Improved Capacity Scaling in Wireless Networks With Infrastructure

Won-Yong Shin, Sang-Woon Jeon, Natasha Devroye, Mai Vu, Sae-Young Chung, Yong Hoon Lee, Vahid Tarokh

Abstract

This paper analyzes the impact and benefits of infrastructure support in improving the throughput scaling in networks of $n$ randomly located wireless nodes. The infrastructure uses multi-antenna base stations (BSs), in which the number of BSs and the number of antennas at each BS can scale at arbitrary rates relative to $n$. Under the model, capacity scaling laws are analyzed for both dense and extended networks. Two BS-based routing schemes are first introduced in this study: an infrastructure-supported single-hop (ISH) routing protocol with multiple-access uplink and broadcast downlink and an infrastructure-supported multi-hop (IMH) routing protocol. Then, their achievable throughput scalings are analyzed. These schemes are compared against two conventional schemes without BSs: the multi-hop (MH) transmission and hierarchical cooperation (HC) schemes. It is shown that a linear throughput scaling is achieved in dense networks, as in the case without help of BSs. In contrast, the proposed BS-based routing schemes can, under realistic network conditions, improve the throughput scaling significantly in extended networks. The gain comes from the following advantages of these BS-based protocols. First, more nodes can transmit simultaneously in the proposed scheme than in the MH scheme if the number of BSs and the number of antennas are large enough. Second, by improving the long-distance signal-to-noise ratio (SNR), the received signal power can be larger than that of the HC, enabling a better throughput scaling under extended networks. Furthermore, by deriving the corresponding information-theoretic cut-set upper bounds, it is shown under extended networks that a combination of four schemes IMH, ISH, MH, and HC is order-optimal in all operating regimes.

Introduction

In the study of wireless information theory, a seminal result by Gupta and Kumar (2000) established a sobering reality for ad hoc networks: as the number of nodes ($n$) grows, the throughput available to each individual node shrinks toward zero. While subsequent research into "Hierarchical Cooperation" (HC) suggested that near-constant scaling was theoretically possible, these gains were largely confined to "dense" networks where nodes are packed into a fixed area. In "extended" networks—where the physical area grows alongside the number of nodes—signal power decays rapidly over long distances (path loss), causing HC to struggle.

This paper addresses a fundamental architectural question: To what extent can infrastructure, in the form of multi-antenna base stations (BSs), overcome these scaling bottlenecks? By introducing $m$ base stations, each equipped with $l$ antennas, the authors provide a rigorous mathematical framework to determine when infrastructure is a necessity and when it is redundant.

Network Model and Methodology

The researchers utilize a stochastic geometry model where $n$ nodes are distributed uniformly at random in a square area. They evaluate two distinct network types:

  1. Dense Networks: The network area remains fixed regardless of $n$.
  2. Extended Networks: The network area grows linearly with $n$, maintaining a constant node density.

The study employs a combination of constructive protocol design and information-theoretic cut-set upper bounds to determine the "scaling exponent"—a measure of how total network capacity grows relative to the number of nodes. To achieve this, two novel infrastructure-supported protocols are introduced:

  • Infrastructure-Supported Single-Hop (ISH): Nodes transmit directly to the nearest base station (uplink), and base stations transmit directly to destinations (downlink). This protocol leverages the Multiple-Input Multiple-Output (MIMO) capabilities of the base stations to manage interference.
  • Infrastructure-Supported Multi-Hop (IMH): Nodes utilize short-range multi-hop relays to reach a base station. This is specifically designed for extended networks to combat high path loss by keeping individual wireless hops short and power-efficient.

These protocols are benchmarked against standard Multi-Hop (MH) and Hierarchical Cooperation (HC) schemes.

Key Insights and Results

The Dense vs. Extended Dichotomy

The research reveals a stark contrast in the utility of infrastructure based on network density. In dense networks, infrastructure is surprisingly unhelpful. Because nodes are physically close to one another, the HC scheme is already near-optimal, achieving a scaling exponent of 1 (linear growth). Adding base stations does not fundamentally alter this scaling law.

In extended networks, however, infrastructure becomes the primary driver of performance. These networks are inherently power-limited. Base stations provide two critical advantages:

  • Power Gain: BSs act as high-power hubs that bridge long distances, preventing the signal-to-noise ratio (SNR) collapse that occurs in pure node-to-node communication.
  • Degrees of Freedom (DoF) Gain: Multi-antenna base stations allow a significantly higher number of simultaneous transmissions compared to traditional multi-hop schemes, provided the number of BSs and antennas is sufficiently large.

Order-Optimality and Phase Transitions

The authors prove that no single strategy is optimal across all conditions. Instead, they demonstrate that a hybrid approach—switching between IMH, ISH, MH, and HC depending on the path-loss exponent ($\alpha$) and infrastructure density ($m$ and $l$)—is order-optimal. This means that this combination achieves the best possible scaling law allowed by the laws of physics and information theory.

Limitations and Considerations

While the theoretical gains are significant, the paper identifies several practical constraints:

  • Backhaul Requirements: The model assumes base stations are connected by a high-capacity, wired backhaul. The paper quantifies the necessary backhaul capacity required to support these wireless gains, noting that the backhaul itself can become a bottleneck.
  • Complexity and Latency: Infrastructure-based routing, particularly the IMH protocol, introduces additional latency and requires sophisticated channel estimation. There is a clear trade-off between maximizing throughput and minimizing the "price" of system complexity.
  • CSI Assumptions: The analysis assumes perfect Channel State Information (CSI) at the receivers. In highly mobile environments, the overhead of maintaining accurate CSI could impact the achievable scaling laws.

Conclusion

This work provides a definitive map for the integration of infrastructure in wireless networks. While ad hoc cooperation suffices for dense deployments, multi-antenna base stations are essential for maintaining capacity in large-scale extended networks. By identifying the specific regimes where ISH and IMH protocols outperform traditional methods, this research offers a theoretical foundation for designing the next generation of hybrid wireless architectures.

Summary updated

Abstract

This paper analyzes the impact and benefits of infrastructure support in improving the throughput scaling in networks of $n$ randomly located wireless nodes. The infrastructure uses multi-antenna base stations (BSs), in which the number of BSs and the number of antennas at each BS can scale at arbitrary rates relative to $n$. Under the model, capacity scaling laws are analyzed for both dense and extended networks. Two BS-based routing schemes are first introduced in this study: an infrastructure-supported single-hop (ISH) routing protocol with multiple-access uplink and broadcast downlink and an infrastructure-supported multi-hop (IMH) routing protocol. Then, their achievable throughput scalings are analyzed. These schemes are compared against two conventional schemes without BSs: the multi-hop (MH) transmission and hierarchical cooperation (HC) schemes. It is shown that a linear throughput scaling is achieved in dense networks, as in the case without help of BSs. In contrast, the proposed BS-based routing schemes can, under realistic network conditions, improve the throughput scaling significantly in extended networks. The gain comes from the following advantages of these BS-based protocols. First, more nodes can transmit simultaneously in the proposed scheme than in the MH scheme if the number of BSs and the number of antennas are large enough. Second, by improving the long-distance signal-to-noise ratio (SNR), the received signal power can be larger than that of the HC, enabling a better throughput scaling under extended networks. Furthermore, by deriving the corresponding information-theoretic cut-set upper bounds, it is shown under extended networks that a combination of four schemes IMH, ISH, MH, and HC is order-optimal in all operating regimes.
BibTeX
@article{shin2011improved,
  author = {Won-Yong Shin and Sang-Woon Jeon and Natasha Devroye and Mai Vu and Sae-Young Chung and Yong Hoon Lee and Vahid Tarokh},
  title = {{Improved Capacity Scaling in Wireless Networks With Infrastructure}},
  journal = {IEEE Trans. Information Theory},
  year = {2011},
  url = {https://openreview.net/forum?id=7Q71AJt2Ugk},
}
Figure