What Should Future Wireless Network Architectures Be?

Lu Yang, Ping Li, Miaomiao Dong, Bo Bai, Dmitry Zaporozhets, Xiang Chen, Wei Han, Baochun LiarXiv 2021 · OpenReview

Abstract

The accelerated convergence of digital and real-world lifestyles has imposed unprecedented demands on today's wireless network architectures, as it is highly desirable for such architectures to support wireless devices everywhere with high capacity and minimal signaling overhead. Conventional architectures, such as cellular architectures, are not able to satisfy these requirements simultaneously, and are thus no longer suitable for the future era. In this paper, we propose a capacity-centric ($C^2$) architecture for future wireless communication networks. It is designed based on the principles of maximizing the number of non-overlapping clusters with the average cluster capacity guaranteed to be higher than a certain threshold, and thus provides a flexible way to balance the capacity requirement against the signaling overhead. Our analytical results reveal that $C^2$ has superior generality, wherein both the cellular and the fully coordinated architectures can be viewed as its extreme cases. Simulation results show that the average capacity of $C^2$ is at least three times higher compared to that of the cellular architecture. More importantly, different from the widely adopted conventional wisdom that base-station distributions dominate architecture designs, we find that the $C^2$ architecture is independent of base-station distributions, and instead user-side information should be the focus in future wireless network architecture designs.

The Scalability Crisis in Wireless Design

As we move toward the 6G era, current wireless architectures are reaching their fundamental limits. Traditional cellular networks are "base-station (BS) centric," defining service areas based on hardware locations. This leads to the persistent "cell-edge problem," where users at boundaries experience poor signal quality due to inter-cell interference.

To address this, researchers have explored Coordinated Multi-Point (CoMP) and fully coordinated (cell-free) networks. While these improve signal quality by allowing multiple base stations to work together, they face a significant scalability crisis. In fully coordinated networks, the signaling overhead—the data required for base stations to coordinate—grows so rapidly that it can overwhelm the network's capacity. The challenge lies in finding a middle ground: achieving high capacity everywhere while maintaining manageable signaling overhead.

The Capacity-Centric ($C^2$) Solution

Researchers propose the Capacity-Centric ($C^2$) architecture as a new paradigm. Instead of fixed, hardware-defined cells, the network is decomposed into non-overlapping clusters based on capacity requirements.

Toy example of the $C^2$ coverage map showing non-overlapping clusters of base stations and users.

The core of this method involves three pillars:

  1. The $C^2$ Optimization Principle: The goal is to maximize the number of clusters ($M$) while ensuring the average capacity for every cluster remains above a predefined threshold ($C_{Th}$). Maximizing clusters is critical because more clusters mean smaller coordination groups, which keeps signaling overhead low.
  2. Average Cluster Capacity Theorem: To avoid grueling high-dimensional matrix math, the authors derived a simplified theorem using continuous density functions for base stations and users. This allows for determining the optimal architecture without needing the exact coordinates of every antenna.
  3. $C^2$ Tessellation: Moving beyond the standard "Voronoi Tessellation" used in cellular networks, $C^2$ Tessellation determines cluster boundaries by capacity requirements rather than simple proximity to a tower.

A Universal Framework for Network Design

A key finding of this research is that $C^2$ is a universal model. As shown in the comparison below, traditional cellular networks and fully coordinated networks are actually extreme cases of the $C^2$ spectrum (where $M=L$ and $M=1$, respectively).

Comparison of wireless network architectures across design principles, scalability, and overhead.

Key Insights and Performance Results

  • User-Side Dominance: Surprisingly, the optimal network architecture is independent of base-station distribution as long as user density is constant. This suggests that future networks should be designed based on where users are located, rather than where hardware is placed.
  • Robustness: Because $C^2$ focuses on capacity thresholds, the architecture remains stable even as base stations are added or moved, provided the user distribution is consistent.
  • Superior Capacity: Simulation results demonstrate that $C^2$ provides an average capacity per base station that is at least three times higher than traditional cellular designs and outperforms CoMP by 17% to 21%.
  • Scalability: By maximizing the number of clusters, $C^2$ maintains low signaling overhead comparable to cellular networks while achieving the high capacity typically reserved for unscalable, fully coordinated systems.

Limitations and Future Work

While the theoretical framework is robust, real-world implementation faces hurdles. The current model relies on asymptotic assumptions (large numbers of BSs and users), which may vary in sparse rural areas. Additionally, future work must address dynamic user movement to allow clusters to shift in real-time with traffic flow. Finally, the sophisticated software-defined networking required for flexible clustering will necessitate advanced backhaul infrastructure beyond current 5G deployments.

Summary updated

Abstract

The accelerated convergence of digital and real-world lifestyles has imposed unprecedented demands on today's wireless network architectures, as it is highly desirable for such architectures to support wireless devices everywhere with high capacity and minimal signaling overhead. Conventional architectures, such as cellular architectures, are not able to satisfy these requirements simultaneously, and are thus no longer suitable for the future era. In this paper, we propose a capacity-centric (C$^2$) architecture for future wireless communication networks. It is designed based on the principles of maximizing the number of non-overlapping clusters with the average cluster capacity guaranteed to be higher than a certain threshold, and thus provides a flexible way to balance the capacity requirement against the signaling overhead. Our analytical results reveal that C$^2$ has superior generality, wherein both the cellular and the fully coordinated architectures can be viewed as its extreme cases. Simulation results show that the average capacity of C$^2$ is at least three times higher compared to that of the cellular architecture. More importantly, different from the widely adopted conventional wisdom that base-station distributions dominate architecture designs, we find that the C$^2$ architecture is independent of base-station distributions, and instead user-side information should be the focus in future wireless network architecture designs.
BibTeX
@misc{yang2021what,
  author = {Lu Yang and Ping Li and Miaomiao Dong and Bo Bai and Dmitry Zaporozhets and Xiang Chen and Wei Han and Baochun Li},
  title = {{What Should Future Wireless Network Architectures Be?}},
  year = {2021},
  eprint = {2110.03157},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2110.03157},
}
Figure