We consider a reconfigurable intelligent surface (RIS)-assisted extremely large-scale multiple-input multiple-output (XL-MIMO) downlink system, where an XL-MIMO array serves two groups of single-antennas users, namely near-field users (NFUEs) and far-field users (FFUEs). FFUEs are subject to blockage, and their communication is facilitated through the RIS. We consider three precoding schemes at the XL-MIMO array, namely central zero-forcing (CZF), local zero-forcing (LZF) and maximum ratio transmission (MRT). Closed-form expressions for the spectral efficiency (SE) of all users are derived for MRT precoding, while statistical-form expressions are obtained for CZF and LZF processing. A heuristic visibility region (VR) selection algorithm is also introduced to help reduce the computational complexity of the precoding scheme. Furthermore, we devise a two-stage phase shifts design and power control algorithm to maximize the sum of weighted minimum SE of two groups of users with CZF, LZF and MRT precoding schemes. The simulation results indicate that, when equal priority is given to NFUEs and FFUEs, the proposed design improves the sum of the weighted minimum SE by 31.9%, 37.8%, and 119.2% with CZF, LZF, and MRT, respectively, compared to the case with equal power allocation and random phase shifts design. CZF achieves the best performance, while LZF offers comparable results with lower complexity. When prioritizing NFUEs or FFUEs, LZF achieves strong performance for the prioritized group, whereas CZF ensures balanced performance between NFUEs and FFUEs.

Extremely large-scale multiple-input multiple-output (XL-MIMO) is a transformative technology to achieve spectral-efficient and energy-saving wireless communication. However, XL-MIMO leads to the near-field propagation becoming dominant. In this paper, we examine a reconfigurable intelligent surface (RIS)-assisted XL-MIMO communication system to serve two distinct groups of users, namely near-field users (NFUEs), directly served by the XL-MIMO, and far-field users (FFUEs), served with the assistance of a RIS. We derive the signal-to-interference-plus-noise ratio (SINR) expressions for whole-array-based precoders, including maximum-ratio transmission (MRT), and zero-forcing (ZF). Moreover, we take into account the spatially, non-stationary channel characteristics, indicating that user terminals may only have visibility of a specific portion of the array, referred to as the visibility region (VR). To further leverage the VR for complexity reduction, we propose a heuristic algorithm designed to determine the VR, while also guaranteeing the individual SINR requirements for both NFUEs and FFUEs. Simulation results indicate that utilizing VR with our proposed heuristic algorithm yields performance comparable to a benchmark utilizing the whole-array, albeit with a notable reduction in the number of antennas and computational complexity.

In this paper, we propose a general framework for Distributed Source Coding (DSC) of multiple sequentially correlated information sources. In particular, we modularize the joint decoding function into a conceptually simple structure by utilizing a notion of "complementarity" on the code space. Assuming a constrained Hamming-distance based correlation, our general framework of DSC achieves arbitrary non-asymmetric-rate compression of multiple sources with a same overall compression rate as asymmetric compression schemes. This DSC framework does not only have flexibility with respect to compression rates, but also accommodates the use of any linear block code and any of its corresponding complements. By investigating the unique connection between syndromes and complementary codewords, we translate the major steps of DSC joint decoding into syndrome decoding followed by channel encoding via a linear block code and also via its complement code. Our aim is to achieve transparency which facilitates practical implementation.