Energy Efficiency of Distributed Signal Processing in Wireless Networks: A Cross-Layer Analysis

TL;DR

This paper presents a cross-layer framework to evaluate how different levels of distributed signal processing impact energy efficiency in dense wireless networks, highlighting hybrid processing as most efficient.

Contribution

It introduces a novel framework for comparing energy efficiency across hybrid, centralized, and distributed signal processing architectures in wireless networks.

Findings

Hybrid processing significantly improves energy efficiency.

Distributed processing can lead to higher energy consumption.

Centralized processing is less efficient than hybrid in practical scenarios.

Abstract

In order to meet the growing mobile data demand, future wireless networks will be equipped with a multitude of access points (APs). Besides the important implications for the energy consumption, the trend towards densification requires the development of decentralized and sustainable radio resource management techniques. It is critically important to understand how the distribution of signal processing operations affects the energy efficiency of wireless networks. In this paper, we provide a cross-layer framework to evaluate and compare the energy efficiency of wireless networks under different levels of distribution of the signal processing load: (i) hybrid, where the signal processing operations are shared between nodes and APs, (ii) centralized, where signal processing is entirely implemented at the APs, and (iii) fully distributed, where all operations are performed by the nodes. We find that in practical wireless networks, hybrid signal processing exhibits a significant energy efficiency gain over both centralized and fully distributed approaches.