Landslide Displacement Monitoring And Positioning Based On GPS And WSNs

In our country,landslide disaster has always occupied the first place in national geological disasters,and seriously threatened the safety of people's lives and properties.In order to reduce the impact of landslide disaster,it is necessary to carry out continuous and long-term monitoring for landslide.There are many kinds of landslide monitoring technologies,which have been gradually developed from traditional manual measurement to high-precision and automatic instrument measurement.Especially,with the development of GPS positioning technology,the GPS receivers can continuously observe the landslide monitoring points all the time.The movement of a monitoring point is reflected by the distance between the initial position and the new position of the monitoring point,and such a method can achieve the purpose of landslide monitoring.However,high-precision GPS receivers are expensive and their deployment in a large-scale landslide monitoring area is costly.Therefore,using low-precision GPS receivers for landslide monitoring is a problem to be studied.In addition,how to transmit the data collected by GPS receivers to a remote server is also a problem to be solved.Traditionally,wired transmissions,such as wires and cables,are used for data transmission.However,many landslide monitoring areas may be geographically remote and their environment could be harsh.It is difficult to implement data transmission in a wired manner,thus wireless transmission is required and using wireless sensor networks(WSNs)to transmitting data is a good choice.WSNs have the advantages of dynamic networking,low power consumption,distributed and low cost.Therefore,it is of great practical significance to combine GPS receivers with WSNs for landslide monitoring.This thesis focuses on the landslide displacement monitoring and positioning,and studies the following two main contents:(1)Aiming at the problem of large single-point positioning error of a low-precision GPS receiver,this thesis combines GPS positioning model with data fusion technology of multi-sensor system,and proposes a distributed federated Kalman filtering fusion method for GPS positioning data processing.After each GPS receiver collects the positioning data,firstly,Kalman filter is used in the corresponding local filter to get the local optimal estimation.Secondly,data fusion is carried out in the fusion center,and global fusion estimation is obtained.Then the global fusion estimation is fed back to all local filter to correct the estimation errors of the abnormal local filters.In addition,when using WSNs for data transmission,considering the problem of data packet loss,a predictive compensator strategy is adopted to deal with data packet loss.Finally,the experimental results show that the proposed method can effectively reduce the positioning error and improve the positioning accuracy.(2)When using a low-precision GPS receiver for landslide displacement monitoring,the low-precision GPS receiver has a large single-point positioning error,and it is impossible to determine whether a landslide occurs by the distance between the initial position and the new position of the GPS receiver.This thesis considers this problem from the point of view of data classification,and proposes a BP neural network methods for landslide displacement monitoring.BP neural network is used to classify the GPS positioning data before and after landslide occurrence into two categories,one belongs to the GPS data before landslide occurrence,the other belongs to the GPS data after landslide occurrence.The experimental results show that the distributed federal Kalman filter fusion method is used to process the GPS receiver's positioning data,and then BP neural network is used to classify it,which can effectively improve the landslide monitoring accuracy.