Cell-free radio access networks (CF-RANs) with network-assisted free-duplex (NA-FD) architecture unify flexible-duplex, hybrid-duplex, and full-duplex operations, but the coupled uplink/downlink interference and centralized processing impose substantial computational and signaling burdens. To address these challenges, this paper studies distributed transceiver design and access point (AP) duplex mode selection under edge distributed unit (EDU)-level information constraints and per- AP power limitations. We develop a partial distributed block coordinate descent (PDBCD) algorithm that decomposes the original sum-rate maximization into three tractable subproblems and iteratively approximates MMSE performance. The proposed design fully leverages EDU computing resources, reducing the computation load at the cloud computing unit (CCU), and significantly lowering fronthaul signaling overhead through an adaptive inter-EDU information sharing mechanism. Furthermore, by integrating a low-complexity greedy search for duplex mode assignment, the algorithm effectively mitigates cross-link interference (CLI) in NA-FD systems. Simulation results show that the proposed scheme improves spectral efficiency compared with conventional duplex mode, and centralized/distributed MMSE baselines, while maintaining strong scalability under practical deployment constraints.

The scalable cell-free radio access network (CF-RAN), built on cell-free massive multiple-input multiple-output (CF-mMIMO) networks, achieves a critical trade-off between computational complexity and system performance in cooperative transmission through the rational division of physical-layer functionalities. However, due to its distributed transmission architecture, unavoidable propagation delay differences arise in signal arrival times across different receivers during cooperative transmission, leading to asynchronous reception effects that severely degrade system performance. In this paper, we analyze the specific impacts of asynchronous reception effects in scalable CF-RAN systems, including accumulated phase offset on received signals, as well as inter-carrier interference (ICI) and intersymbol interference (ISI). We investigate channel estimation under non-ideal channel state information (CSI) acquisition caused by non-orthogonal pilot sequences and asynchronous reception effects, deriving closed-form expressions for the achievable uplink and downlink spectral efficiency (SE) in scalable CF-RAN systems under asynchronous conditions. To mitigate asynchronous reception effects, we introduce a beam division multiple access (BDMA) transmission scheme into scalable CF-RAN systems, leveraging large-scale antenna arrays at remote radio units (RRUs) to achieve beam-domain multi-user spatial multiplexing. Building on this framework, we implement per-beam time delay compensation (PBTDC) on RRU antenna arrays to approximate asynchronous received signals as synchronized and derive closed-form expressions for the achievable SE of uplink/downlink in asynchronous scalable CF-RAN systems with PBTDC architecture. Numerical simulations demonstrate that asynchronous reception effects significantly degrade channel estimation and data transmission performance in scalable CF-RAN systems. In contrast, the proposed PBTDC architecture based on BDMA transmission effectively mitigates these adverse effects and substantially enhances system performance.

Orthogonal Time Frequency Space (OTFS) modulation can effectively enhance transmission efficiency in doubly selective channels and is considered as one of the modulation techniques for next generation wireless communication. To meet the demands of machine-type communication in high-speed scenarios, this paper conducts research on a massive random access scheme based on OTFS modulation. Specifically, the transmission and reception model of multiple-input multiple-output (MIMO) OTFS signals is analyzed, where channel estimation is formulated as a block-sparse signal recovery problem. Therefore, based on existing superimposed and embedded pilot schemes, we propose a novel hybrid preamble scheme. This scheme utilizes the superimposed preamble and embedded preamble to achieve rough active user detection (AUD) and precise AUD, respectively, enabling accurate detection and channel estimation while supporting a large number of device access. On this basis, this paper employs a generalized approximate message passing with pattern-coupled sparse Bayesian learning (GAMP-PCSBL) algorithm that can capture the block sparsity characteristics of the channel matrix and achieve accurate estimation. The results of numerical simulations verify the effectiveness and superiority of the proposed scheme.

As a critical component of beyond fifth-generation (B5G) and sixth-generation (6G) mobile communication systems, ultra-reliable and low-latency communication (uRLLC) imposes stringent requirements on both latency and reliability. In recent years, with the evolution of mobile communication networks, centralized and distributed processing schemes for cell-free massive multiple-input multiple-output (CF-mMIMO) have attracted significant attention. This paper investigates the performance of a novel scalable cell-free radio access network (CF-RAN) architecture featuring multiple edge distributed units (EDUs) under the finite block length regime. Closed-form expressions for the upper and lower bounds of the expected sum spectral efficiency (SE) are derived, where centralized and fully distributed deployment can be treated as two special cases, respectively. Furthermore, the spatial distributions of user equipment (UE) and remote radio units (RRUs) are analyzed, revealing that interleaving RRUs deployment associated with the EDUs can enhance SE performance under finite block length constraints with a specified transmission error probability. This paper also compares Monte-Carlo simulation results with multi-RRU clustering-based collaborative processing, validating the accuracy of the space-time exchange theory in the scalable CF-RAN scenario. By deploying scalable EDUs, a practical tradeoff between latency and reliability can be achieved through the spatial degree-of-freedom (DoF), thereby offering a distributed and scalable realization of the space-time exchange theory.

In recent years, with the improvement of mobile communication network performance, cell-free mMIMO with fully integrated and distributed processing has been widely studied, and wireless access networks have also become a widely studied topic in the academic community. This article analyzes the spectral efficiency of a new full spectrum cell-free wireless access network architecture with multiple EDUs and derives the performance upper and lower bounds of its traversal achievable rate. The full set formula and distributed processing can be used as two special cases; Secondly, the article further considers the distribution of users and large-scale fading models and studies the location distribution of RRUs. It is concluded that a uniform distribution of RRUs is beneficial for user traversal and speed improvement, and RRUs in multiple EDUs need to be as intertwined as possible, which is different from traditional multi-node clustering centralized collaborative processing; The article further proposes an modified genetic algorithms (GA), which simulates the performance of a new full spectrum cellfree wireless network with pilot pollution. The performance is compared with that of multi-RRU clustering group collaboration processing through Monte Carlo simulation.