The variational Bayesian method solves nonlinear estimation problems by iteratively computing the integral of the marginal density. Many researchers have demonstrated the fact its performance depends on the linear approximation in the computation of the variational density in the iteration and the degree of nonlinearity of the underlying scenario. In this paper, two methods for computing the variational density, namely, the natural gradient method and the simultaneous perturbation stochastic method, are used to implement a variational Bayesian Kalman filter for maneuvering target tracking using Doppler measurements. The latter are collected from a set of sensors subject to single-hop network constraints. We propose a distributed fusion variational Bayesian Kalman filter for a networked maneuvering target tracking scenario and both of the evidence lower bound and the posterior Cramér–Rao lower bound of the proposed methods are presented. The simulation results are compared with centralized fusion in terms of posterior Cramér–Rao lower bounds, root-mean-squared errors and the 3 bound.
A tree $T$, in an edge-colored graph $G$, is called {\em a rainbow tree} if no two edges of $T$ are assigned the same color. A {\em $k$-rainbow coloring}of $G$ is an edge coloring of $G$ having the property that for every set $S$ of $k$ vertices of $G$, there exists a rainbow tree $T$ in $G$ such that $S\subseteq V(T)$. The minimum number of colors needed in a $k$-rainbow coloring of $G$ is the {\em $k$-rainbow index of $G$}, denoted by $rx_k(G)$. Graph operations, both binary and unary, are an interesting subject, which can be used to understand structures of graphs. In this paper, we will study the $3$-rainbow index with respect to three important graph product operations (namely cartesian product, strong product, lexicographic product) and other graph operations. In this direction, we firstly show if $G^*=G_1\Box G_2\cdots\Box G_k$ ($k\geq 2$), where each $G_i$ is connected, then $rx_3(G^*)\leq \sum_{i=1}^{k} rx_3(G_i)$. Moreover, we also present a condition and show the above equality holds if every graph $G_i (1\leq i\leq k)$ meets the condition. As a corollary, we obtain an upper bound for the 3-rainbow index of strong product. Secondly, we discuss the 3-rainbow index of the lexicographic graph $G[H]$ for connected graphs $G$ and $H$. The proofs are constructive and hence yield the sharp bound. Finally, we consider the relationship between the 3-rainbow index of original graphs and other simple graph operations : the join of $G$ and $H$, split a vertex of a graph and subdivide an edge.
Cardiovascular disease is a sudden and extremely disabling and deadly disease that hinders the development of public health care in China and poses a serious threat to the health of users. Therefore, this paper proposes a deep neural network-assisted diagnosis of cardiovascular disease based on TAdam algorithm optimization under a distributed medical monitoring network to enable rapid identification of cardiovascular patients, timely detection and reporting of the condition. This paper addresses the problems that deep neural network models are susceptible to the influence of their structure and hyperparameters and poor generalization convergence, firstly, the mainstream optimization algorithms based on deep learning are sorted out, and finally the T Adam algorithm is used to optimize the model. By comparing the common gradient optimization algorithms SGD (Stochastic Gradient Descent), Adagrad (Adaptive Gradient Algorithm), RMSprop (Root Mean Square propagation), Adam (Adaptive Moment Estimation) and the improved TAd am optimization algorithm trained and tested on the same dataset, the experimental results show that the TAdam algorithm exhibits the fastest convergence speed and the best generalization performance on the cardiovascular disease dataset. The method proposed in this paper enables rapid diagnosis of the user's disease risk and feedback to the relevant personnel to maximise user safety.