Abstract. Due to the limited energy, storage space and computing ability, data fusion is very necessary in Wireless Sensor Networks (WSN). In this paper, a new variable weight based fuzzy data fusion algorithm for WSN is proposed to improve the accuracy and reliability of the global data fusion. In this algorithm, the weight of each cluster head node in global fusion is not fixed. Time delay, data amount and trustworthiness of each cluster head will all affect the final fusion weight. We get the fusion weights by variable weight based fuzzy comprehensive evaluation or fuzzy reasoning. In the variable weight based fuzzy comprehensive evaluation, by increasing the weight of the factor with too low value, we can give prominence to deficiency and the clusters with too long time delay or too small amount or too low trustworthiness will get smaller weights in data fusion. And therefore, the cluster head node with deficiency will have a small influence in global fusion. Simulation shows that this algorithm can obtain a more accurate and reliable fusion results especially when there are data undetected or compromised nodes compared with traditional algorithms.
The vibration signal of rolling bearing in mechanical equipment is nonlinear and nonstationary under the influence of various excitation sources. This paper combines empirical wavelet transform (EWT) with the fuzzy function and gives a method of fault signal recognition. Several modal components of the original signal can be obtained by decomposition. Components with more features of the original signal can represent some features of the original signal. The mutual information between each modal component and the original signal can be calculated. The noise and useful information can be identified according to the value of the mutual information, and then the noise of the signal can be filtered out to reconstruct the original signal The fuzzy functions of the known class signals and the signals to be identified are calculated, and the correlation coefficients of the fuzzy functions of the signals to be identified and the signals to be identified are calculated.
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.