POCET最优相位搜索的数值算法
发布时间：2018-11-17 07:19
【摘要】：现代化的卫星导航信号要求在星上高功率放大器之前恒包络复用同频点甚至临近的双频点/三频点的多个导航信号分量。最优相位恒包络发射(POCET)技术能够恒包络复用任意路数信号且达到最高复用效率。已见诸报道的POCET最优相位搜索的数值算法存在计算量大、收敛速度慢、当迭代点远离最优解或要求提高计算精度时难以收敛到局部最优解等问题。针对导航信号最优恒包络复用论证的需求,首先在优化目标函数中引入增广拉格朗日乘子法以解决当终止误差减小无法收敛到局部最优解的问题;其次对于搜索步长的确定摈弃了已有的精确线搜索算法而采用基于Armijo准则的非精确线搜索算法,并比较研究了最速下降法、共轭梯度法、拟牛顿法(包括BFGS法和对称秩1法)等多种搜索方向优化算法的优缺点和适用性;最后通过对BDS B1频点不同功率分配下的最优相位搜索和合成损耗评估,验证了改进后算法的精度高、计算量小、收敛性强等优点,为导航信号调制复用方案的设计和优化提供参考。
[Abstract]:Modern satellite navigation signals require constant envelope multiplexing of multiple navigation signal components at the same frequency point or even adjacent dual-frequency / three-frequency points before the on-board high-power amplifier. Optimal phase constant envelope transmit (POCET) technique is capable of multiplexing any number of signals with constant envelope and achieving the highest multiplexing efficiency. The reported numerical algorithm for POCET optimal phase search has many problems, such as large computational complexity, slow convergence speed and difficulty in converging to local optimal solution when iteration points are far away from the optimal solution or when the computational accuracy is required to be improved. Aiming at the demand of optimal constant envelope multiplexing for navigation signal, an augmented Lagrangian multiplier method is introduced into the optimization objective function to solve the problem that when the termination error is reduced, it can not converge to the local optimal solution. Secondly, the inexact line search algorithm based on Armijo criterion is adopted to determine the search step size, and the steepest descent method and conjugate gradient method are compared. Quasi-Newton method (including BFGS method and symmetric rank 1 method) and other search direction optimization algorithms have their advantages and disadvantages and applicability. Finally, through the optimal phase search and synthesis loss evaluation of BDS B1 frequency points with different power distribution, the advantages of the improved algorithm are verified, such as high accuracy, low computational complexity, strong convergence, and so on. It provides a reference for the design and optimization of the modulation and multiplexing scheme of navigation signal.
【作者单位】： 北京航空航天大学电子信息工程学院;
【基金】：国家自然科学基金(61271197)~~
【分类号】：TN967.1
本文编号：2336941
[Abstract]:Modern satellite navigation signals require constant envelope multiplexing of multiple navigation signal components at the same frequency point or even adjacent dual-frequency / three-frequency points before the on-board high-power amplifier. Optimal phase constant envelope transmit (POCET) technique is capable of multiplexing any number of signals with constant envelope and achieving the highest multiplexing efficiency. The reported numerical algorithm for POCET optimal phase search has many problems, such as large computational complexity, slow convergence speed and difficulty in converging to local optimal solution when iteration points are far away from the optimal solution or when the computational accuracy is required to be improved. Aiming at the demand of optimal constant envelope multiplexing for navigation signal, an augmented Lagrangian multiplier method is introduced into the optimization objective function to solve the problem that when the termination error is reduced, it can not converge to the local optimal solution. Secondly, the inexact line search algorithm based on Armijo criterion is adopted to determine the search step size, and the steepest descent method and conjugate gradient method are compared. Quasi-Newton method (including BFGS method and symmetric rank 1 method) and other search direction optimization algorithms have their advantages and disadvantages and applicability. Finally, through the optimal phase search and synthesis loss evaluation of BDS B1 frequency points with different power distribution, the advantages of the improved algorithm are verified, such as high accuracy, low computational complexity, strong convergence, and so on. It provides a reference for the design and optimization of the modulation and multiplexing scheme of navigation signal.
【作者单位】： 北京航空航天大学电子信息工程学院;
【基金】：国家自然科学基金(61271197)~~
【分类号】：TN967.1
【相似文献】
相关会议论文 前1条
1 何春江;郭速学;;积分变换中卷积的数值算法[A];数学及其应用文集——中南模糊数学和系统分会第三届年会论文集（上卷）[C];1995年
相关硕士学位论文 前1条
1 刘亚允;一般二值信源算术码码谱研究[D];西北农林科技大学;2016年
，本文编号：2336941
本文链接：https://www.wllwen.com/kejilunwen/xinxigongchenglunwen/2336941.html
