Providing islands of video expertise in a sea of IP
All broadcasters and network operators need the flawless delivery of high-quality content to TVs and other video devices. More and more operators are achieving this through using IP in the core network and are starting to deploy IP at acquisition and at the studio. This article looks at how the move to IP brings advantages to broadcasters through important new techniques such as intelligent switching, advanced forward error correction (FEC), remultiplexing, monitoring and video compression. These techniques combine to not only make video transport more efficient but also provide important advantages in video network operations and workflow.
Even though the bandwidth advantages of moving to IP are so persuasive, the changes in workflows enabled by IP will be even more important moving forward. Many solutions now use two-way IP communications for enhanced control and management. The combination of the move to digital TV and the two-way nature of IP networks increasingly make it possible to build intelligence into the video network. This is a trend that's only just starting. The first steps have been taken with systems such as seamless redundancy switching, which is already making life a lot easier for a number of major operators. This trend will only increase, with IP providing a raft of advantages, including:
- Increased operational and workflow efficiencies;
- Maximized network bandwidth utilization;
- Minimized management resource requirements;
- Ensuring QoS throughout the video transport chain; and
- Improved cost effectiveness through redundancy and automation.
The initial criticism of IP in broadcast was that it was designed for data traffic and not for applications demanding very high quality, such as live HD video. However, these questions have now been pretty much resolved in the contribution and distribution domains, through engineering IP networks to fulfill the QoS requirements for live video and by deploying technology such as advanced FEC and Virtual Private LAN Service/Multiprotocol Label Switching (VPLS/MPLS).
Looking at just one area, outside broadcast, IP connectivity in general has a much lower cost than previous satellite or video telco networks. IP also means that video can be distributed over generic Ethernet networks, rather than the whole distribution network being video- or broadcast-centric. This saves money and also makes it far easier to contribute content from wherever it is being generated, particularly for news footage or for sports and events at smaller venues that may not have dedicated links already installed.
The multidevice scenario
Today, broadcasters are faced with the challenge of delivering their content to a vast array of devices including large HD TV sets, SD TV sets, PCs, mobile phones and tablets. At the same time, whatever the viewing device, the customer expects a high-quality viewing experience. To meet the quality requirements demanded by consumers, the broadcaster has to deliver a product compatible with the highest video quality device, the large TV set in the living room. This means that the broadcaster has to design at least the first part of his content delivery network, the contribution network, to deliver high video quality. There is a clear trend toward broadcasters designing their contribution infrastructures to support 10-bit/1080p even though most of their current contribution is still SD or HD at 1080i or 720p. The transcoding to prepare the content for different viewing devices happens in the studio, allowing the quality and format to be tailored for the various device screens and resolutions and at the same time archiving the high-quality content for future repurposing. The use of IP networks makes it economically viable to keep the content at high quality for as long as possible throughout the video chain, which in turn makes transcoding more effective and efficient when handled centrally.
JPEG 2000 in IP video transport
The JPEG 2000 codec is particularly suited for IP video transport. JPEG 2000 is a wavelet-based compression technology that provides a number of benefits over DCT compression methods such as MPEG-2 and H.264. JPEG 2000 has many inherent strengths and benefits that make it a better choice for high-quality video contribution. When run at bit rates from 120Mb/s to 200Mb/s, JPEG 2000 will deliver video quality comparable to uncompressed HD video, which otherwise requires bandwidths of 1.5Gb/s.
One major difference compared with the DCT compression of the MPEG family of codecs is that the wavelet transform of JPEG 2000 is carried out over the complete picture in one transform, and not by blocks of pixels, avoiding the MPEG macroblocking effect, which can create impairment easily noticed by the consumer.
DCT is well-suited to distribution because it was devised as a means of compressing broadband video to a small bit stream that could fit in narrow broadcast or satellite transmission channels. The MPEG family of video compression schemes can employ other tools, in addition to DCT, such as motion vector coding of image differences and entropy coding. Entropy coding can be compared to zipping a file. The computing power required for DCT processes increases significantly in proportion to the size of the image.
As an intraframe encoding method, JPEG 2000 encodes each video frame independently. This approach offers many benefits to live video contribution applications. Unlike DCT, JPEG 2000 produces very low latency of less than 1.5 frames for encode or decode. Since each video frame is a key frame containing all of its own picture information, the transport stream can also be edited with frame accuracy. JPEG 2000 can allocate 10 bits or even 12 bits at 4:4:4 quality — compared with MPEG-4's 10-bit/4:2:2 quality — a level in line with the demands of video contribution.
JPEG 2000 offers two compression modes. The mathematically lossless mode uses reversible integer wavelet filtering to ensure that the compressed data has all the information of uncompressed SDI video and therefore provides video transport of equal quality, but with a typical bandwidth saving of at least 60 percent. The visually lossless mode employs floating point filtering and quantization techniques to provide greater compression with no perceived loss of video quality. Typically, visually lossless JPEG 2000 uses from 120Mb/s to 150Mb/s, while the backhaul of uncompressed SDI video needs a 1.5Gb/s pipe.
The advantages of JPEG 2000 in contribution include:
- Low latency;
- Picture-by-picture encoding;
- Fewer visual artefacts;
- Robustness in case of transmission errors;
- Sustains multiple code/decode steps; and
- Two high-quality options, through support for both mathematically lossless (equal to uncompressed video) and visually lossless encoding.
IP robustness: meeting operator needs
A fundamental requirement for IP contribution networks is to flawlessly handle any single-point failures, both in the network and in the video processing devices. IP networks are typically designed with automatic and fast rerouting in the case of a node failure. The rerouting delays are highly dependent on the network design, but can be as low as 50ms in a well-designed MPLS network.
There are several mechanisms that can be employed to increase robustness in the IP network. One mechanism is to use external redundancy technology, e.g. have video edge devices controlling external switches and thereby manage how content is sent through the network. Another mechanism is to have video edge devices more closely integrated with the IP network. A typical scenario sees cost-based routing mechanisms used to automatically select the best video signal where redundant video sources are available.
Of course, IP networks can still suffer from packet loss, so the video edge devices have to be able to cope with these losses and still deliver uninterrupted video signals. Advanced FEC mechanisms are now built into the video edge devices to ensure the reconstruction of any lost packets.
The market is moving toward building intelligence into the edge devices, thereby enhancing the robustness of the entire system. Cost-based routing mechanisms are already being integrated into video devices, but there is much more to come. Resource ReSerVation Protocol (RSVP) is a transport layer protocol designed to reserve resources across a network and is capable of improving both flexibility and robustness in contribution networks. Session Initiation Protocol (SIP) is widely used for controlling communication sessions such as voice and video calls over IP, and it is possible that this protocol will also find its way into professional video networks.
Networks can support automated redundancy switching with seamless switching between transport streams. Solutions are also available for SFN and DVB-T2 frame-aligned seamless switching for terrestrial networks. In addition, TS monitoring solutions are now available to handle multiple TS monitoring and error detection both over ASI and IP. These solutions deliver:
- Automatic input stream redundancy;
- Improved error resilience in IP networks, with advanced FEC; and
- Important QoS improvements for satellite network operators through automated traffic shaping designed to protect bit rate overflow in one service affecting other services in the multiplex.
Providing video expertise in a sea of IP
It is clear that content distribution networks need to cater for more than just broadcasting content to the TV. As operators need to reach multiple device types as well as merging broadcast services with broadband content distribution, the flexibility of IP contribution networks will become even more important, with intelligent edge devices becoming a necessity as complexity increases.
The video industry is migrating to architectures that can be described as islands of video expertise in a sea of IP. This revolution is now unstoppable. JPEG 2000 compression, redundancy mechanisms and FEC, TS monitoring, and intelligent switching all combine to make IP-based video-centric networks more flexible, powerful and easier to manage.
We are already seeing large-scale deployments of systems designed to exploit video-centric IP networks. They are expected to deliver the highest possible quality, meet the most stringent QoS requirements, enable innovative workflows and result in cost-effective business practices for broadcasters and video professionals.
As IP becomes pervasive throughout the professional video world, continued innovation in IP video transport technologies will accelerate adoption and enable vendors with the crossover skill set of IP and video transport to ride the waves of change.
Janne T. Morstøl is chief operating officer at T-VIPS.