Spatial Resources Optimization in Distributed MIMO Networks with Limited Data Sharing

TL;DR

This paper addresses the challenge of optimizing spatial resources in distributed MIMO networks with limited data sharing, proposing exact and approximate solutions to improve capacity and QoS.

Contribution

It introduces a novel optimal and an approximation algorithm for resource allocation in distributed MIMO networks with limited data sharing, outperforming existing methods.

Findings

Exact optimal solution for small networks

Polynomial-time approximation algorithm for larger networks

High QoS levels achieved with reasonable complexity

Abstract

Wireless access through a large distributed network of low-complexity infrastructure nodes empowered with cooperation and coordination capabilities, is an emerging radio architecture, candidate to deal with the mobile data capacity crunch. In the 3GPP evolutionary path, this is known as the Cloud-RAN paradigm for future radio. In such a complex network, distributed MIMO resources optimization is of paramount importance, in order to achieve capacity scaling. In this paper, we investigate efficient strategies towards optimizing the pairing of access nodes with users as well as linear precoding designs for providing fair QoS experience across the whole network, when data sharing is limited due to complexity and overhead constraints. We propose a method for obtaining the exact optimal spatial resources allocation solution which can be applied in networks of limited scale, as well as an approximation algorithm with bounded polynomial complexity which can be used in larger networks. The particular algorithm outperforms existing user-oriented clustering techniques and achieves quite high quality-of-service levels with reasonable complexity.