Indoor PDR Positioning Assisted by Acoustic Source Localization, and Pedestrian Movement Behavior Recognition, Using a Dual-Microphone Smartphone

Algorithm 1

Algorithm for detecting the movement transition zone based on D-S evidence theory and SVM.

Input: The movement state of the pedestrian’s current gait;

Output: Update the movement state of the first 5 steps of the pedestrian’s current gait.

1: Determine the current movement state

if movement state = upstairs or downstairs

 1.1: for i =1 to 5 do:

 1.2: Fusion of all categories:

  1.2.1: Set A=Walk, B=Upstairs, C=Downstairs;

  1.2.2: for n =1 to 3 do:

   1.2.2.1:if n =1, X = A;

else if n =2, X = B;

    else if n =3, X = C;

    end if

   1.2.2.2: Obtain ;

   1.2.2.3: Obtain ,;

   1.2.3.4: Obtain ;

  1.2.3: end

 1.3: Fusion between categories:

  1.3.1: Obtain the probability ,,

  1.4: Decision output.

  1.5: end

end if

2: Update the first 5 steps of the movement state