Every public or private network operator needs to understand how its network is performing for its customers. Although operators configure QoS ( Quality of Service) parameters throughout their network infrastructures, the EXPERIENCE of the network is ultimately the customer’s and is referred to as Quality of Experience (QoE).
QoS: A set of parameters and settings configured in devices at the various network functions.
Network equipment offers configurable QoS parameters to enforce particular performance QoS, such as: traffic throughput, transit delays (latency), jitter, and other performance shaping parameters. For example, a network operator may put in place QoS to shape virtual Ethernet circuit service offerings to businesses. Today’s operator networks offer various service levels of QoS to satisfy particular customer requirements.
QoE: The experience delivered to a network-connected end-user, device, or application.
QoE measures how the service is experienced at the edge of the network, where the customer connects. Although some legacy applications such as batch data transfer may not be as sensitive to QoE, others, such as real–time remote surgical procedures, are. For any IoT/IIoT, mission–critical and time–sensitive applications, to score well on QoE, the complete service, including connectivity from end to end, must be considered regardless of the QoS applied by the network operator(s).
In short, QoS refers to the network parameter settings configured by service providers to deliver various service level offerings to their customers. QoE is a measurement of the network service delivery as experienced by the customer.
QoS and QoE Together
The network operator‘s provisioning function defines and sets QoS parameters for various network functions, such as the RAN (radio access network), the backhaul, the core network, other network functions, and the cloud(s). A QoE tool operates on top of the QoS-enabled infrastructure to report what is happening throughout the network, from the edge to the cloud.
Take a video call as a simple consumer example: the network operator’s QoS settings will help dictate how the service offering is shaped within the service provider’s network and sold to the customer. The QoE tool will evaluate the network quality assurance and genuine network performance delivered and experienced by the customer’s people, devices, and applications.
In this way, QoE provides a broader, higher-level view of what is going on in the network from the end-user perspective, incomparison to the more specific view of individual network parameters that QoS provides. Public and private network operators need both types of tool today.
Actionable, Real-Time IoT Network Analytics: QoE for Mission-Critical Applications
Industry, government, and business increasingly leverage and rely on new and more sophisticated systems and applications to automate critical operational efficiencies, such as handling hazardous processes or material and safety systems. The very foundation of these systems depends on heterogeneous networks provided by both public and private network operators.
The ability to visualize, understand, and maintain safe and proper network QoE is mandatory to any network operator that serves mission-critical machines and people.
For these customers, the QoE score is the only important metric. For all parties, understanding the customer’s experience of the network is key to isolating issues, optimizing all parts of a network, and predicting and pre-empting application or process failures due to inadequate network QoE, even if only for a moment in time.
ML and AI-Driven QoE
A QoE solution driven by ML (machine learning) and AI (artificial intelligence) can help identify where in the network data path issues occur. It can also suggest how to resolve any issues, regardless of the network supplier or equipment vendors, and without the collection of network probes that network operators have traditionally deployed. An ML- and AI-driven QoE tool provides enhanced visualization of network health. With a proper QoE tool, it becomes possible to pinpoint performance issues anywhere they occur and to predict and pre-empt network issues before these affect the end-user.
Successful public and private network operators will invest in ML and AI-driven QoE tools to ensure their networks deliver as needed. These demands come in part from IoT and IIoT network customers seeking to leverage new, real-time applications that make use of remotely operated, semi-autonomous, and autonomous equipment and functions.
At Cheetah Networks, we empower public and private network operators to observe, experience, analyze, and act on—in real-time—the QoE delivered by their networks. The ML and AI-driven PulseView™ Solution correlates inputs from the network edge to the cloud, then analyses, correlates, and presents on a single pane of glass actionable, real-time network analytics.