Hadoop平台下基于优化X-means算法的大数据聚类研究

针对WSNs中现有路由算法存在的各种不足,本文提出了一种基于占空比间隔优化的延迟约束路由算法。具体来说,在提出的算法中,首先,将端到端延迟分布估计为占空比间隔和潜在转发器数量的函数,在给定的网络模型和参数下,其分布可以近似地估计。然后,选择满足延迟约束成功率(Delay-Constrained Success Ratio,DCSR)要求的占空比间隔最大值,每个节点独立地调度其休眠和唤醒时间,发送端节点将数据包转发给潜在转发器中最先唤醒的节点,从而确保数据包以要求的概率即DCSR到达接收器,同时最大化占空比间隔；仿真实验结果表明,提出的路由算法不仅能够满足要求的DSCR,并在ETE延迟、数据包交付率和实际得到的DCSR方面都优于现有的先进算法。

Research on Large Cluster Analysis Based on Optimized X-means Algorithm under Hadoop Platform

Aiming at all kinds of the shortcomings of existing routing algorithms in WSNs, this paper proposes a delay-constrained routing algorithm based on duty cycle interval optimization is proposed in this paper. Specifically, in the proposed algorithm, firstly, the end-to-end(ETE) delay distribution is estimated as a function of the duty cycle interval and the number of potential forwarders, and its distribution can be approximately estimated under the given network model and parameters. Then, the maximum duty cycle interval that met the delay-constrained Success Ratio (DCSR) requirement is selected. Each node independently schedules its sleep and wake up times, and the sending node forwards the data packet to the node that is woke up first in the potential forwarders. This ensures that the packets arrive at the receiver with the required probability, namely DCSR, and maximizes the duty cycle interval. Simulation results show that the proposed routing algorithm can not only meet the requirements of DSCR, but also outperform the existing advanced algorithms in terms of ETE delay, packet delivery rate and the obtained actually DCSR.