Abstract
We consider the problem how to build the dataset model which supports to track the target in wireless sensor network (WSN) for online tranining phase. Our current works introduced particle filter (PF) based on Kullback-Leibler Distance (KLD) with adjusted bound error method to ameliorate the negative Received Signal Strength (RSS) variations by creating a sample set within the high-likelihood region. The bound error of this method is one of three parameters that decide to enhance the estimation accuracy and convergence rate of declining number of particle used. Therefore, in this paper, we propose the dataset model and apply a typical supervised machine learning such as K-Nearest Neighbor (KNN) to predict the bound error. The first iteration, using the observation information via KLD resampling optimal bound error to conduct a resampling on the basis of the initial bound error. From the second to the end iteration, we propose the KNN technique to search the predicted bound error value that fulfills the minimum of mean number of particle used between at the current and the next iteration. Our experiments show that our dataset model applied KNN-KLD method imptoves the estimation accuracy and the efficient number of particles compared to the traditional methods.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have