简介：以便改进传播精确性和在事情(IoT)的因特网察觉到并且激活信号的效率并且保证系统稳定性，一个适应资源分配算法被建议，它动态地根据他们的信号频率域特征在部件之中分配网络带宽和优先级。一遥远察觉到并且控制无人的地面的车辆(UGV)路径追踪试验台被开发，追踪错误信号的多重UGV为表演评估在模拟被测量。与一样的网络带宽限制，建议算法能减少的结果表演积累并且UGV路径与常规静态的算法相比在超过60%追踪的最大的错误。

简介：Astructuredlow-rankmatrixrecoverymodelforRGBDsalientobjectdetectionisproposed.Firstly,theproblemisdescribedbyalow-rankmatrixrecovery,andthehierarchicalstructureofRGBimageisaddedtothesparsityterm.Secondly,thedepthinformationisfusedintothemodelbyaLaplacianregularizationtermtoensurethattheimageregionswhichsharesimilardepthvaluewillbeallocatedtosimilarsaliencyvalue.Thirdly,avariationofalternatingdirectionmethodisproposedtosolvetheproposedmodel.Finally,bothquantitativeandqualitativeexperimentalresultsonNLPR1000andNJU400showtheadvantageoftheproposedRGBDsalientobjectdetectionmodel.