Perfomance Research of Improved MDS-MAP Algorithm in Wireless Sensor Networks Localization

The algorithms based on classical Multidimensional Scaling (MDS) only need 3 or 4 anchor nodes and can provide higher accuracy than other schemes. But it's high computational complexity and large energy consumption cause its shorter lifecycle, so an improved positioning algorithm for Wireless Sensor Networks (WSN) is presented in the paper. It reduces distance matrix involved and fully uses nodes with ranging capability. We conduct simulation studies under square shaped topologies with different communication radius, different connectivity levels and different number of anchors. Simulation results show that improved MDS provides a lower position estimation error than classical MDS and have less running time. In the end, some parameters are suggested, which can make WSN get the optimal network performance.