An Algorithm with Iteration Uncertainty Eliminate Based on Geomagnetic Fingerprint under Mobile Edge Computing for Indoor Localization

. 2022 Nov 22;22(23):9032. doi: 10.3390/s22239032

Algorithm 1 Algorithm.

1:
procedureBuilding
2:
$p a r a m e t e r s \leftarrow initial value$
3:
$m a p \leftarrow g e o m a g n e t i c m a p$
4:
$s t e p D a t a \leftarrow t r a j e c t o r y d a t a$
5:
end procedure
6:
procedureMatching
7:
using geomagnetic threshold
8:
if stepData[current].magnet not in map then return false
9:
end if

return $P_{1} \leftarrow$ possible position area
10:
end procedure
11:
procedureFiltering
12:
Input past trajectory
13:
Output smooth trajectory, prediction
14:
end procedure
15:
procedureMotion model
16:
for step in stepData do
17:
using angle, speed in previous steps, add uncertainty
18:
if prediction positions by inference not in $P_{1}$ then return false
19:
end if

return $P_{2} \leftarrow$ intersection of prediction and $P_{1}$
20:
end for
21:
end procedure
22:
procedureHandling outlier positions and error
23:
Input each trajectories, estimation points
24:
Output trajectory, point by correction
25:
end procedure
26:
procedureEvaluation
27:
Compute euclidean metric between test path and train, using DTW to compare trajectory
28:
Compute euclidean metric of final position
29:
end procedure