A selective neighbour channels in Wi-Fi networks based on adaptive machine learning techniques

Abstract

Due to the growing commercial exploitation of WiFi-based technologies in recent years and the lack of solutions for effective WiFi orchestration, spectrum utilization and user performance are often sub-optimal. The integration of WiFi-based radio resource management (RRM) and radio environmental maps (REMs) may create a cost-effective Smart-WiFi solution that optimizes underlying spectrum utilisation and network performance. The REM enables effective use of radio environmental data such as device location, estimated channel models, real-time network interference levels, WiFi channel occupancies, and so on. In WiFi-related settings, this information may be used to make intelligent and optimum RRM decisions. This study provides a new REM-based RRM strategy for managing and optimizing commercial WiFi devices that makes use of the underlying radio environmental data. The research uses a commercially available platform to illustrate the suggested solution, which includes on-the-fly radio environmental data gathering and optimal WiFi RRM allocation. In comparison to traditional WiFi networks, the simulation study findings reveal that the proposed Smart-WiFi leverages considerable performance advantages for large-scale situations.