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.