Evaluation of Channel Assignment Algorithms in a Dense Real World WLAN

Abstract

This paper addresses the problem of Access Point (AP) channel assignment in dense IEEE 802.11 Wireless Local Area Networks (WLANs) implemented in a real world scenario, based on a housing complex located in Oslo, Norway. Currently, the APs composing this housing complex are centrally configured through a channel selection approach based on a genetic algorithm that aims to minimize the cumulative interference experienced by each AP. In this work, we present the performance of an alternative channel assignment algorithm in comparison to the existing genetic one. More specifically, the algorithm investigated in this work aims to minimize a networkwide parameter called interference impact, which represents the interference caused by an AP to all the other APs in the neighbourhood. Moreover, the algorithm is implemented using the spectrum programming architecture Wi-5 based on Software-Defined Networking (SDN). Although the benefits of this algorithm have been demonstrated in simulated environments, this work presents its first evaluation in a dense real world scenario. The performance analysis illustrates the important gains obtained in terms of the data transmissions quality through the proposed algorithm compared against the AP channel selection approach currently implemented in the considered scenario.

On Optimizing WiFi RSSI and Channel Assignment using Genetic Algorithm for WiFi Tuning

“…However, even with a well-designed WLAN, some issues are exacerbated such as interference and growth in network size from new nodes being introduced into the system exacerbated. Interference issues can be resolved in the physical layer by applying the appropriate channel selection strategies proposed in [6][7][8][9][10]. It can also be mitigated at the medium access level by applying a well-designed carrier sensing technique [11].…”

“…During the WLAN design process, many research works have adopted GA for Wi-Fi planning and design to provide optimal wireless access point placement under certain conditions and constraints. In [1,6], node placement was jointly investigated with path-loss from indoor obstacles. In [2], GA was adopted to provide a cost-effective WLAN in terms of coverage and capacity under budget constraints.…”

“…• Defining Population: Since the values to be adjusted during Wi-Fi tuning include transmission power and the frequency channel, each gene represents a set of random values between the power level and channel number. Based on the setting of Wi-Fi access points deployed at the 2.4 GHz frequency band, power values ranging from 1 and 6 were chosen, represented by P t ∈ [1,6] while a channel number C h was picked from 1, 6, and 11, represented by C h ∈ {1, 6, 11}.…”

Deep Reinforcement Learning-Based Channel Allocation for Wireless LANs With Graph Convolutional Networks

“…This algorithm achieved an improvement in the throughput by operating the bandwidth of each channel without overlapping. Raschellà et al [7] evaluated a channel assignment algorithm. The objective was to minimize the parameters that represented the interference among the APs.…”

Joint Access Point Placement and Power-Channel-Resource-Unit Assignment for 802.11ax-Based Dense WiFi with QoS Requirements

“…Lei et al [32] took into account both the channel overlapping degree and the distribution of stations and designed a channel selection scheme to improve the network throughput. Raschella et al [33] presented a channel assignment algorithm to minimize the interference impact and improve the data transmission quality. Moreover, to improve both throughput and fairness, Oh et al [34] and Kasasbeh et al [35] studied the co-channel interference problem and proposed efficient channel assignment methods to eliminate the interference between adjacent APs.…”