科研成果详情

In WSNs (Wireless Sensor Networks), data storage and retrieval is a challenging problem because of the limited resource and the short communication radius of the sensor nodes. Most of the existed schemes choose one or more static sensor nodes or Sinks to act as the rendezvous nodes, which can be seen as the connectors between the data producers and the data consumers. However, those schemes cannot avoid both the “hot spot” problem and the “bottleneck” problem, which refer to the much higher load balance of the sensor nodes around the rendezvous nodes. Moreover, most of the existing schemes never consider the dynamic nature of WSNs, which leads to the lack of adaptability. In this paper, we propose a novel dynamic-optimization-based framework named SRMSN using mobile Sinks to solve such a problem. SRMSN utilizes two heuristic methods, which are based on the virtual-grid-division technology and the diversity-factor-analysis technology, to determine the optimal target locations of the mobile Sinks in each time interval when each Sink node stay at a certain position and the optimal length of each of the time intervals adaptively, aiming at improving the adaptability and the efficiency of WSNs on data storage and retrieval. Simulation results show that SRMSN can reduce and balance the energy consumption greatly as well as decrease the average delay of data storage and retrieval in comparison with the state-of-the-art scheme on data storage and retrieval in WSNs.