Algorithm 2: CP Estimation on $\mathit{L}\mathit{O}\mathit{S}\mathit{L}\mathbf{\left(}\mathit{i}\mathbf{\right)}$ based on Geometric Formulation.

Initialization: Obtain the coordinates of all wireless sensor nodes, as well as the topology information of all LOSLs, that is, their formulas; roughly know the pedestrian speed $v_{knw}$;

Step 1. Launch the algorithm when $LOSL\left(i+1\right)$ is triggered; Step 2. Find the unique possible segment where the CP may be located as the restricted region; Step 3. Build the matching model in Equation (9) between the ASL and measured distances travelled $v_{knw}·T\left(i\right)$ and $v_{knw}·T\left(i+1\right)$; Step 4. Optimize the model (9) to determine the best-fitting ASL; Step 5. Obtain the intersection of the best-fitting ASL and $LOSL\left(i\right)$ as the estimated CP.