A Segmented Off-Grid Near-Field Wideband Channel Estimation Approach for XL-MIMO

With the coming of the 6G era, extremely large-scale multiple-input-multiple-output (XL-MIMO) will be one of the key technologies in the near future. By that time, the number of antennas in a base station (BS) will have increased to hundreds or even thousands. In addition, terahertz (THz) and millimeter wave bands will also be widely used in 6G due to their excellent performance. According to electromagnetic theory, the Rayleigh distance increases as the number of antenna aperture and the frequency increase. This means that the near-field distance expands. However, accurate near-field wideband channel estimation is very difficult, both the beam split effect and the recovery of both distance and angle dimensions have to be taken into consideration. To solve this problem, inspired by the traditional near-field narrowband channel estimation algorithm. We propose a segmented off-grid polar-domain simultaneous iterative gridless weighted (St-P-SIGW) algorithm to estimate the wideband channel, which not only considers the beam split effect but also can efficiently improve the estimation accuracy. At the end of this paper, we compare the proposed algorithm with other algorithms. simulation result shows that the algorithm in this paper achieves better results.