Research On Key Techniques Of Indoor Localization For Wireless Sensor Network

With the rapid development of Internet of Things,mobile Internet and smart mobile devices,the demand for location-based services is growing.While the Global Positioning System(GPS)has been widely used with excellent localization performance in outdoor positioning,it still may not be able to provide positioning services with sufficient localization accuracy in indoor environments due to the lack of line of sight(LOS)transmission channels between a satellite and a receiver.Therefore,indoor positioning technology in wireless sensor networks(WSNs)has attracted much attention in recent years.To overcome the defects of existing work and to relax the challenge of indoor localization,five localization solutions which satisfy the practical requirements of high precision,high efficiency and low cost are proposed in this dissertation based on the analysis and conclusions of domestic and overseas researches.The main contributions of this dissertation are listed as follows:An indoor NLOS(Non-line-of-sight)localization algorithm based on Gaussian mixed model is proposed since the obstacles are easy to cause the NLOS propagation in indoor environment.This algorithm uses Gaussian mixed model(GMM)to train the distance measurements containing NLOS errors,Expectation Maximization(EM)method is employed for the more accurate range estimations.Finally,the proposed algorithm employs the residual weighting algorithm with lower complexity to estimate the coordinates of target nodes with range estimations.The proposed algorithm could mitigate the NLOS error effectively and improve the accuracy of localization.In pertinence to the application of multidimensional scaling(MDS)methods in ranging-based positioning systems,an analysis is firstly conducted by the classical MDS algorithm.Modified MDS algorithm and subspace method are also applied for wireless localization.We also depicted the unified framework and general solutions of MDS methods.However,the least square solutions under this framework are not optimal.Their performance is still related to selection of coordinate reference points.To address this problem,a minimum residual MDS algorithm based on particle swarm optimization(PSO)is proposed to derive a new solution for indoor robot localization under the unified framework.The result of analysis indicates that the performance of minimum residual MDS method is immune to selection of reference points.Furthermore,the localization accuracy for indoor robot has been enhanced as compared with the classical MDS algorithm.Localization algorithms based on received signal strength indicator(RSSI)are widely used in the field of target localization due to its advantages of convenient application and independent from hardware devices.Unfortunately,the RSSI values are susceptible to fluctuate under the influence of NLOS in indoor space.Existing algorithms are easily affected by environments,leading to low accuracy and low effectiveness in indoor target localization.Moreover,these approaches require extra prior knowledge about the propagation model.As such,we focus on the problem of localization in mixed LOS/NLOS scenario and propose a novel localization algorithm.In the proposed method,the RSSI is estimated using a GMM.The dissimilarity matrix is built to generate relative coordinates of nodes by a MDS approach.Finally,based on the anchor nodes' actual coordinates and target's relative coordinates,the target's actual coordinates can be computed via coordinate transformation.The experimental verification shows that the proposed method effectively reduces NLOS error and is of higher accuracy in indoor mixed LOS/NLOS environment and still remains effective when we extend single NLOS to multiple NLOS.Existing 3D localization methods suffer from low accuracy and low stability,especially for moving target localization and tracking.The main objective of this paper is to design a novel 3D localization algorithm for GPS-denied environments that can achieve higher stability and accuracy of mobile localization without knowledge of both measurement noise statistics and target motion information.Within each sampling period,the MDS localization algorithm is adopted to make a preliminary estimation of the target's position,according to available distance measurements.Then,a polynomial data fitting method is employed to fit the result of the target's position estimation within a spell of time.The polynomial resulted from fitting is utilized to rectify the estimated result of the current position.The proposed method is evaluated by using our 3D Ultra-wide band(UWB)measurement based indoor localization testbed.The experimental results evince that the algorithm proposed can achieve better performance in terms of higher stability and higher localization accuracy by comparing with existing approaches.Fingerprinting based indoor positioning system(IPS)is gaining more and more research interest under the umbrella of location-based services.However,existing solutions are lacking because of factors such as noisy measurements,high computational complexity and so on.A feedforward neural network named extreme learning machine(ELM)is applied to address these issues.In the proposed system,KELM(Kernel Extreme Learning Machine)is selected as the indoor fingerprint positioning algorithm to both reduce the computational complexity and boost the generalization ability.Extensive real-world indoor localization experiments are conducted on users with smartphone devices and a performance comparison with existing solutions verifies the superiority of the proposed IPS in terms of accuracy and real-time.In summary,the indoor localization techniques have been systematically researched for wireless sensor networks in this dissertation.A series of experiments show that the proposed five solutions are feasible,available and advanced.