A new approach for personal safety forewarning based on GPS positioning and spatiotemporal analysis

In this paper, a new personal safety forewarning approach (PSFA) using GPS positioning and spatial analysis is presented. Specifically, we discussed the spatial topological relationships between the spatiotemporal objects related to personal safety. On that base, safety state is assessed through the calculation of a state matrix and then how to affirm the factors related to forewarning response is explored. Experimental results of buffer analysis show that sufficient accuracy (more than 95%) can be obtained when the target is 20 meters or farther away from the border of the test regions (even with poor GPS reception). A compromise is also adopted for the detected trajectory points near the border to decrease the error rate. Thus the practicability of the proposed method is validated.