Cross-layer Optimization for Next Generation Wi-Fi

TL;DR

This paper develops a detailed cross-layer simulator for IEEE 802.11ac, evaluates MU-MIMO performance realistically, and proposes a PHY layer acknowledgment frame to reduce overhead in IEEE 802.11ah.

Contribution

It introduces the first detailed cross-layer simulator for IEEE 802.11ac and assesses MU-MIMO gains considering PHY and MAC interactions.

Findings

MU-MIMO provides less gain than ideal estimates suggest.

Channel sounding interval significantly affects MU-MIMO performance.

A short PHY acknowledgment frame reduces overhead in IEEE 802.11ah.

Abstract

From the initial 1997 specification to the undergoing IEEE 802.11ac standardization, a leap in throughput has been observed with every new generation. The expectations for next generations on issues like throughput, range, reliability, and power consumption are even higher. This is quite a challenge considering all the work already done. Cross-layer optimization of physical (PHY) and medium access control (MAC) layers can be an interesting exploration path for further enhancement. During this thesis we have studied cross-layer optimization techniques, with a focus on the IEEE 802.11ac standard. A new multichannel aggregation scheme involving cross-knowledge between PHY and MAC layers has been proposed to improve performance in collision-prone environments. We have shown that some functionalities directly involved PHY and MAC layers. An accurate modeling of both PHY and MAC mechanisms is thus needed to have a realistic characterization of such functionalities. A cross-layer simulator, compliant with IEEE 802.11ac specifications, has thus been implemented. To the best of our knowledge, this is the first simulator incorporating detailed PHY and MAC functionalities for the IEEE 802.11ac standard. The multiple-user multiple-input, multiple-output (MU-MIMO) technique, which is one of the main innovations of the IEEE 802.11ac, needs both PHY and MAC layer considerations. We have thus used the implemented cross-layer simulator to evaluate the performance of MU-MIMO and compared it with the single-user MIMO (SU-MIMO). The aim of these studies was to evaluate the 'real' gains of MU-MIMO solutions (accounting for PHY+MAC) over SU-MIMO solutions and not the generally accepted ones. The impact of the channel sounding interval has particularly been studied. Finally, we have proposed a short PHY layer version of acknowledgment frames for overhead reduction in IEEE 802.11ah communications.