摘要：This paper proposes a modified fingerprinting hybrid location algorithm for homogeneous and heterogeneous indoor environment. The new method is designed based on fingerprinting to improve the location accuracy. It works in two phases: one is offline phase, which opts for the location of reference point by different kind of complex regions. Another one is online phase, which proposes dynamic K for finding the nearest reference point to improve location accuracy. Sensor- assisted tracking hybrid positioning function is utilized to further improve the low location accuracy problem of heterogeneous regions. Simulation results show that the new method obtains approximately 37.7% improvement comparing with the classical fingerprinting-based algorithms, the location accuracy reaches to 0.5m in 200mx200m heterogeneous network.
