中文摘要：

纸建议过滤机制(PMF ) 解决浪费的精力在无线传感器网络(WSN ) 由时间冗余的数据引起了的传播的问题的 prediction-mode-based，根据在数据收集的采样系列上的时间空间的关联的特征。优先的工作建议了几条途径到精力经由数据聚集树结构在数据传播过程期间花费了的减少。在上面的研究与那些方法区分开来，我们的建议计划主要集中于在事件来源减少时间的冗余的度经由自我适应的过滤完成节省精力的效果结构。为精力有效的收集的 PMF 的框架为采矿由预言模块组成时间领域的变化法律，为更新模型，并且为控制数据过滤驾驶模块的自我学习的模块操作。与驾驶散布统治的统治和阀值的错误的设计结合了，它是在上述过滤机制的中间件，在网络的传播负担的数量能极大地在服务(QoS ) 的质量的前提上被禁止保证和精力消费能因而被减少。最后，试验性的结果证明 PMF 的表演能显著地在节省精力的效果和自我适应性上超过一些古典数据收集算法。

英文摘要：

The paper proposes a prediction-mode-based filtering mechanism（PMF） to solve the problems of transmission energy wasting caused by time-redundant data in wireless sensor networks（WSN）,according to the characteristic of spatio-temporal correlations on sampling series in data-collection.Prior works have suggested several approaches to decrease energy cost during data transmission process via data aggregation tree structure.Distinguish from those methods in above researches,our proposed scheme mainly focus on reducing the temporal redundant degree in event-source to achieve energy-saving effect via self-adaptive filtering structure.The framework of PMF for energy-efficient collection is composed of prediction module for mining the change law of time domain,self-learning module for updating model,and driving module for controlling data filtering operation.Combined with the design of error driving rule and threshold distributing rule,which is the middleware in the above filtering mechanism,the quantity of transmission load in networks can be greatly inhibited on the premise of quality of service（QoS） assurance and energy consumption can be reduced consequently.Finally,the experimental results show that the performance of PMF can significantly outperform some classical data-collection algorithms on energy-saving effect and self-adaptability.