Research On Multi-target Wireless Tomography Positioning Method Based On Wireless Sensor Networ

In recent years,the wireless tomography localization technology based on Wireless Sensor Networks(WSN)has become a hot research topic in the field of passive localization.That is,the target does not need to carry any electronic device,and the system can realize the localization and tracking only by analyzing the wireless signal disturbance caused by the target.This technology has broad research value and application prospects.However,the existing technology can not solve the problem of indoor multi-target localization and tracking.In order to effectively solve the problem of indoor multi-target localization,a multi-target localization algorithm based on wireless tomography is proposed.Firstly,the UWB wireless sensor network is introduced,and the RSS value is calculated by using the single channel impulse response value to determine the line-of-sight component of the wireless communication link.Secondly,combing wireless tomography technology with Gaussian bilateral filter to obtain the wireless propagation field image of indoor monitoring areas.Then,the radar CFAR is used to adaptively adjust the threshold to obtain the target candidate location.Finally,the density clustering algorithm without a priori number of clusters is used to obtain the localization information of multiple targets.With the purpose of dealing with the problem of estimating the number of targets in indoor environment,an indoor monitoring area target number estimation method based on dynamic threshold detector is proposed.By setting different numbers of targets(people),analyzing the influence of the received signal strength(RSS)of the communication link between the UWB wireless sensor nodes at a fixed distance,and calculating the pixel value of the RTI image position where the target is located,the number of targets is finally estimated.For the problem of trajectory crossing and parallel interference of multiple targets(people),a multi-target tracking method based on joint probabilistic data association and Kalman Filter is used.Firstly,the association tracking gate is introduced,and then transform the multi-target tracking problem into multi-target localization association problem.Secondly,considering the trajectory crossing and parallel interference of multi-targets(people),the joint probabilistic data association algorithm is used to associate the multi-targets localization with the existing target trajectory in real time.Finally,the Kalman Filter corresponding to the JPDA algorithm is used to make the multi-target tracking trajectory closer to the real trajectory.Field experimental results show that the method proposed in this paper can effectively locate multiple targets(people),estimate the number of targets and track the trajectory of targets in indoor environment.