SIP: Add focus to your bird’s eye view

Favicon_Qr_Qrystal_86

2021-02-21

SIP: Add focus to your bird’s eye view

Sometimes there is just too much data and focusing on the really important stuff becomes harder and harder. What helps is aggregation. It improves clarity. And decision making.
Voipfuture’s Qrystal aggregates detailed measurement data to any dimension of interest. What you get is high-level yet accurate information. On regions, branch offices, interconnection partners, VIPs, routes and destinations.
You simply get things done more efficiently. Means: reduced stress and improved results. More info? Happy to help.

Stop looking for SIP faults. Start finding them

Favicon_Qr_Qrystal_86

2021-02-21

SIP: Add focus to your bird’s eye view

The Voipfuture way of doing things is: Go deep. Recognize failure patterns. And finally, automate everything through appropriate indicators. This immediately makes issues visible. On call, trunk or cell level.

It empowers the operational team to achieve efficiency. So no time is wasted on manual, routine and repetitive tasks. Because both really is important – signaling quality and the team’s time too.

Critical Loss Densitiy

Favicon_ai2__86

Intelligent Indicators

2021-01-14

The worst loss isn’t burst loss

Everybody knows that bursts of lost packets negatively impact speech quality of VoIP calls. The gaps in the packet stream directly translate to gaps in the audio. But there is an even worse form of packet loss.

Imagine a series of frequent but individual packet losses, e.g. packet lost, packet received, packet lost, packet received, etc. We call this pattern Critical Loss Density (CLD).

It’s clear that such a loss pattern is bad for speech quality, because loss concealment mechanisms – for example, replaying the last received packet – cannot mask a multitude of gaps stretching over time. This loss pattern is often caused by low level issues with network links such as old or broken fibers.

Qrystal is the only passive monitoring solution on the market that can detect CLD and identify the root cause.

 

Happy to help.

Carsten

Beware of all call drops

Favicon_v_voipfuture

2020-11-26

Beware of all call drops

As a user you know for sure when your call drops. Yet, as a mobile service provider you probably don’t. The reason is that conventional call drop detection relies on data provided by individual network elements, e.g. in the RAN. This view often does not match the user experience.

We at Voipfuture care about user experience and thus implemented call drop detection based on the analysis of user behavior. When your 4G call drops, you will typically try to re-establish the call. Qrystal takes these redial calls and inspects their RFC 3326 SIP reason headers. Machine-learning is then used to determine the headers that indicate call drops.

The result: Qrystal’s dropped call ratio (DCR) is in line with values determined by drive tests. Even better: the data is available 24/7, for your entire network.

Bearer Loss Events

Favicon_Qr_Qrystal_86

2020-11-11

Bearer Loss Events

How to efficiently deal with dropped calls

For customers, there is little that is more dissatisfied than call drops. But the equation is simple: more mobility (i.e. more handovers) and more traffic load lead to more dropped calls. Extremely helpful in understanding them are SIP reason headers (RFC 3326). They indicate why a SIP request was issued. Specifically, when a call is dropped for technical reasons, the final BYE message includes a reason header specifying the actual drop cause. Qrystal’s statistical analysis of these reason headers is thus key to identifying and solving dropped calls in VoLTE services. A recent customer case solved using Qrystal impacted millions of customers – and demonstrated that the network is not kidding when it reports “Dedicated Bearer Lost”.

SIA – more than a million calls per week

Favicon_ai2__86

2020-07-08

SIA – more than a million calls per week

Hi all,

 

In my last post about the SIA (SDP in ACK) I said that the problem doesn’t come up that often. The question remained open: What does “not that often” mean?

The screenshot shows two things. Firstly, the problem is minor in percentage terms. On the other hand: In this case, more than a million calls per week are affected.

If the problem is regionally concentrated, the operations team definitely should know about it.

 

Take care,

Carsten

SIA Indicator

Favicon_ai2__86

Intelligent indicators

2020-06-23

Cut! The SIA indicator, media interworking and call clipping

You surely have experienced this: You call a friend, ringing stops and then there is silence. You say “Hello…?” and immediately your friend responds with “Yes, I’m here, I already said hello!”.

You missed that first “Hello” from your friend which caused an awkward moment.

Why’s that? When you initiate a call, the endpoints will use to the SDP protocol to negotiate which codec to use. Sometimes this negotiation is still ongoing, when the call gets connected. In effect, the endpoints do not know which codec to use & to expect and the first second(s) of speech are clipped.

 

Voipfuture Qrystal has a special indicator to automatically detect such clipped calls. It’s called “SDP in ACK” or SIA for short. There are call routing scenarios, e.g. involving back-to-back SIP user agents, where the late end of codec negotiation cannot be avoided, and where it’s technically perfectly legal.

 

However, to the users the clipping effect is a nuisance. You want to know how often it happens and want to eliminate it wherever possible. Qrystal’s SIA indicator is here to help.

Take care, Carsten

Just in case you want to know more:

carsten.niepmann@voipfuture.com

How to improve VoIP Monitoring with Media KPIs

Favicon_v_voipfuture

How to improve VoIP Monitoring with Media KPIs

LIVE WEBINAR
Tuesday, June 30, 2020 – 2:00pm ET | 11:00am PT

Quality of service is high on the agenda of VoIP service providers. Many KPIs for measuring the signaling performance exist, but with voice you also have to look for the in-call user experience. Typically, we measure packet loss, jitter and the ominous “MOS” for each call.

Unfortunately, this data will very often not match the experience of actual users.

This webinar discusses how a new generation of Media KPIs can improve VoIP monitoring. These KPIs are already being used today by tier-1 communication service providers around the world to monitor the media plane quality.

What Attendees will learn:

  • Why MOS per call is misleading: The need for new KPIs to measure the in-call user experience
  • How RTP timeslicing provides essential insights
  • Big data: aggregation and slicing of user experience information
  • Examples of VoIP & VoLTE service quality reports using timeslice KPIs

Who should attend:

  • Telecom Engineers
  • VoIP service/network operations & optimization
  • Decision makers with technical background assessing state-of-the-art tools

#telecom #voip #telecomengineer #servicequality #transportquality #kpi

The seminar is already over - watch video

Objective:

Engagement

  • Visit TMC registration page – click on link
  • Create follower – click follower button

 

New Webinar: How to improve VoIP Monitoring with Media KPIs
Tuesday, June 30, 2020 – 2:00pm ET | 11:00am PT

 

Quality of service is high on the agenda of VoIP service providers. Many KPIs for measuring the signaling performance exist, but with voice you also have to look for the in-call user experience. Typically, we measure packet loss, jitter and the ominous “MOS” for each call.

Unfortunately, this data will very often not match the experience of actual users.

 

This webinar discusses how a new generation of Media KPIs can improve VoIP monitoring. These KPIs are already being used today by tier-1 communication service providers around the world to monitor the media plane quality.

 

What Attendees will learn:

  • Why MOS per call is misleading: The need for new KPIs to measure the in-call user experience
  • How RTP timeslicing provides essential insights
  • Big data: aggregation and slicing of user experience information
  • Examples of VoIP & VoLTE service quality reports using timeslice KPIs

 

Who should attend:

  • Telecom Engineers
  • VoIP service/network operations & optimization
  • Decision makers with technical background assessing state-of-the-art tools

 

#Telecommunications #VoIP #Quality #NANO #Network #Operations

Some more on the Network Overload Indicator

Favicon_ai2__86

Some more on the Network Overload Indicator

Another short post about the Network Overload Indicator. Because the question arose how it looks exactly in Voipfuture’s Qrystal, here is a short screenshot for those who are not familiar with it.

Of course, this is just a short extract from our pretty long list of indicators. But I think the principle becomes clear.

All automatic indicators tell you:

  • What is the problem
  • And how big is it

You can see how much traffic there is in a selected period. The individual indicators show which impairments there are. In this case it is the network overload indicator and the related jitter.

It is shown as absolute values, as ratio of minutes and you can select which indicators to plot in a chart. And as always with Qrystal, there is a click-through to get to the bottom of it.

If you want to know more or a short life demo – a short email is enough.

Stay sane,

Carsten

How to detect Network Overload

Favicon_ai2__86

How to detect Network Overload

Recently we spoke about increasing voice traffic due to Corona and mentioned the network overload indicator. This has led to inquiries, which we are happy to answer this time.

It is important to measure jitter of voice traffic – excessive jitter cannot be compensated for by the receiver and causes gaps in the audio. But what is the reason for a specific jitter event and how do you prevent jitter from happening?

One common source of jitter in the network is buffering. Overloaded network elements like routers and SBCs may be too busy to forward packets in time and start queueing up packets. This creates a very unique pattern in the packet delays of affected RTP streams.

The pattern is automatically detected by Qrystal and flagged by the Network Overload indicator.

The recent increase in call volume due to the Corona crisis underlines the importance of this indicator. Most subscribers won’t notice yet, but the rising traffic is pushing some networks to their limit and eventually the user experience will suffer. Qrystal’s Network Overload indicator will give you the extra time needed to identify bottlenecks and increase network capacity.

More? Just drop me an email.

Stay sane, Carsten

carsten.niepmann@voipfuture.com

Data protection
We, VOIPFUTURE GmbH (Registered business address: Germany), would like to process personal information with external services. This is not necessary for the use of the website, but allows us to interact even more closely with them. If desired, please make a choice:
Data protection
We, VOIPFUTURE GmbH (Registered business address: Germany), would like to process personal information with external services. This is not necessary for the use of the website, but allows us to interact even more closely with them. If desired, please make a choice: