the test point and returns the coordinates of the test point. Step 4