一种基于GM(1,1)的分簇WSNs拥塞检测与控制方案

针对分簇结构WSNs下中继簇首流量负担过重易引发局部拥塞的问题,提出了一种基于簇首数据吞吐量短期预测的拥塞检测与控制算法CMETR.该算法通过建立GM(1,1)灰色模型分析流经各簇首的当前流量,预测簇首未来的拥塞程度,并以调整簇内节点数据采集频率的方式减小簇首的数据传输压力,从而达到控制网络拥塞的目的.仿真结果表明:该算法有较好的预测精度,对即将到来的网络拥塞能够进行提前处置,且在网络繁忙时能够缓解链路压力,相对CODA算法有更好的稳定性和能耗特性.

A congestion detection and control algorithm for clustering WSNs based on GM (1,1)

Aiming at the problem of local congestion caused by relay cluster heads with heavy forwarding workload in clustering WSNs,a congestion detection and control algorithm(CMETR)based on the short-term prediction of data throughput for a cluster head is proposed.In this algorithm,a GM(1,1)grey model is established to analyze the current network traffic of cluster heads and predict the future congestion degree of cluster heads. In order to control the network congestion,the algorithm reduces the transmission pressure of the cluster head by adjusting the data sampling frequency of the nodes in the cluster. The simulation results show that the algorithm has good prediction accuracy,can handle the incoming network congestion in advance,and relieves the link stress when the network is busy;moreover,it has better stability and lower energy consumption compared with the CODA algorithm.