Optimal Multi-Base-Station Selection for Belt-Area Wireless Sensor Networks

Since underground environments, such as urban subways, tunnels, underground pipe corridors, and mine roadways, etc., are complex and changeable, the software functions of the sensor nodes in underground spaces must be updated regularly to satisfy the requirements of the monitoring center. Software updates cannot be traditionally conducted due to the large number of nodes, the wide distribution range and the poor underground environment; they can only be made using wireless reprogramming. Current reprogramming protocols are all designed for topologically unconstrained networks and are inefficient in routing constrained underground belt-area wireless sensor networks (BAWSNs). We propose a new reprogramming mechanism for reducing the energy and time consumption of the data downstream transmission process in BAWSNs. For network energy optimization, we identified the highest energy efficiency transmission radius. For network base-station location time optimization, an approximate (1 + ε) algorithm that is based on gradient cyclic descent is proposed, which is of complexity O (kn
²
). The simulation results demonstrate that, compared with classical algorithms, the BAWSNs approximation algorithm can locate the optimal base station accurately with low time consumption.