用于WSNs的间歇型参数估计算法

Intermittent LMS parameter estimation algorithm for WSNs

无线传感器网络(WSNs)用于目标定位与追踪、频谱感知、自动雷达、导航及机器视觉等领域时,常常需要节点协同估计同一个未知参数。节点用一组输入观测未知参数,而未知参数会受到噪声干扰。本文提出了一种间隙式参数估计算法,采用分布式自适应扩散式最小均方(LMS),节点无需直接与网关通信,仅与邻居节点交换自身的估计信息,与集中式算法相比,降低了负载。然后,进一步引入间歇参数,仅在部分时刻交换估计信息,在损失少许性能的情况下大大降低通信负载。

When wireless sensor networks(WSNs)are used in target positioning and tracking,spectrum sensing,automatic radar,navigation,and machine vision are often required nodes to collaboratively estimate the same unknown parameter.The node observes unknown parameters with a set of inputs,and the unknown parameters are disturbed by noise.A gap least mean square(LMS)parameter estimation algorithm is proposed.This algorithm uses distributed adaptive diffusion LMS.Nodes do not need to communicate directly with the gateway,and only exchange their own estimation information with neighboring nodes.Compared with the centralized algorithm,the communication load is reduced.Then,the intermittent parameters are further illustrated,and the estimated information is only exchanged at some times,which greatly reduced the communication load with a little loss of performance.