ID 原文 译文
8514 针对LEO/MEO/GEO三层卫星网络,首先根据业务类型划分优先级,其次运用自回归积分滑动平均(autoregressive integrated moving average,ARIMA)模型预测小区内新呼叫业务用户量。 For LEO/MEO/GEO satellite network, three layer first priority.the depending on the type of business, the second using autoregressive integrating moving average (autoregressive integrated moving average, ARIMA) model to predict the new call business users in the district.
8515 然后根据预测用户量和卫星剩余信道带宽为用户选择接入层,再利用博弈模型判断是否允许用户接入,最后根据降级因子选择降级用户让出信道带宽。 And then according to the prediction residual channel bandwidth for the user to select users and satellite access layer, using the game model to determine whether to allow users to access, according to the degradation factor to select downgrade user release channel bandwidth.
8516 仿真结果表明,该方法能够降低整体阻塞率,提高整体效用。 The simulation results show that the method can reduce overall model.next, improve overall utility.
8517 主要研究以交替方向法为基础的总变分图像恢复模型,结合约束优化问题以及快速迭代技术,提出了一种约束总变分图像恢复的快速算法。 Research based on the alternating direction method of total variational image restoration model, combining with the constrained optimization problem and fast iterative technique, this paper proposes a constraint total variational fast algorithm of image restoration.
8518 对总变分模型添加范围约束,利用交替方向法进行求解,把原问题转化为3个子问题,分别用迭代阈值法、快速傅里叶变换法以及投影法进行求解。 Add range constraint on total variational model, the use of alternating direction method for solving, to convert the original problem into three sub problems, respectively, using iterative threshold method, fast Fourier transform method and projection method for solving.
8519 把快速迭代技术应用于迭代阈值法来提高计算效率,利用非精确计算法来克服系数矩阵为随机投影阵带来的傅里叶变换的计算费时问题。 The fast iterative techniques applied in the iterative threshold method to improve the efficiency of calculation, the accurate calculation method is used to overcome the coefficient matrix of the random projection matrix Fourier transform to calculate the time consuming problem.
8520 数值试验结果表明,针对随机投影阵下的约束总变分问题,新方法在提高计算效率的同时还能得到很好的图像恢复效果。 Numerical test results show that, in view of the random projection matrix under the constraints of the total variational problem, a new method to improve calculation efficiency at the same time also can get very good effect on image restoration.
8521 基于分布式多输入多输出雷达,针对目标跟踪精度的优化问题提出了一种联合资源优化分配算法。 Based on distributed multiple input multiple output radar, the optimization problem in target tracking precision a joint resources optimization allocation algorithm is proposed.
8522 首先,推导了机动目标跟踪误差的贝叶斯克拉美罗下界(Bayesian Cramer Rao lower bound,BCRLB),由BCRLB可知其跟踪精度主要由信号发射功率、带宽和信号有效时宽决定。 First, this paper derived the luo bei Ye Sike Latin America and lower bounds of the maneuvering target tracking error (Bayesian Cramer Rao lower bound, BCRLB), by BCRLB knowable its tracking accuracy is mainly composed of signal transmission power, wide signal bandwidth and effective decision.
8523 然后,以最小化目标的BCRLB为目标函数,建立了包含相应的3个资源变量的优化模型,分析可知该模型的求解是一个非凸问题的求解。 Then, in order to minimize the target BCRLB as objective function, established variable contains the corresponding three resources optimization model, the analysis of the model solution is a non-convex problem solving.