QoE Fairness and Resilience Strategies for Access Point Failures in Smart Greenhouses

Abstract

Networked Control Systems (NCSs) are often used in smart greenhouses for remote monitoring and control of various environmental factors. The crop yield is highly dependent on the proper monitoring and control of said factors. Often these NCSs utilize wireless sensors communicating with a controller. In this paper, the network involves Wi-Fi Access Points (APs) linking the sensors to the switched Ethernet network, on which the controller resides. The greenhouse is split into multiple cells, each with a dedicated AP. This research work revolves around the impact of AP failures. Spanning various failure scenarios, sensors may redirect their traffic to other functioning cell APs. In this paper, Riverbed Modeler is used to simulate a comprehensive and representative set of failure scenarios, with different traffic redirection schemes. The simulations produce measures of Packet Loss Rate (PLR), which is factored into a measure of Quality of Experience (QoE). Finally, QoE Fairness is calculated for the various redirection schemes, to identify the optimal redirection scheme, for each failure scenario. This serves as a case study for traffic redirection schemes in similar scenarios, and in a broader sense for the application of QoE Fairness in fault-tolerance planning.