Measuring IPTV QOS (Quality of Service)
March 1st, 2010IPTV QOS is a topic that has been a problem of confusion for many companies, let me clarify.
Quality of service, which is something new 's environment makes people automatically think of using techniques of pre-measurement. This basic approach for measuring QOS is where most of the confusion generated.
Similarly, when companies began switching from analog to digital signals, the natural tendency of existing engineers who want to meet newdigital signal to be converted back to analog and then use their existing equipment. IPTV QOS has caused much the same method, where engineers with a background network want to measure network statistics, and engineers with a background video for the metrics for measuring video. The former (network engineers) is can fortunately their measurements of existing network infrastructure, but no feeling for all packets on the network refers to any videosignals. People want to convert the video signal to restore the IPTV digital video format (conversion from IP-to-video) that really misses the point, what you really know is how well does the conversion device (a piece of test equipment is not will be compared with the way in which an STB (STB) to decode the signal. So you have two different approaches to the same problem – not one of them is really ideal.
Now there is room for an existing test(test equipment network equipment is great for data traffic, as has always been, and Transport Stream (Digital Video) analyzers are great in your head-end (where the video content originates) to confirm the video across the Web IP which was good) so it is not the time to throw it away, not just the right tool for IPTV QoS.
With the observations of the ways in which we can go forward (it is difficult to move when you still have one foot in the oldthought).
Depending on who you are, you may well use only a fraction of an IPTV system, or involve the entire system, so we will break the heart of the problem and what it means in any position of the network (assign a network of 4 key points: 1) The main purpose 2) Core Network 3) Network Edge 4) Customer Home).
1) Head End
You can do alone are responsible for creating, delivering, or receiving video from a head-end.
A Fine Head may consisteverything from professional video encoder for servers VOD (Video On Demand), and one of the many video formats, the types of printing, bit rate, etc. They can be Unicast and Multicast, UDP, RTP, or a private facility (as in If MSTV).
Whatever the situation, it is a good idea to take measures to ensure that the head end is solid and that the video-coding device is reliable. A problem with the head end will affect all in a row, from right tocustomer. (assuming that the various "redundant" systems are in place to prevent this type of problem, as far as possible)
He built the system so the head with a solid architecture, the last thing (and the most important for us), the Head End IP-screen video output stream to ensure that the first point where the video is embedded IP is probably done and that the rest of the IPTV infrastructure may rely on this input.
Note: A common mistake in thispoints (and elsewhere) is a kind of round-robin system in place where not all the video streams are measured at the same time – must be carried out only if absolutely necessary as one of the 'problems' with the nature of IP delivery over a network is the damage caused signal in the IP field is a non-deterministic influence on the pattern yet. This means that while you look at 5 of 100 rivers, one can see that there are problems at some random digitsother flows, can not see – unless you check all streams simultaneously.
2) Core Network.
Hopefully, the above actions will be done, so if you are interested in the network core, your main task is to make your own statement that the flow into the network is ok (you can count on the provider End head for not doing it for you, and it is much easier to get out of focus when a problem occurs when you can easily confirm your input), and to ensurethat the transition of the network is not an injury or excessive movement due (only 2 components that can not stop the video network, we reach the end remained.
Now that we are now in the IP domain, the problem of packet loss is ultimately the number 1 thing to watch (all IP packets that are lost will result in the loss of video content, because all the mechanisms for video packets into IP packets for delivery to add some even up to 7 video packets in an IP packet).But with that said, each network device (and ultimately, STB) has buffer, which means that the movement may cause excessive packet loss. Because we really do not want the packet loss, this means that the movement is important for us to follow our system.
The real kicker is that if you're old school IP surveillance, will be completely satisfied with what I have said so far – but there's one thing that does something a bit 'more "interesting"' s. And 'quite possible to lose the "average"packages, but not IP packets. When an infrastructure that includes items such as MPEG multiplexer MUX and a couple Brooks merged into one unless you are a kind of deep packet inspection "(see the titles of the media to ensure the continuity counters are correct) does not could be the IP packet loss, but still problems yet. This means that the solution may come from a strategy or another, but not on the monitoring of IP –domain, while confirming that the packages on average intact.
This complication is one of the things that many test equipment manufacturers have not taken into account, especially because of the fact that it is still a relatively new area, and many equipment suppliers have focused on creating functions, rather than addressing the problems of customers to provide the services that are truly robust solution is needed.
3) EDGE network.
As before, our first step is to confirmour entrance is closely monitoring all streams simultaneously for jitter and packet loss and then make sure that the "last mile" mechanism for the customer's home is only as strong as possible.
How this action could easily lead to the conversion of IP to RF (cable companies, our RF (Radio Frequency) signals rather than copper or fiber that the most widely used network equipment must not have any test equipment, the appropriate interface for this (the most common interface here is a QAM (Quadrature AmplitudeModulation), of which there are 3 main types (it is actually called "Annex's') Annex A, B and C for U.S., Europe and Asia.
So depending on the facilities here, you may or may not have an IP network to the customer at home.
4) Client Home.
The final, and some might say the most important part of the system.
As before, we see our entry (IP video streams to go to the STB client). Since we are an over-IP, when it comesrelocation and packet loss that occurred after these video streams on their way to this house. Since the video quality as encoded in the end the head is turned on, we know that as long as the movement is not too much to cope with STB, and no packet loss – the video will be exactly as it was when was encoded.
If you're wondering how to get around this – there are suppliers of equipment with devices that go in the house of the customer, and the workload adstractby the STB, and even a couple of customers that allow the press of a button to signal when they saw a problem (no matter what your test equipment may or may not have explained – what is needed to simulate a customer experience)
Ci – Pretty simple, really.
True, but in real life, most companies are not in possession of, control or even access to the entire system. This makes the deployment of an IPTV deployment a bit 'of a nightmare, if you understand the issues and are the proper testequipment (remember that some people still have a foot in the network or video of the old world).
When companies do not have access to much of the system or work with a welcoming society, who do what they can get headaches a lot easier if the equipment used, can obtain information on a central video monitoring. This way, the 2 common problems 1) How is the problem 2) It 'a matter of IP is visible at a glance, and much waste of time and effort onlyreach the point where even know if the problem can be prevented.
When it comes to wanting to quantify the quality of the system, there are different standards for the assessment of IPTV, the most common are: 1) V-Factor 2) MOS 3) MDI
1) V-Factor
V-model is a system that uses Moving Picture Quality Metrics (MPQM) to do research to try to simulate what a man would decide the video quality was right.
This is an interesting approach, and is a way of approachingproblem, but requires a lot of therapy may not be realistic for most of the network is done (because the process of working hard, is not suitable for a nuclear control or head-end), so it can serve as a useful measure for to integrate into STBs.
When we look at a holistic approach IPTV QOS, only a monitoring solution that gives us the framework and the methods it will.
2) MOS (Mean Opinion Score)
Again, these statistics are designed to try to makeapproximation of what a human would see.
As for the V-factor, an idea is fresh and technically excellent, but we can not know what is wrong with the system (it's nice to have a quality "score", but in reality is what we need to know what to dealing with a "poor" rating).
3) MDI (Media Delivery Index)
As the name suggests, you get a statistic that tells us something about the delivery. (XX / YY where XX is equal to the total and jitter JJ is equivalent to packet loss) This time, instead of tryinganalysis of the video and "score" no, we get information on jitter and packet loss on the point measured. Although it is not possible to study the video signal is decoded, it means that we tell you how good the video is delivered – if you can remember what is really the key, when the correct code.
MDI is a apporiate metric anywhere in the system and tell us immediately if there were supply problems. MDI because values are based onTransmission rate of video streams, which also gives us valuable information about how the different schools will be affected by our network (for example, if we're already running 50 SD (standard definition), streams, and we want to replace with HD (High Definition) flows, a V-model of the MOS score of a particular point in our network, we will not tell you what to expect, while the MDI measurements, the network will show us how much difference can be expected to know. Movement on SD Web affectedand HD streams in a different way (in fact, all streams with different bit rates have been otherwise affected by the transition – which causes many problems), so that information about how the movement influenced over IP is really useful information, just don 't get with other measurement systems.
I hope you find this article useful and take the necessary steps to create a reliable system to ensure, before launching a headache. "Another article will follow a brief description of how to build aIPTV network robust.
