Fairness and Applications’ transport protocol aware frame aggregation using programmable WLANs

Abstract

Internet of Things (IoT) is emerging as a prominent technology, and a broad range of IoT applications are deployed using advanced Wi-Fi technologies like 802.11n, ac, and ax. Wi-Fi technologies with the support of frame aggregation techniques are highly useful for throughput hungry applications such as e-education, and e-conference to improve wireless channel utilization, and throughput of Wi-Fi stations (STAs). In existing works by considering mobility, and channel conditions, various frame aggregation schemes were investigated at STA level for finding optimal aggregation size. But, the local frame aggregation schemes could lead to unfairness issues among STAs with poor signal strength. Moreover, IoT applications such as e-education, e-conferences can be implemented using either transmission control protocol (TCP) or user datagram protocol (UDP). Hence, enabling frame aggregation in Wi-Fi networks can lead to poor throughput, and fairness issues for TCP STAs from UDP STAs transmitting at higher rates due to lack of flow control and congestion control. Having enterprise wireless local area networks (WLANs) with programmable access points (Pro-APs), it is possible to control fairness issues using global knowledge of connected STAs, and their channel, and traffic details. In this work, we investigate major reasons for fairness issues, and propose a Fairness and Applications’ transport protocol aware frame aggregation (FAFA) scheme using Pro-APs. In our work, we design and implement the FAFA scheme at Pro-AP to solve fairness issues among STAs connected to it. The FAFA scheme is evaluated against existing schemes using extensive simulations with the Network Simulator-3 (NS-3). In various test scenarios, the results corroborate that the proposed FAFA is able to maintain Jain’s Fairness Index (JFI) around 0.9 for STAs in the network, whereas existing works are able to maintain JFI 0.3 to 0.7 only in terms of throughput, delay, and jitter. © 2022, The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.