Case For Static AMSDU Aggregation in WLANs

TL;DR

This paper argues that in WLANs, maintaining a simple static AMSDU aggregation approach is preferable to complex adaptive schemes, as traditional error control mechanisms can effectively manage packet error rates.

Contribution

The study demonstrates that simple static AMSDU aggregation, combined with traditional error control, is more practical than adaptive aggregation mechanisms.

Findings

Static AMSDU aggregation is sufficient for error management.

Complex adaptive mechanisms offer limited practical benefits.

Traditional error control mechanisms effectively influence packet error rates.

Abstract

Frame aggregation is a mechanism by which multiple frames are combined into a single transmission unit over the air. Frames aggregated at the AMSDU level use a common CRC check to enforce integrity. For longer aggregated AMSDU frames, the packet error rate increases significantly for the same bit error rate. Hence, multiple studies have proposed doing AMSDU aggregation adaptively based on the error rate. This study evaluates if there is a \emph{practical} advantage in doing adaptive AMSDU aggregation based on the link bit error rate. Evaluations on a model show that instead of implementing a complex adaptive AMSDU frame aggregation mechanism which impact queuing and other implementation aspects, it is easier to influence packet error rate with traditional mechanisms while keeping the AMSDU aggregation logic simple.