Beyond Near-Field: Far-Field Location Division Multiple Access in Downlink MIMO Systems

Exploring channel dimensions has been the driving force behind breakthroughs in successive generations of mobile communication systems. In 5G, space division multiple access (SDMA) leveraging massive MIMO has been crucial in enhancing system capacity through spatial differentiation of users. However, SDMA can only finely distinguish users at adjacent angles in ultra-dense networks by extremely large-scale antenna arrays. For a long time, most research has focused on the angle domain of the space, overlooking the potential of the distance domain. Near-field location division multiple access (LDMA) was proposed based on the beam-focusing effect yielded by near-field spherical propagation model, partitioning channel resources by both angle and distance. To achieve a similar idea in the far-field region, this paper introduces a far-field LDMA scheme for wideband systems based on orthogonal frequency division multiplexing (OFDM). Benefiting from frequency diverse arrays (FDA), it becomes possible to manipulate beams in the distance domain. Combined with OFDM, the inherent cyclic prefix ensures a complete OFDM symbol can be received without losing distance information, while the matched filter of OFDM helps eliminate the time-variance of FDA steering vectors. Theoretical and simulation results show that LDMA can fully exploit the additional degrees of freedom in the distance domain to significantly improve spectral efficiency, especially in narrow sector multiple access (MA) scenarios. Moreover, LDMA can maintain independence between array elements even in single-path channels, making it stand out in MA schemes at millimeter-wave and higher frequency bands.

Deep Learning Model for Demodulation Reference Signal based Channel Estimation

In this paper, we propose a deep learning model for Demodulation Reference Signal (DMRS) based channel estimation task. Specifically, a novel Denoise, Linear interpolation and Refine (DLR) pipeline is proposed to mitigate the noise propagation problem during channel information interpolation and to restore the nonlinear variation of wireless channel over time. At the same time, the Small-norm Sample Cost-sensitive (SSC) learning method is proposed to equalize the qualities of channel estimation under different kinds of wireless environments and improve the channel estimation reliability. The effectiveness of the propose DLR-SSC model is verified on WAIC Dataset. Compared with the well know ChannelNet channel estimation model, our DLR-SSC model reduced normalized mean square error (NMSE) by 27.2dB, 22.4dB and 16.8dB respectively at 0dB, 10dB, and 20dB SNR. The proposed model has won the second place in the 2nd Wireless Communication Artificial Intelligence Competition (WAIC). The code is about to open source.