This paper presents a Failure-Aware Access Point Selection (FAAS) method aimed at improving hardware resilience in cell-free massive MIMO (CF-mMIMO) networks. FAAS selects APs for each user by jointly considering channel strength and the failure probability of each AP. A tunable parameter \(α\in [0,1]\) scales these failure probabilities to model different levels of network stress. We evaluate resilience using two key metrics: the minimum-user spectral efficiency, which captures worst-case user performance, and the outage probability, defined as the fraction of users left without any active APs. Simulation results show that FAAS maintains significantly better performance under failure conditions compared to failure-agnostic clustering. At high failure levels, FAAS reduces outage by over 85\% and improves worst-case user rates. These results confirm that FAAS is a practical and efficient solution for building more reliable CF-mMIMO networks.
Cell-free massive multiple-input multiple-output (CF-mMIMO) has emerged as a leading architecture for beyond-5G/6G wireless networks, generalizing classical massive MIMO into a distributed, cell-less paradigm [1], [2], [3]. By deploying a large number of distributed access points (APs) that jointly serve all users, CF-mMIMO leverages macro-diversity, mitigates inter-cell interference, and ensures uniformly high data rates [4], [5]. Unlike traditional cellular systems, CF-mMIMO significantly reduces cell boundaries and associated edge effects, enabling consistent quality of service (QoS) and ultra-reliable links [6]. At the same time, the large number of distributed APs supports efficient MU-MIMO transmission, allowing the system to exploit spatial multiplexing gains that boost spectral efficiency while maintaining uniform coverage and link reliability [7], [8], [9].
Despite these touted reliability benefits, the resilience of CF-mMIMO networks in the face of hardware failures has received surprisingly limited attention, in contrast to recent works highlighting resilience-by-design as a crucial paradigm for ensuring robust 6G communication networks [10]. In practice, however, AP hardware can malfunction
The work presented in this paper was funded by the UK Department for Science, Innovation and Technology under project YO-RAN.
or fail (e.g. due to power outages, equipment faults, or maintenance issues), which poses a serious challenge to any distributed antenna system. Conventional cellular networks suffer outages when a base station fails, but a distributed CF network could be more fault-tolerant by design. For example, recent architectural proposals like the “radio stripes” concept suggest that node failures can be tolerated via internal routing mechanisms, thereby improving network robustness. This assumption holds particularly in high-density deployments, where overlapping AP coverage ensures that the failure of a few nodes has only a marginal impact on system-wide performance due to the inherent spatial redundancy of CF-mMIMO [11].
While resilience in CF-mMIMO is often discussed qualitatively [12], only a few works offer detailed analysis. Sadreddini et al. [13] use Markov models to show how limited fronthaul capacity and long routing paths can disconnect UEs or degrade SINR. Weinberger et al. [14] demonstrate that RIS can passively enhance resilience by providing alternative paths, even without optimized phase settings. Elkeshawy et al. [15] propose a data-driven activity detector at the central processing unit (CPU) that remains accurate under impairments, highlighting robustness against practical impairments. In addition, [16] addresses hardware nonlinearity by modeling PA distortions and optimizing user association and power control to mitigate their effects. Overall, there remains a significant theoretical gap in understanding how probabilistic AP failures influence CF-mMIMO performance and what can be done to design resilient cell-free networks.
The theoretical novelty of this work lies in integrating hardware failure resilience into CF-mMIMO for the first time in a systematic way. Rather than deriving closedform analytical expressions for performance under failures, which remain highly complex due to combinatorial failure patterns, we propose a tractable modeling framework that incorporates probabilistic AP failures into AP selection and evaluation. By explicitly defining failure-aware user-AP associations and resilience metrics, our work bridges the gap between purely qualitative discussions of resilience and quantitative system-level analysis. We show that even under moderate AP failure rates, intelligent AP selection can preserve much of the system’s arXiv:2602.16546v1 [eess.SP] 18 Feb 2026 spectral efficiency, whereas traditional failure-agnostic approaches suffer more pronounced degradation. The proposed Failure-Aware AP Selection (FAAS) methodology offers a blueprint for making CF architectures failureaware: network controllers can use failure probabilities (obtained from hardware health monitoring or historical data) to optimize user-AP associations proactively.
To validate these claims, we evaluate FAAS under various probabilistic failure scenarios and compare the resulting user rates and fairness against baseline schemes without failure awareness. The introduction of resilience into CF-mMIMO, as pursued in this work, opens a new research direction to ensure that the next generation of CF-mMIMO networks can deliver on their promise of ubiquitous, reliable connectivity even in the presence of inevitable hardware failures. In the following, we detail the system model and assumptions, then present the FAAS strategy and its theoretical performance analysis under AP failure conditions.
As illustrated in Fig. 1, APs in a CF-mMIMO network are susceptible to hardware failures caused by power loss, component degradation, or synchronization issues. If each AP fails independently with a proba
This content is AI-processed based on open access ArXiv data.