原子范数和ULA的关系

起因

是在论文里看到说In the existing atomic norm-based methods, since a Toeplitz matrix is formulated, only the ULA can be used for the DOA estimation, limiting the application of atomic norm-based methods.[1]因为有toeplitz矩阵所以现存原子范数方法只能用于ULA，但在RIS辅助的通信系统需要考虑UPA场景，并且UPA是与实际更匹配的。

但这个论文来着RIS的无人机领域，不确定是为什么这样表达。

搜索一些论文得到以下叙述：

[2]

关于gridless DOA的一维估计，表明ANM用于一维DOA估计，因为相干矩阵有Toeplitz结构，需要vandermonde分解，gridless方法局限于在ULA情形下取抽样测量。

论文的主题是将相关的方法从一维ULA转化到一维NUA（non-uniform array）

Currently, gridless DOA estimation is solved via convex relaxation, and is applicable only to uniform linear arrays (ULA). The ULA sample covariance matrix has Toeplitz structure, and gridless DOA is based on the Vandermonde decomposition of this matrix. The Vandermonde decomposition breaks a Toeplitz matrix into its harmonic components, from which the DOAs are estimated.

The weakness of gridless methods is that they are limited to regularly sampled measurements that can only be taken from a uniform linear array (ULA).

Next, atomic norm minimization (ANM) for gridless DOA is generalized to NUA using the IVD.

[3]

全程的叙述是描述ULA和SLA，即还是处于一维，只是是否均匀分布。

Note that SPA/GLS does not require any knowledge of the number of sources and is applicable in both uniform linear array (ULA) and sparse linear array (SLA) cases. 

结论

ANM相关的算法仅针对一维，[1]中的语句意思应该是指不能用于不均匀阵列（SLA），只用于ULA。

---

[1]Reconfigurable Intelligent Surface Aided Sparse DOA Estimation Method With Non-ULA

[2]Gridless DOA Estimation and Root-MUSIC for Non-Uniform Linear Arrays

[3]A Toeplitz Covariance Matrix Reconstruction Approach for Direction-of-Arrival Estimation