A multivariate analysis-based for range-free localization algorithm
Hao Yan1, Xiaoyong Yan2, Rulin Dou1
1School of Software Engineering, Jinling Institute of Technology, Nanjing 211169, China
2School of Intelligence Science and Control, Jinling Institute of Technology,Nanjing 211169, China
Proposed an improved DV-Hop localization algorithm (PLS-DVHop) based on partial least squares, which uses the partial least squares to model of hop-count and the Euclidean distances, along with the maximum covariance of input matrix and output matrix to estimate the location of unknown nodes. PLS-DVHop has strong adaptability for different deployment network, and overcomes the shortage of only suitable for isotropic networks in the original algorithm. Simulation results show that PLS-DVHop algorithm has high estimate precision and stable performance, can adapt to different network topologies, and is very suitable for large scale deployment network.