当前位置: X-MOL 学术Digit. Signal Process. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Gridless DOD and DOA estimation in bistatic MIMO radar using 2D-ANM and its low complexity algorithms
Digital Signal Processing ( IF 2.9 ) Pub Date : 2020-11-01 , DOI: 10.1016/j.dsp.2020.102900
Wen-Gen Tang , Hong Jiang , Qi Zhang

Compared with the traditional subspace methods, the application of compressed sensing (CS) to the direction-of-departure (DOD) and direction-of-arrival (DOA) estimation in bistatic multiple-input multiple-output (MIMO) radar can achieve robustness to coherent targets and high localization accuracy with a limited number of snapshots. However, the performance of grid-based CS reconstruction methods degrades since there is an unavoidable basis mismatch between the actual DOD/DOA and the assumed basis. As a gridless CS method, the atomic norm minimization (ANM) has attracted much attention. Specifically, the use of multiple snapshots in ANM has improved the parameter estimation accuracy in contrast with a single snapshot. In this paper, we address the problem of gridless DOD and DOA estimation in bistatic MIMO radar, and develop a multiple-snapshot 2D-ANM algorithm and its two low complexity versions. We start with a rigorous derivation on how to convert the multiple-snapshot 2D-ANM into a semi-definite programming problem, and then explore its dual problem. To overcome the heavy computational burden of the 2D-ANM when the numbers of snapshots and array elements increase, we further propose a 2D-ANM algorithm with snapshot reduction (2D-ANM-SR), as well as an improved 2D-ANM-SR algorithm based on alternating direction method of multipliers (2D-ANM-SR-ADMM). Numerical examples show that with similar accuracy in comparison to 2D-ANM, much lower computational complexity can be achieved via 2D-ANM-SR and 2D-ANM-SR-ADMM.



中文翻译:

使用2D-ANM及其低复杂度算法的双基地MIMO雷达中的无网格DOD和DOA估计

与传统子空间方法相比,在双基地多输入多输出(MIMO)雷达中将压缩感知(CS)应用于出发方向(DOD)和到达方向(DOA)估计可以实现鲁棒性连贯的目标和高定位精度,快照数量有限。但是,基于网格的CS重建方法的性能会降低,因为实际的DOD / DOA与假定的基准之间不可避免地存在基础不匹配。作为无网格CS方法,原子范数最小化(ANM)引起了很多关注。具体而言,与单个快照相比,在ANM中使用多个快照提高了参数估计的准确性。在本文中,我们解决了双基地MIMO雷达中的无网格DOD和DOA估计问题,并开发了多快照2D-ANM算法及其两个低复杂度版本。我们首先严格地推导如何将多快照2D-ANM转换为半定编程问题,然后再探讨其双重问题。为了克服快照和数组元素数量增加时2D-ANM的繁重计算负担,我们进一步提出了一种具有快照减少功能的2D-ANM算法(2D-ANM-SR),以及一种改进的2D-ANM-SR乘法器交替方向法的二维算法(2D-ANM-SR-ADMM)。数值示例表明,与2D-ANM相比,其准确性相似,通过2D-ANM-SR和2D-ANM-SR-ADMM可以实现更低的计算复杂度。我们首先严格地推导如何将多快照2D-ANM转换为半定编程问题,然后再探讨其双重问题。为了克服快照和数组元素数量增加时2D-ANM的繁重计算负担,我们进一步提出了一种具有快照减少功能的2D-ANM算法(2D-ANM-SR),以及一种改进的2D-ANM-SR乘法器交替方向法的二维算法(2D-ANM-SR-ADMM)。数值示例表明,与2D-ANM相比,其准确性相似,通过2D-ANM-SR和2D-ANM-SR-ADMM可以实现更低的计算复杂度。我们首先严格地推导如何将多快照2D-ANM转换为半定编程问题,然后再探讨其双重问题。为了克服快照和数组元素数量增加时2D-ANM的繁重计算负担,我们进一步提出了一种具有快照减少功能的2D-ANM算法(2D-ANM-SR),以及一种改进的2D-ANM-SR乘法器交替方向法的二维算法(2D-ANM-SR-ADMM)。数值示例表明,与2D-ANM相比,其准确性相似,通过2D-ANM-SR和2D-ANM-SR-ADMM可以实现更低的计算复杂度。我们还提出了一种具有快照减少功能的2D-ANM算法(2D-ANM-SR),以及一种基于乘法器交替方向方法的改进2D-ANM-SR算法(2D-ANM-SR-ADMM)。数值示例表明,与2D-ANM相比,其准确性相似,通过2D-ANM-SR和2D-ANM-SR-ADMM可以实现更低的计算复杂度。我们还提出了一种具有快照减少功能的2D-ANM算法(2D-ANM-SR),以及一种基于乘法器交替方向方法的改进2D-ANM-SR算法(2D-ANM-SR-ADMM)。数值示例表明,与2D-ANM相比,其准确性相似,通过2D-ANM-SR和2D-ANM-SR-ADMM可以实现更低的计算复杂度。

更新日期:2020-11-13
down
wechat
bug