Random forest is an excellent ensemble learning method, which is composed of multiple decision trees grown on random input samples and splitting nodes on a random subset of features. Due to its good classification and generalization ability, random forest has achieved success in various domains. However, random forest will generate many noisy trees when it learns from the data set that has high dimension with many noise features. These noisy trees will affect the classification accuracy, and even make a wrong decision for new instances. In this paper, we present a new approach to solve this problem through weighting the trees according to their classification ability, which is named Trees Weighting Random Forest (TWRF). Here, Out-Of-Bag, which is the training data subset generated by Bagging and not involved in building decision tree, is used to evaluate the tree. For simplicity, we choose the accuracy as the index that notes tree's classification ability and set it as the tree's weight. Experiments show that TWRF has better performance than the original random forest and other traditional methods, such as C45, Naïve Bayes and so on.
Tag localization for asynchronous wireless sensor networks requires the development of a scheme for clock synchronization. This remains a difficult and open problem since the performance of tag localization can be adversely affected by complications such as reply time and relative clock skew. Joint clock synchronization and a tag localization algorithm that implements a multi-anchor compensated timeof-flight (TOF) to the asynchronous wireless sensor network is a possible and viable solution. Although previous methods that leverage TOF measurements are effective and easily conducted, their performance is not always superior due to the relative clock skew. In this paper, we propose to extend the joint clock/tag synchronization/localization algorithm by introducing a compensation factor that can cancel relative clock skews from multi-tag anchor pairs. We apply a least squares estimation (LSE) algorithm to both the time of emission (TOE) and time of arrival (TOA) for the clock synchronization step. Under the assumption of a Gaussian measurement noise model, the tag localization problem is approximately solved by maximum likelihood estimation (MLE). To assess the performance of our algorithm, we derive the mean square error (MSE) of both relative clock skew and tag location and numerically evaluate the Cramér-Rao lower bound (CRLB) as a benchmark. The simulation results show that the accuracy of the relative clock skew-based estimation and tag localization are significantly improved over traditional algorithms when the appropriate reply time is selected. This is what our proposed algorithm focuses on: it is robust to tag mobility to some extent. We test the performance of proposed algorithm using a well-designed experiment. Based on the experiment results, the localization algorithm can achieve high accuracy without an additional restriction on the reply time and the clock skew.INDEX TERMS Source localization, wireless sensor network (WSN), time-of-flight (TOF), symmetric double-sided two-way ranging (SDS-TWR), relative clock skew.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.