论文部分内容阅读
针对无线多传感器网络数据在进行聚类时,一直存在聚类效率低。提出基于定量递归调度的无线多传感器网络数据聚类算法。建立无线多传感器网络数据信息流模型,将数据分类中分类误码率映射成一组概率密度函数,通过定量递归分析生成特定随机时间序列,计算多传感器网络数据定量递归特征,把特征函数分配到每个分类频点,通过定量递归特征分类调度,实现对网络数据聚类算法的优化。仿真结果表明,采用该方法能提高多传感网络数据聚类性能,平均误分率较低,优于传统算法。
For clustering in wireless multisensor network data, clustering efficiency is always low. A data clustering algorithm based on quantitative recursive scheduling for wireless multisensor networks is proposed. A data flow model of wireless multi-sensor network is established. The classification error rate in the data classification is mapped into a set of probability density functions, and a specific random time series is generated by quantitative recursive analysis to calculate the quantitative recursive features of the multi-sensor network data. A classification frequency, through quantitative recursive feature classification and scheduling, to achieve the network data clustering algorithm optimization. The simulation results show that this method can improve the data clustering performance of multi-sensor networks, and the average error rate is lower, which is better than the traditional algorithm.