Technology

Bad user experience can take a number of different forms. A VoIP call can suffer from cut words at the beginning, choppy voice, single-sided voice, “cross-talk” and unexpected termination just to name a few impairment effects. One of the biggest challenges when operating a VoIP service is to identify such effects automatically before customer care is hit by complaints. Closely related to this is the challenge of identifying the root cause of quality impairments. This is a particularly difficult task, if operators cannot reproduce quality degradations, because it is temporary or depending on unknown factors.

Monitoring Approach

VOIPFUTURE’s passive RTP Monitoring solution provides the information required to evaluate the VoIP service quality and to identify the root cause of problems. To this end, it makes use of its own measurements and own patented algorithms instead of relying on RTCP (or other end-to-end approaches). The benefit of this is greater accuracy and detail, the ability to locate problems and the ability to measure voice traffic from arbitrary sources, which do not make use of RTCP or proprietary reporting protocols.

VOIPFUTURE’s elementary data structure is the quality data record (QDR), containing highly condensed statistical information for every 5 s segment of an RTP stream. The QDR metric is generated by the probes and consists of values, ratios, KPIs and indicators. Indicators enable effective root cause analysis, as they automatically identify QDRs that suffered from specific impairment types, such as network overload conditions. The key to this is VOIPFUTURE’s patented pattern matching technology which analyses RTP streams in real-time.

Quality Data Records

The QDR information is subdivided into the following groups:

• Network transport performance in terms of jitter, packet loss, burst loss, loss density, etc.
• Conformance of RTP senders to standard behavior and best practices
• Conformance of RTP senders to network policies, e.g. correct DSCP settings
• User experience in terms of MOS & R-Factor values according to ITU-T G.107 (E-Model) and P.564

The unique 5 s detail of VOIPFUTURE’s QDRs overcomes the limitations and inaccuracies of conventional metrics using average values per call. Consider a perfect quality 5 minute call of which the last 10 seconds were so impaired, that the users hang up. The average MOS value for such a call would be nearly perfect, yet the users would judge the call quality by the impaired last seconds. The VOIPFUTURE metric allows to quickly identify the exact time and location of an impairment, making troubleshooting and service quality management highly effective.

MOS Minutes The detailed QDR information is the foundation for the reports, alarms and statistics generated by the VOIPFUTURE system. For example, the MOS value of each QDR can be aggregated to MOS Minutes, i.e. a quality rating for each call minute. MOS Minutes allow for a very concise presentation of the overall voice service quality. The MOS Minute Histogram is an exclusive VOIPFUTURE feature and shows the quality distribution of stream minutes transported through a network. MOS Minutes are also used for voice quality SLAs, where they allow to tie the service quality to the billing unit. MOS Minutes allow for simple agreements, stating for example that only call minutes above a certain MOS value will be charged for. In offering such detailed information, the VOIPFUTURE technology acts as an enabler for proactive network performance and customer experience management in VoIP networks.

back to top