A fuzzy-based localization in range-free wireless sensor network using genetic algorithm&Sinc membership function

In this paper a novel localization technique is proposed using fuzzy logic and genetic algorithm to get nearly precise location of sensor nodes, in a range free localization system. This localization technique gives a stable system in which both magnitude and range of the error are very low. Various membership functions (MF) are tested among which the Sinc MF provides the best results for the system. Genetic algorithm is used to optimize the knowledge base of the fuzzy system and the simulations are done employing Mamdani fuzzy interference method. A thorough analysis compares the accuracy of the proposed system to that of the basic Centroid localization methods. Error analysis is shown using violin plot i.e. a combination of box plot and kernel density plot for the validation of the proposed model.