Spectrum Coexistence, Network Dimensioning, and Cell-Free Architectures in 5G and 5G-Advanced Wireless Networks

Fifth-generation (5G) wireless networks introduce new architectural paradigms, spectrum usage models, and optimization challenges to support enhanced mobile broadband, massive machine-type communications, and ultra-reliable low-latency communications. This survey provides a comprehensive overview of key technologies and design challenges in 5G systems, with a focus on spectrum coexistence and interference management, network dimensioning and planning, cell-free massive MIMO architectures, fronthaul-aware user management, and power allocation strategies. Representative analytical, simulation-based, and optimization-driven approaches are reviewed, fundamental trade-offs are highlighted, and open research challenges relevant to 5G-Advanced and beyond are identified.

💡 Research Summary

This survey provides a comprehensive overview of the intertwined challenges and solutions that arise in fifth‑generation (5G) and emerging 5G‑Advanced wireless networks. It begins by describing the shift from traditional, cell‑centric architectures to highly flexible, software‑driven designs such as centralized RAN (C‑RAN), distributed RAN (D‑RAN), and cloud‑native virtualized networks. These architectures enable advanced features like network slicing and dynamic functional splits, but they also introduce stringent fronthaul capacity and latency constraints that must be accounted for in system‑level design.

The paper then examines spectrum coexistence across low‑band, mid‑band, and millimeter‑wave (mmWave) frequencies, covering licensed, unlicensed, and shared regulatory regimes. It introduces the key radio‑frequency metrics—adjacent channel leakage ratio (ACLR), adjacent channel selectivity (ACS), and the combined adjacent channel interference ratio (ACIR)—that quantify out‑of‑band emissions and receiver filtering imperfections. A detailed analysis of TDD frame alignment versus conflict shows how cross‑link interference (BS‑to‑BS or BS‑to‑UE) can dominate when uplink and downlink subframes are misaligned. The authors illustrate these effects with concrete CBRS (3.5 GHz) case studies, highlighting the impact of inter‑site distance (ISD) and indoor‑outdoor penetration loss on interference levels and SINR.

Next, the survey turns to network dimensioning and planning using stochastic geometry. By modeling user locations as a Poisson point process (PPP) and incorporating service‑specific traffic models (eMBB, mMTC, URLLC), the authors derive analytical expressions for the overload probability of time‑frequency resources. These expressions enable designers to size the number of subcarriers or resource blocks so that the probability of resource shortage stays below a target threshold (e.g., 1 %). The analysis also discusses how heterogeneous QoS requirements can be accommodated through multi‑class Erlang‑B formulations and spatial load balancing.

The fifth section focuses on cell‑free massive MIMO (also called user‑centric) architectures. In this paradigm, many distributed radio units cooperate to serve each user as if it were at the center of a “virtual cell.” The paper reviews the trade‑offs between centralized processing gains and fronthaul limitations, proposing partial cooperation, compressed front‑haul transmission, and user clustering as ways to preserve spectral efficiency when fronthaul bandwidth is scarce. It also points out the increased overhead for channel state information acquisition and coordination compared with traditional cell‑centric designs.

Section six addresses power allocation and resource optimization under the combined constraints of fronthaul capacity, ACIR‑driven interference, and diverse QoS demands. The authors formulate a multi‑objective optimization problem and compare solution techniques such as Lagrangian dual decomposition, the alternating direction method of multipliers (ADMM), and reinforcement‑learning‑based policies. Simulation results demonstrate that adaptive power control that reacts to real‑time fronthaul load can significantly improve overall system throughput and user fairness.

Finally, the survey identifies open research directions: the need for integrated simulation‑testbeds that jointly model RF emissions, stochastic spatial traffic, and fronthaul dynamics; AI‑driven, real‑time spectrum sharing mechanisms; joint design of cell‑free massive MIMO and fronthaul‑aware scheduling for ultra‑reliable low‑latency communications; and the extension of these concepts to the upcoming 5G‑Advanced era, where tighter integration of radio, transport, and compute resources will be mandatory. In summary, the paper argues that successful 5G‑Advanced deployments require a holistic approach that simultaneously considers spectrum coexistence, spatial network dimensioning, user‑centric architecture, and power/resource optimization.