Near-Field Analysis of Extremely Large-Scale MIMO: Power, Correlation, and User Selection

With the fast development of communication technology, mobile networks have been evolving from the fifth generation (5G) to the sixth generation (6G). One of the most important technologies in 5G is massive multiple input multiple output (MIMO). In 6G, it has been extended to extremely large-scale MIMO (XL-MIMO) over the TeraHz band, which makes it easier for users to fall into the near-field communication range. However, the previous performance analysis based on the far-field assumption can be very inaccurate under the near-field scenario. Hence, it is necessary to use the near-field channel models to redo these analyses. In this work, we summarize previous analytical results on received signal-to-noise ratio for specific near-field wave models. Then, we derive the generalized formula for the received power of different wave models and antenna structures. We newly derive our closed-form formula for the correlation between different users by the stationary phase method. These results can be applied to different beam-forming schemes and the multipath case. Based on these analytical results, we manage to make a sum rate analysis for different antenna arrays and near-field channel models in a multi-user XL-MIMO system. Finally, with the modification by our analytical result, we show a dramatic speed-up of the previous user selection algorithm, while reaching the same sum rate.