As we migrate to IP, broadcasters are thinking about how they will interface their SDI, MADI, and AES systems together. Many see the benefit of IP and one day all devices will be IP-enabled. But until that time arrives, we need to understand how to interface new and old together. In this article we investigate real applications on how to connect existing and new technologies.
Few broadcasters have the benefit of implementing IP in a greenfield site and will instead be slowly and cautiously migrating IP into their existing infrastructures to meet their business demands. Upgrading from analogue to SDI had its challenges but could be quickly and easily achieved. And moving from SD to HD SDI was relatively straight forward.
Coaxial cable for both migrations differed in their characteristics but they required the same broadcast engineering skillset to make it work. Although processing equipment, such as distribution amplifiers and vision switchers, operated completely differently between analogue, SD and HD, the interfaces and underlying cabling structures were very similar.
Natural Analogue to Digital Migration
The basic video-audio coax and twisted pair cabling infrastructures have stood the test of time and serviced the broadcast industry well since the 1930’s. As each video technology emerged, the quality and characteristics of the cable had to be improved to address the higher frequencies needed. But essentially, the cable looked the same and could easily be installed into existing infrastructures without too much inconvenience.
Although Ethernet networks have their history in coaxial cabling, twisted pair infrastructures such as CAT-5 have dominated the landscape since the 1990’s. Simple to use and easy to install, CAT-5 could achieve data rates of 100Mbit/sec.
CAT-5 Provides IT Connectivity
Various improvements to CAT-5 occurred over the years culminating with CAT-8 to achieve 25Gbits/sec and 40Gbits/sec. But as fiber costs came down and even higher data rates were needed, fiber infrastructures started to dominate IT infrastructures.
From a cabling perspective, any type of signal can be distributed over fiber or CAT-5. This is different from an Ethernet network operating over fiber or CAT-5 as the distributed signal must comply with the specification of the underlying Ethernet protocol.
We can easily install singlemode fiber backbones with their virtually unlimited bandwidth capabilities and finally future-proof our infrastructures since fiber is agnostic when it comes to the signals it can carry.
Use IT Infrastructure
During IP migration, broadcasters can take advantage of fiber infrastructures and connect high speed devices together using proprietary protocols. Broadcasters still benefit from the COTS infrastructures since the technicians installing the cabling for IT can easily install the same cabling for broadcasters.
Image 1 – Riedel’s MediorNet real-time network uses dynamic bandwidth allocation to maximize fiber usage in their proprietary networks. MicroN can provide a gateway to IP networks to deliver the best of both worlds
When IP connectivity becomes available for such equipment, patching the fiber into an Ethernet switch is an easy process thus making the broadcast equipment ready for connectivity to the Ethernet network when the vendors make the interfaces available.
Two dominant media IP standards have emerged; ST2022-6 and ST2110. Both distribute over IP and can deliver lightweight compressed and uncompressed media.
ST2110 is more efficient than ST2022 as it removes the unnecessary overhead of line, field, and framing information, and has delivered the most significant change to broadcasting since the 1930’s as it abstracts away the media essence from the underlying transport stream. And for the first time in the history of television, we can process video, audio, and metadata all independently of each other while still maintaining frame-accurate synchronization.
The key to making ST2110 work over an IP network is providing an accurate timing reference. PTP (IEEE1588:2004), usually locked to an external source such as GPS, provides the sub-microsecond accuracy needed for reliable distribution of media across an IP network.
Timing Timing Timing
During migration, a hybrid network of IP-Ethernet and SDI-Video-Audio will soon develop and maintaining accurate timing between the two is critical.
SDI, AES, and MADI networks also require a single source of timing reference. Sync Pulse Generators (SPG) have been the traditional origin of black-and-burst for video and word-sync for digital audio. As well as ensuring that adequate bandwidths are available on the IP network links, a timing relationship between the PTP grand master and SPG must be established.
If we consider a studio that has SDI cameras and IP video playout servers, both must be frame synchronized to achieve seamless switching between the camera and playout server video output.
Diagram 1 – When integrating ST2110 and SDI, the PTP Grand Master and SPG should be locked to guarantee frame accurate switching. In this case, the SPG field counter is running slightly faster than the video servers field counter and the video fields do not align. This will cause video disturbance when switching between the two.
When a PTP slave, such as a video playout server, synchronizes to a PTP grand master, the value of the counters representing the number of nanosecond tic-events that have occurred since the epoch are adjusted to make them the same as the grandmaster PTP clock.
The video servers internal media clock is running at 90KHz for video processing and 48kHz for audio processing and this is used to gap each IP packet to maintain compatibility with ST2110-10 and keep the 40 millisecond frame rate. The 90KHz clock is derived from PTP by counting the number of nanoseconds.
Frame Lock Video
The PLL oscillator in the SDI camera locks to the clock frequency of the SDI signal from the SPG. For SD this is 270MHz and for HD progressive is 2.97GHz. The camera will then derive its frame, line, and pixel signals from this input. To achieve a 25Hz frame rate in SD, the camera’s digital electronics will use counters to just divide 270,000,000 by 10,800,000 to get 25.
In the case of SDI, all equipment connected to the SPG will be phase and frequency locked, resulting in an exact number of frames per second.
However, if the video servers slave nanosecond counter and the camera’s local PLL are not exactly time synchronous, and the playout server is running fast relative to the camera, then there will be more video frames from the playout server than the camera. When switching between them on a vision switcher, frames will be dropped or duplicated resulting in a significant problems.
Buffers Add Latency
To overcome this, either buffers and frame synchronizers must be used, thus adding to latency, complexity and cost, or more efficient solutions can be used. Some PTP Grandmasters have black and burst SPG’s built into them and products such as Riedel’s Mediornet have the ability to synchronize black-and-burst to PTP. This is a true hybrid solution that covers the transition from SDI to IP.
A PTP Grandmaster is used to synchronize the IP-compatible equipment. Mediornet synchronizes to the PTP Grandmaster through the MicroN interface and all connected Mediornet frames through the fiber infrastructure to provide a timing frame for SDI equipment.
This provides the best of all both worlds as the IP and SDI equipment are frame synchronous.
We are just at the infancy of migrating to IP and to truly take advantage of the benefits of COTS, broadcasters should consider moving to ST2110. Although we can use existing fiber cabling infrastructures to migrate cautiously, to be ready to move, broadcasters should consider their timing designs before any equipment is procured.
Timing has always been at the core of any design. It is the first consideration as each frame of video must arrive synchronous to the vision switcher, and each sample of audio must arrive synchronous to the sound console. PTP and IP is no different, especially as we migrate from SDI, AES, and MADI, and timing must be considered and understood at the outset.
You might also like...
Today’s broadcast engineers face a unique challenge, one that is likely unfamiliar to these professionals. The challenge is to design, build and operate IP-centric solutions for video and audio content.
Broadcasting used to be simple. It required one TV station sending one signal to multiple viewers. Everyone received the same imagery at the same time. That was easy.
Are you an IT engineer having trouble figuring out why the phones, computers and printer systems work but the networked video doesn’t? Or maybe you have 10-15 years of experience with video production equipment but really don’t understand why…
As broadcasters migrate to IP, the spotlight is focusing more and more on IT infrastructure. Quietly in the background, IT has been making unprecedented progress in infrastructure design to deliver low latency high-speed networks, and new highly adaptable business models,…
In principle, IP systems for broadcasting should not differ from those for IT. However, as we have seen in the previous nineteen articles in this series, reliably distributing video and audio is highly reliant on accurate timing. In this article,…