Massive MIMO wireless networks: Theory and methods

Project leader

Prof Jonathon Chambers

Dates

April 2015 to April 2018

Sponsors

EPSRC (EP/M015475)

Description

Spectrum is a precious but scarce natural resource. In the UK, Ofcom will free up the analogue TV spectrum at 800MHz for 4G, together with the available 2.6GHz band. This has already raised £2.34 billion for the national purse. Each month, Britons already consume 20m gigabytes of data on the move. This is mainly due to user engagement with video and TV while on the move. By 2020, system capacity will need a 1000-times increase to avoid mobile networks grinding to a halt.

Interference and fading limit spectral efficiency. Maximising efficiency for wireless networks, including 4G, is thus a major issue. An emerging idea is that of a massive MIMO antenna system. This has received serious consideration by vendors and operators. It is already championed by Alcatel-Lucent. This technology has the potential to unlock the issue of spectrum scarcity. It would enable simultaneous access of tens or hundreds of terminals in the same time-frequency resource. This would provide enormous enhancements to spectrum usage.

A MIMO is a wireless network that can transmit and receive more than one signal at the same time, over the same channel. It is a multiple-input multiple-output system. A Massive MIMO has many antennas. They may number into the hundreds.

Massive MIMO technology has yet to meet its potential. There are various challenges, such as:

• channel estimation and acquisition due to pilot contamination
• fast spatial-temporal variations in signal power and autonomous resource allocation
• simultaneous access by a large number of users

This project will tackle these fundamental challenges. To do this, we will advance aspects of information theory, estimation theory and network optimisation. We will:

• model massive MIMO channels underpinned by heterogeneous correlation structures
• carry out information theoretic analysis using random matrix theory through shrinkage estimators
• develop a precoder design which is robust in the presence of channel estimation errors
• develop a novel channel estimation technique in the presence of severe pilot contamination
• explore game theoretic algorithms for autonomous resource allocation and pilot assignments

All the concepts and algorithms developed will be integrated. We will assess the radio link layer performance using a simulation reference system based on LTE-Advanced standards and its evolution towards 5G. Industrial partners will be engaged throughout the project to ensure the industrial relevance of our work.