Fortunately, dealing with the Telcos became less of a burden because they often could be avoided with new technology. The first adopted solution was line-of-sight microwave systems. That technology subsequently gave way to satellites, but each called for carefully positioned (and expensive) additional vehicles on site--along with a trained operator.

Permanent video and audio circuits were deployed for regularly used venues, but even these proved limiting as the nature of broadcasting changed. Now production companies are expected to deliver multiple feeds back to base, with different feeds for different users.

Telco’s ramp up IP

While broadcasters were agonizing over this problem, Telco’s were busy installing huge amounts of data capacity in the form of IP networks. The latter seemed to be an obvious solution to the former, but broadcast engineers generally remained skeptical of using IP for mission-critical contribution.

Security and ease of access were considerations, but the fundamental issue was the difference in philosophy between broadcast and IP. As a technology, IP was designed around the best-effort principle, i.e. that the content will likely get there as soon as possible. In contrast, broadcast television expects a new picture every 40 milliseconds, without fail. If it doesn’t get one, everybody notices – including the viewers who chose that channel and the advertisers who are paying for it.

IP contribution was not a viable consideration until large amounts of deterministic bandwidth was available. Today we have that availability. More importantly, we have it at a price that is very attractive to broadcasters. Furthermore, with companies like Nevion providing technology, including advanced monitoring and protection mechanisms, to bring a broadcast-level of reliability, and an application layer to manage the network, it’s now simple to set up and manage ad hoc circuits as well as defined networks over IP.

IP for broadcast - the critical management layer

IP-based transport’s inherent flexibility is also its primary challenge. The process can be complex and may require a knowledge base to properly build and operate an IP contribution broadcast network. Fortunately a technology overlay can overcome the difficulties posed by IP-based media transport and bring simplicity to something that is inherently complex.

We’re in the midst of a shift, a convergence between broadcast and the IT industry for real-time broadcast applications. We have the tools to maximize the efficiencies of IP and the availability of “commercially off-the-shelf” components to build highly efficient broadcast contribution networks based on IP. Ultimately, the single most important element that ties all the moving parts together is an overarching media management layer.

Software Defined Networking’s (SDN) philosophy of centralized control of all parts of a network is the key ingredient that anchors any IP-based media network. With centralized media management, all systems and services are constantly monitored and actively controlled to ensure end-to-end managed services. This allows users to identify all services running though the network and to pre-empt any potential faults that may occur or to quickly resolve faults when they do occur.

Protecting Media Over IP

The technology exists to protect streams carried over IP networks, ensuring that the content is delivered in a timely fashion and without disruption.

Forward Error Correction (FEC), defined by SMPTE as ST2022-1 for transport streams and JPEG 2000 streams and ST2022-5 for uncompressed SDI, provides basic protection against random packet losses and short burst packet losses. However, for more severe faults, such as link or node failures in the network, other technologies need to be deployed instead of, or in addition to, FEC to avoid outages that would impact the streams and potentially harm revenues.

Protection mechanisms, such as RTP seamless switching or Streaming Intelligent Packet Switching (SIPS), leverage network redundancy and can increase the performance of media networks and greatly increase resiliency.

Codecs and encapsulation

In the digital age, there is no single definition of a “broadcast quality” contribution link. For the most prestigious events no compromises are acceptable, so uncompressed HD (at the very least) from the venue to the broadcast center is standard. Other productions can happily accept a good mezzanine connection, and for yet another group or situation, more or less any signal is better than nothing. Often the bandwidth available - and its cost - will determine what broadcast quality makes the most sense.

If dark fiber is available from the venue to the broadcast, center then this can carry multiple uncompressed HD feeds. IP can also be used to carry many HD signals uncompressed. IP is so versatile and ubiquitous though, it can also be used to carry signals with various degrees of compression. This is particularly relevant for occasional-use or temporary connections, for which IP networks are ideally suited.

The choice of codec will depend upon the bandwidth available and the number of signals to be transported. JPEG2000 is an excellent choice for very high quality, albeit at relatively high bitrates. Because it is a wavelet compression scheme, it degrades fairly gracefully. If any part of the contribution circuit comes under pressure, the visible effect would simply be a slight softening of the image.

For other applications, H.264/AVC will provide the right balance of image quality, bitrate requirement and latency. Fast, very high quality H.264 encoders/decoders are now available.

Service provisioning

The IP network provides anywhere-to-anywhere connectivity. Adding or taking down connections is relatively quick and easy, in theory. The problem has been more that the ability to set up lines was restricted to those with the technical expertise to navigate Telco networks, a skill which was largely confined to Telco’s themselves.

To realize IP’s full advantages, broadcast engineers must be able to establish circuits themselves. Engineers need a simple, open interface that mimics the network capabilities and allows routings to be set with just a few clicks.

This can be achieved with a focused management platform. While edge hardware devices provide encoding/decoding and network interfacing, a management platform provides a network abstraction layer for the engineer to establish and monitor services without having to consider detailed configuration settings in each network element.

Such a tool must be also capable of performing network provisioning and using the optimal media transport path through the network infrastructure from source to destination(s). The path-finding algorithm should be technology agnostic, allowing the system to support path finding at different layers in the network stack, at the baseband, Ethernet or MPLS layers.

With the proper level of automation and easy provisioning, inventory management and system monitoring even non-specialists can manage their media transport services on their own, removing traditional user/supplier boundaries.

Service level agreements

Broadcasters can trust IP circuits, even for high quality contribution, and can largely take control of their set-up and management. But they do need to work with their providers to agree on Service Level Agreements (SLAs) that are appropriate for the task.

The nature of a data network is that the Telco will work to fill its capacity. If the broadcaster is not using it then the Telco will find traffic from somewhere else. So broadcasters and Telco’s need to reach an agreement that will guarantee bandwidth when it is needed. Even armed with the appropriate network management tool, the broadcaster or production company still needs to understand the nature of the Telco’s business case for data circuits.

The flexibility of IP connectivity, though, is the real benefit to broadcasting, and it is a transformational change in the way of working. Given there is access to the data network, highly sophisticated two-way broadcast quality links can be established quickly from any location.

The combination of powerful hardware and good SLAs from service providers can combine to deliver very low latency and high resilience broadcast links. Such circuits are extremely cost effective, can be established quickly, and are much more practical than driving satellite or microwave trucks to the broadcast location. IP seems certain to become the dominant connection for contribution circuits.