The performance of a WLAN station heavily depends on the selection of an Access Point (AP). A wrong AP selection might yield to a bad Quality of Experience (QoE) and therefore unsatisfied users. The current IEEE 802.11 association approach based the association decision on the strongest Received Signal Strength Indicator (RSSI). Since this value does not represent the real features or capacities of the APs, this decision does not lead to the selection of the most suitable AP for a specific user.
The project MetroQoE introduces QoE-based mechanisms, in order to guarantee the expected QoE when the users are accessing services and applications in Wireless Metropolitan Area Networks (WMN).
In WMSN scenarios, the lack of a centralized management, and the existence of many administrative domains make it difficult to guarantee the QoE that users wanted. The first MetroQoE mechanism is able to enhance the association decision in WLAN through the use of additional metrics in the selection of the AP that provides the best connectivity index for a specific user.
The second MetroQoE mechanism is called Gossipmule. Under the current scenario, it is not possible that a user has the whole image of the network topology and therefore of the performance of the APs. Gossipmule uses mobile crowdsensing between the wireless stations to collect and disseminate information regarding the network. The station uses this information in order to have a more assertive association decision.
The third mechanism is a QoE estimation tool. The goal of the estimator is to sense different QoE factors such as delay, packet loss, RSSI, application requests, battery consumption, among others, in order to estimate a QoE value that reflects the user satisfaction. Additionally, the tool is able to made a customized estimation since it takes into account the user preferences using a Facial Expression Recognition component to evaluate the user satisfaction.