Trajectory Approximation for Resource Constrained Mobile Sensor Networks

摘要: Low-power compact sensor nodes are being increasingly used to collect trajectory data from moving objects such as wildlife. The size of this can easily overwhelm the storage available on these nodes. Moreover, transmission extensive over wireless channel may prove be difficult. memory and energy constraints platforms underscores need for lightweight online compression albeit without seriously affecting accuracy mobility data. In paper, we present a novel Polygon Based Approximation (PBA) algorithm that uses regular polygons, which is determined by allowed spatial error, smallest unit approximating raw GPS samples. PBA only stores first sample reference. Each subsequent point approximated centre polygon containing point. Furthermore, coding scheme proposed encodes relative position (distance direction) each with respect preceding in trajectory. resulting thus series bit codes, have pair-wise dependencies at reference It possible reconstruct an approximation original decoding chain codes starting Encoding single O (1) operation, overall complexity (n). requires two samples any given time. low small footprint make it particularly attractive low-power evaluated using traces capture actual flying foxes wild. Our results demonstrate achieve up nine-fold savings compared Douglas-Peucker line simplification heuristic. While context devices, equally useful other GPS-enabled devices smartphones car navigation units.