Channel Estimation for Extremely Large-Scale MIMO: Far-Field or Near-Field?

Extremely large-scale multiple-input-multiple-output (XL-MIMO) is promising to meet the high rate requirements for future 6G. To realize efficient precoding, accurate channel state information essential. Existing estimation algorithms with low pilot overhead heavily rely on sparsity in angular domain, which achieved by classical far-field planar-wavefront assumption. However, due non-negligible near-field spherical-wavefront property XL-MIMO, this domain not achievable. Therefore, existing schemes will suffer from severe performance loss. address problem, paper, we study exploiting polar-domain sparsity. Specifically, unlike angular-domain representation that only considers information, propose a representation, simultaneously accounts both and distance information. In way, also exhibits polar based which, on-grid off-grid XL-MIMO schemes. Firstly, an simultaneous orthogonal matching pursuit (P-SOMP) algorithm proposed efficiently estimate channel. Furthermore, iterative gridless weighted (P-SIGW) improve accuracy. Finally, simulations are provided verify effectiveness of our