Coverage Analysis and Scaling Laws of Ultra-Dense Networks

TL;DR

This paper introduces a comprehensive framework for analyzing ultra-dense wireless networks, revealing new insights into coverage probability scaling laws and the benefits of multi-antenna deployment at higher frequencies.

Contribution

It develops a unified analytical approach for coverage analysis in ultra-dense networks considering general channel models and demonstrates that increasing antenna height is not necessary for densification success.

Findings

Coverage probability can be maintained with densification by increasing antenna height and frequency.

Deploying multiple antennas and higher frequencies can improve coverage without lowering antenna height.

Simulation results confirm the theoretical predictions across various propagation environments.

Abstract

In this paper, we develop an innovative approach to quantitatively characterize the performance of ultra-dense wireless networks in a plethora of propagation environments. The proposed framework has the potential of significantly simplifying the cumbersome procedure of analyzing the coverage probability and allowing the remarkable unification of single- and multi-antenna networks through compact representations. By harnessing this key feature, we develop a novel statistical machinery to study the scaling laws of wireless network densification considering general channel power distributions including the entire space of multipath and shadowing models as well as associated beamforming gain due to the use of multiple antenna. We further formulate the relationship between network density, antenna height, antenna array seize and carrier frequency showing how the coverage probability can be maintained with ultra-densification. From a system design perspective, we present a new innovative theoretical discovery stipulating that if multiple antenna BS are deployed and moved to higher frequencies, then monotonically increasing the coverage probability by means of ultra-densification is possible, and this without lowering the antenna height. Such findings are completely different from the conclusions in existing works, who suggest to lower the BS height as to leverage the potential of network densification. Simulation results substantiate performance trends leveraging network densification and antenna deployment and configuration against path loss models and signal-to-noise plus interference (SINR) thresholds.