基于WSN的两种气体源定位算法研究

Study of the two plume source localization algorithms based on WSN

本文基于气体污染源浓度衰减模型,首次采用极大似然预估算法、直接三边测距算法对气体污染源定位进行了对比研究。仿真实验对比了两种算法在不同数量传感器节点以及背景噪声情况下对预估定位误差的影响。仿真表明在环境背景噪声较大的情况下,采用极大似然算法比采用直接三边测距算法有着更强的鲁棒性;而直接三边测距算法简单,在背景噪声较小时,定位效果同样有效。

Based on the attenuation model of plume source, this paper studies the location of the plume source using maximum likelihood estimation （MLE） algorithm and direct trilateration （DT） algorithm. The effects of these two algorithms on the location estimation errors are researched through simulation with different sensor number and different background noise conditions. Simulation result shows that compared with DT algorithm, MLE algorithm is more robust to stronger background noise, and can get better accuracy in plume source localization. DT algorithm is simple and effective when the noise is less.