Why is quality of service now so important for network operators?
In a market in which growth rates are declining, customer retention has never been so important. As network operators continuously deploy new services such as LTE and Voice over IP (VoIP), it becomes increasingly more important to provide a highQuality of Service (QoS) to ensure customers get the same level of satisfaction they have become accustomed to.
However as new services are introduced, they bring increased complexity (IP and CS) and huge amounts of data to the network operator. Voice over IP for example has stringent performance requirements such as high data throughput capacity, low latency and loss. Thus, service providers need service quality analysis that can proactively and continuously monitor voice over IP calls in multiple areas (eNodeB, EPC, policy, charging, IMS) and on top of that, detect and mitigate service quality degradation of VoIP traffic across the entire network.
Why probes?
Probes are one of the most effective ways to gain insights of the end-to-end performance that is perceived by the end user. They provide real-time and historical end-to-end call tracing, proactive alarming and KPI’s that constantly monitor the network status and Quality of Experience (QoE).
Active and passive probe measurements are the two fundamental approaches that can enrich the end-to-end (from the user terminal to the core network) analysis and increase performance. Probes form an important part of the ever-increasing complexity of modern telecom networks, as they provide granular data that allows service operators to determine the service quality at a per-service (QoS) and per-user granularity (QoE), across multiple transport technologies.
Active probes vs. passive probes?
Active probes emulate VoIP traffic through the network and measure the service quality of the network, providing an end-to-end measurement mechanism of the network, which can be used to detect service degradations. However, active probing alone cannot help pin-point the root cause of failure or congestion. It can detect an excessive delay through the network, but it cannot identify the cause of the failure.
Passive probes provide a different perspective of the network, as they are installed on links within the network, and they ‘snoop’ on all the traffic that flows through the link being monitored. They filter individual VoIP calls and compute the service quality received by each one of them. In this way, they segment the network for sources of failures or congestion. However, they do not give an end-to-end perspective of the network performance. They can compute the performance of the network only between the points of installation of the probes.
So, active and passive probes, when used in combination give service providers the capability of effectively monitoring their network for VoIP performance.
Combining probe data with network and customer data for a true end-to-end view of the network
For a complete view, these probe measurements should be correlated with multiple node-to-node performance data and customer data to reach the end-to-end quality metrics goal.
The Zen software solution from SysMech can proactively manage the network by constantly monitoring end-to-end KPIs and KQIs, created from different perspectives in the network, providing an integral view from customer, network, service or terminal perspective, immediately highlighting any degrading trends (e.g. dropped-call ratio and set-up times).
Network parameters (delay, loss and jitter) that can be used to compute the call quality using a voice quality metric can also be added to identify network impairments that cause service quality degradation, and operators can then take corrective measures in real time so that the impact on the degradation perceived by end users is minimal.
In addition to assuring a high level of customer satisfaction, a proactive approach can yield savings by minimizing the number of trouble-tickets and hence optimising resource allocation. Zen’s pre-defined reports can also visualize service quality, facilitating root-cause analysis across all network layers.