During the course of my PhD I came to understand Quality of Experience (QoE). This is the term that often refers to how users experience a service. Whether it is in a restaurant, making a phone call, using a company network to complete daily tasks, play games, watch iPlayer or YouTube. Our experience is subjective, it is an emotional response. How we feel at any given time about what we are experiencing, dictates what we think about the service received. Our experience usually deteriorates when resources become stretched and waiting times increase; this could be due to lack of staff serving, or in the case of networks when links start to become congested. When networks perform badly, packets can be lost and delayed. Data loss and delays are tangible, they can be measured, whereas QoE is subjective to the user. The challenge faced by network operators is translating how you feel about the service and relating it to the level of loss and delay on the network. Poor QoE has an economic impact, users with bad experience, will move service providers, call the support centre, and in the work place, suffer from lack of productivity and stress.
One solution often adopted by network operators is to run network links at moderate capacity, for example 50%, adding more capacity as the carried traffic exceeds this threshold. The 50% margin also provides backup paths in the event of link failures. However, depending on the application and service type, QoE can degrade at applied loads less than 50%, providing network operators with a challenge of not knowing what there users are experiencing at any given point in time. Fortunately there are tools which can measure application specific user QoE. Using such tools enables network operators to improve network performance and enhance users QoE, thereby reducing economic impact.
Some of the tools I use in my day to day work enable me to quantify QoE and determine where in the supply chain impairments are occurring for the user. This has enabled me to not only identify performance impairments but more importantly, I’ve been able to understand where these impairments were and resolve the issues once the problem had been identified.
“If you cannot measure it, you cannot improve it.” Lord Kelvin