Conventional gyrocompass alignment methods are based on relatively small azimuth misalignment angles.However,a marine strapdown inertial navigation system may face large azimuth misalignment angle caused by a failed coarse alignment algorithm.This paper provides a novel gyrocompass alignment method to solve the problem.Effects of system parameters are analyzed and the proper scenario of parameter switch based on the classic control theories is derived.Test results show that compared with the conventional methods,our method can accomplish the initial alignment quickly and accurately under large azimuth misalignment angle.
Community discovery of complex networks,esp.of social networks,has been a hotly debated topic in academic circles in recent years.Since actual networks usually contain some overlapping nodes that are difficult to assign to a certain community,overlapping community discovery is under great demand in practical applications.However,at present network community discovery is mainly done by non-overlapping community discovery methods,overlapping discovery methods are not common.In this context,an overlapping community discovery method is proposed hereby based on topological potential and specific algorithms are also provided.This method not only considers the spread of the uncertainty of community identity of the overlapping nodes in the network,but also realizes a quantified representation,i.e.,uncertainty measure,of the community identity of the overlapping nodes.The experiment results show that this method yields the results that are consistent with those by the classic methods and are more reasonable.
The studies show that numerous complex networks have clustering effect.It is an indispensable step to identify node clusters in network,namely community,in which nodes are closely related,and in many applications such as identification of ringleaders in anti-criminal and anti-terrorist network,efficient storage of data in Wireless Sensor Network(WSN).At present,most of community identification methods still require the specifications of the number or the scale of community by user and still can not handle overlapping nodes.In an attempt to solve these problems,a network community identification method based on utility value is proposed,which is a function of each node's clustering coefficient and degree.This method makes use of individual-centered theory for reference and can automatically determine the number of communities.In addition,this method is an overlapping community identification method in nature.It is shown through contrastive experiments that this method is more efficient than other methods based on individual-centered theory when they control the same amount of information.Finally,a research direction is proposed for network community identification method based on the individual-centered theory.
Mining important nodes in the complex network should not only consider the core nodes, but also consider the locations of the nodes in the network. Despite many researches on discovering important nodes, the importance of nodes in the structural holes is still ignored easily. Therefore, this paper proposes a method of local centrality measurement based on structural holes, which evaluates the nodes importance both by direct and indirect constraints caused by the lack of structural holes around the nodes. In this method, the attributes and locations of the nodes and their first-order and second-order neighbors are taken into account simultaneously. Deliberate attack simulation is carried out through selective deletion in a certain proportion of network nodes. Calculating the decreased ratio of network efficiency is to quantitatively describe the importance of nodes in before-and-after attacks. Experiments indicate that this method has more advantages to mine important nodes compared to clustering coefficient and k-shell decomposition method. And it is suitable for the quantitative analysis of the nodes importance in large scale networks.
