Quality of Service (QoS) and Quality of experience (QoE)

The International Telecommunication Union (ITU) characterizes Quality of Service (QoS) as "the collective attributes of a telecommunication service that influence its capacity to satisfy the user's expressed and implicit needs" (source). QoS is primarily assessed through objective metrics such as bandwidth, packet loss ratio, latency, etc. However, QoS measurement may vary depending on the location within the network where it is evaluated.

The ITU defines Quality of Experience (QoE) as "the user's level of satisfaction or frustration with an application or service" (source). While QoE is linked to QoS, it also relies on subjective factors like the user's mood, the device used, and time of day. There is an ongoing discussion within the telecommunications industry about the best way to quantify QoE. Several models exist, but the connection between QoE and QoS can be simplified as follows:

$QoE=QoS+subjective factors$

Considering that users prioritize QoE, which influences their decisions to lodge complaints with their service provider or switch providers, it's critical to establish a method for measuring this in a system that monitors internet performance. We propose several principles that must be fulfilled to ensure QoS data measurements align as closely as possible with QoE:

• Extensive Data Coverage: The user experience should be assessed over a wide geographical area and with enough independent data points to support statistical analysis.

• Customer-Level Measurement: Testing should occur at network points where the user interacts with the internet, such as the user's home, office, coffee shop, etc.

• Independent Unbiased Data: Test results should be impartial, and analysis should be conducted independently of industry stakeholders to mitigate conflict of interest concerns.

• Data Sharing: The collected data should be shared more openly than is currently standard in the industry, to prevent duplicate efforts and establish a single source of truth.

• Actionable Insights: Data should be analysed to identify major events and ongoing service-impacting issues. The collected data and insights should be available to relevant stakeholders for necessary actions.