ID 原文 译文
55837 将该模型与基于遗传算法优化的RBF神经网络预测控制器、基于粒子群优化的多步SVR模型预测控制器和基于遗传算法优化的多步SVR模型预测控制器进行比较分析,仿真结果表明该预测控制模型优于其他控制器,具有良好的预测性能,可有效的对非线性系统进行预测控制。 The model and the RBF neural network predictive controller based on genetic algorithm optimization, based on particle swarm optimization and multi-step SVR model predictive controller based on genetic algorithm to optimize the multi-step SVR model predictive controller for comparative analysis, the simulation results show that the predictive control model is superior to the other controller, has a good prediction performance, can be effective to forecast the nonlinear system control.
55838 提出了一种基于关联规则的信道编码识别方法,解决了线性分组码的盲识别问题。 This paper proposes a channel coding identification method based on association rules, can solve the problem of blind identification of linear block code.
55839 在分析线性分组码中的信息位和监督位之间存在一一对应关系这一特性的基础上,利用关联规则寻找出隐藏在二进制序列中的对应关系,实现了分组码的盲识别。 Based on the analysis of linear block code bits of information and supervision of a one-to-one correspondence between the relations on the basis of this feature, USES association rules to find out the corresponding relationship, hidden in the binary sequence block code blind identification is realized.
55840 首先将按照一定规则排列的序列进行二分段划分,并依此构建相关事务项集,然后计算二分段项集中每项事务的支持度和置信度,在满足支持度条件下,通过遍历的方式寻找最大置信度和最少映射关系种类,达到正确识别编码的目的。 First to arrange according to certain rules of sequence for two block division, and in accordance with this build related transaction item set, and then calculate 2 section on each transaction support and confidence, under the condition of meet the support, through the way of traversing the looking for the biggest confidence and least mapping relation types, achieve the goal of correct identification code.
55841 由于采用了二分段划分的方法,因此极大地降低了关联规则搜索的复杂度,而且所提算法与现有识别方法相比对误码率要求不高。 Due to adopt the method of the second block division, thus greatly reduces the complexity of the search, the association rules and the proposed algorithm is compared with the existing recognition methods on bit error rate is not high.
55842 仿真实验验证了算法具有较高的鲁棒性和识别正确率,表明该算法具有一定的工程应用价值。 Simulation results verify the effectiveness of the algorithm has higher identification accuracy and robustness, show that the algorithm has certain engineering application value.
55843 分析了卡尔曼滤波算法在目标状态发生突变和运动模型建立不准确时估计精度降低,甚至发散的原因,对比自适应渐消卡尔曼滤波算法,提出了一种通过直接修正预测值来提高卡尔曼滤波算法精度、改善算法性能的修正算法。 Analyzed the kalman filtering algorithm in target state mutates and motion model establishing inaccurate estimation precision is reduced, or even divergence, contrast adaptive fading kalman filter algorithm, proposed a modified procedure is to improve precision of the kalman filtering algorithm, improves the algorithm performance correction algorithms.
55844 修正的算法通过设置判定准则和修正准则,实时修正预测值,在滤波初始阶段可迅速降低估计误差、提高稳态时的滤波精度、缩短收敛时间;当目标发生状态突变时,可消除或降低由于目标状态突变造成的滤波跟踪精度下降、滤波发散的问题;当目标运动建模不准确时,可消除或降低由于建模不准确带来的模型误差。 Correction algorithm by setting criteria and guidelines, real-time correction procedure is in initial stage of filtering can quickly reduce the estimation error, improve the filtering precision of the steady-state and shorten the convergence time;When the target state mutation occurs, can eliminate or reduce as a result of target state mutations cause the loss of tracking accuracy of filtering, filtering divergence problem;When the target movement model is not accurate, can eliminate or reduce the model error caused by the inaccurate modeling.
55845 仿真实例说明了算法的有效性和较强的实际应用指导意义。 The simulation example is given to illustrate the effectiveness of the algorithm and strong practical significance.
55846 针对双重决策信息下的群决策问题,提出了一类基于群体意见交互式修正的信息联动决策方法。 Under double decision-making information for group decision making problems, puts forward the information linkage decisions based on a group opinion interactive correction method.