Performance evaluation of SOM based localization for anisotropic wireless ad-hoc networks

In wireless communications, localization has attracted attention from many researchers. Most of the researches focus on isotropic topologies, but in practical scenarios, wireless ad-hoc networks usually have irregular shapes (anisotropic topologies). For anisotropic topologies, many localization algorithms have failed to estimate the location of nodes. This paper proposes a new algorithm based on Self-Organizing Maps (SOM) to deal with this issue. Our proposed algorithm utilizes only connectivity information and information from some heard anchors in the network to determine the location of nodes. By efficiently utilizing the information from anchors and surrounding neighbors, our algorithm minimizes the error in anisotropic networks. From our intensive simulations, the results show that the proposed scheme improves the localization accuracy for anisotropic networks.