This paper addresses common situations where more intelligent traffic management techniques based on specific traffic flow attributes can be used to optimize Quality of Experience (QoE). We present several user stories to illustrate these situations. We explain how effective traffic management requires in-network traffic classification/identification or
application-based self-policing of traffic based on subscriber profiles to optimize traffic flows. It is important to note that application-based self-policing techniques should not inherently be trusted without verification by the network that the application is actually enforcing the policy. By properly identifying traffic flows that can be more intelligently optimized, effective traffic management can be implemented while optimizing QoE. We also present and analyze several potential solutions and provide recommendations and analysis for their use.

Traditionally network operators optimized video using a network-based approach where all the identification and “optimization” was done in the network without application collaboration.  However, given the increase in the use of encryption for this type of traffic, this approach is increasingly ineffective. A more collaborative approach as highlighted in this paper is needed whereby the network works with applications allowing the application itself to perform the optimization.

The next step is to further develop and deploy the identified solutions or new industry solutions to better meet the needs of users, network providers, and application developers.