Outdoor 2.5D-Terrain IoT Deployment Algorithms: Repairing Partitions While Mitigating Coverage Loss

We present and evaluate new hybrid recovery algorithms for outdoor IoT networks. Rather than using the 2D disk model, the algorithms use 2.5D terrain data to compute intersections of commsheds that identify potential positions for recovery nodes, and they compute unions of viewsheds to prioritize potential recovery positions. The algorithms are evaluated using actual terrain data. Results show that on average, TARCI-GP achieves post-recovery coverage that approaches that of a similar algorithm that incorporates an exhaustive step, while using two orders of magnitude less CPU time during pre-failure planning. Also, TARCI-GP achieves better post-recovery coverage compared to a proxy for 2D recovery algorithms. And finally, since the TARCI variants consider coverage when choosing recovery positions, we discovered that on average, TARCI variants that consider more distant recovery positions during pre-failure planning tend to achieve better post-recovery coverage.