一种面向区域数据的机会群智感知数据分发策略

面向时空敏感的局部数据，结合区域数据采集的背景，设计了一个基于地图分区的机会式群智感知数据分发策略. 该策略分为四部分:首先对节点进行区域的划分和周期性地采集数据；其次在节点相遇时不同位置属性的节点之间进行边缘节点的判断和位置更新；然后进行时空敏感的区域数据之间的共享和传输过程；最后当缓存区满或缓存数据超过有效时间时进行缓存更新和丢弃过程. 该数据分发策略可以实现针对局部区域数据进行数据采集的目标，同时具有很好的边缘检测控制功能. 仿真实验结果表明:基于地图分区的数据分发算法具有较好的数据采集率与较低的网络开销，可以在性能接近Epidemic算法的前提下，提供可靠的区域数据采集和数据共享功能，且某些条件下性能可以超越Epidemic算法.

An Opportunistic Data Dissemination Strategy of Crowd Sensing based on Local Data

An opportunistic data dissemination strategy based on map division is designed in this paper. The data dissemination strategy is combined with the temporal and spatial sensitivity features of effective local data and suitable for the scene of regional data acquisition. The strategy is divided into four parts. Firstly, nodes with different regional attributes collect data periodically, then, the judgment of edge nodes and position update of nodes are executed between nodes with different location attributes when nodes encounter each other, also, transmission of regional data with space-time sensitivity is completed, finally, cache updates and discarding are implemented when the buffer memory is full or the data stored in the cache is invalid. The data dissemination strategy can achieve local data collection and has a great edge detection function. The simulation results show that the data dissemination algorithm based on map partition has good data acquisition rate and low network overhead. It can accomplish reliable regional data collection and data sharing under the premise that the performance of our proposed algorithm is close to that of Epidemic algorithm. In some conditions, the performance of proposed algorithm even can exceed that of Epidemic algorithm.