Media and broadcast companies are preparing for the introduction of ultra-high resolution video signals in their contribution networks. This is not such an easy task as the transport of native HD signals is very demanding on bandwidth. While HD-SDI (High-Definition Serial Digital Interface) and 3G-SDI are satisfied with 1.5 and 3 Gbit/s respectively, emerging ultra-high density interfaces such as UHDTV 4k and 8k (Ultra-High Definition TV) demand connection capacity of several tens of Gbit/s.

Will those bandwidth needs require media companies to compromise video quality and introduce quality-impacting codecs prior to transmitting HD signals over their contribution networks?

Not at all! Technological innovation with optical transmission and photonic switching results in cost reductions with high-capacity transport. This is however only a part of a more complex equation. Media and broadcast companies will need to take a closer look at their existing networks and ask themselves whether the currently applied technologies can scale to meet the upcoming requirements or whether a more cost efficient technology needs to be considered.

Designing optimized network architecture should be based on a thorough CAPEX/OPEX and performance analysis. The most cost efficient architecture needs to be identified, which combines highest availability, lowest latency, best scalability and most simple operations. The key functions of the contribution network such as video signal adaptation function, provisioning flexibility and resiliency have to be evaluated in a holistic way including the optical transmission network.

Typical contribution network architecture is shown in Figure 1. Generally, an operator or specialized managed service provider offers fixed bandwidth services using standard interfaces while the adaptation of the video signal and functionality to enable the required flexibility and resilience is the responsibility of the broadcast or media company.

Let’s look at a realistic growth scenario. A broadcaster is faced with the need to migrate to high-capacity native video signals including SD-SDI to HD-SDI / 3G-SDI. What options should be considered? An analysis starts with understanding the requirements. Table 1 lists the currently applied and emerging native video interfaces as well as audio and video codecs. In addition, inter-office traffic needs to be supported; high capacity file transfer defines the required capacity.

For video applications, capacity growth of emerging (U)HDTV is tremendous. As most interfaces are rarely supported by connectivity service providers today, an adaptation function is required as shown in the figure above. Flexibility and resiliency are typically enabled by OTN, MPLS or proprietary technologies. These networks provide multi-service capability and are designed to deliver guaranteed performance (QoS).

A continuation of existing network design strategy will require scaling the transport, aggregation and adaptation layer. It is fair to assume that cost will grow linearly with bandwidth. This might not be the best option as the additional revenue opportunity will not justify such a cost increase. Caught in a cost-revenue trap, alternatives become vital.

Recent technological innovation with optical transmission and photonic switching might open up opportunities addressing this challenge.  As high-speed video interfaces are introduced, the switching granularity required to maintain flexibility and resiliency in contribution networks can become coarser.

Figure 2 proposes a simple method for introducing cost-efficient connectivity for high-speed native video interfaces. Instead of mapping onto an electrically switched flexibility layer, the signals are directly mapped onto wavelengths using standard compliant OTN framing. Resiliency can now be provided by photonic switching, which is a widely applied approach and a perfect fit for protecting high-capacity signals. SD and HD video signals can be mapped into a common data stream together with MADI digital audio signals and Ethernet traffic, resulting in a very cost-efficient connectivity solution. As a side effect, media and broadcast companies can benefit from lower latency and higher availability as complexity is reduced and network layers get thinner.

Picture3

Complex layering as shown in Figure 1 is now avoided by a straight forward approach, which meets availability requirements by applying cost efficient resiliency mechanisms and does not compromise quality by allocating dedicated resources. Switching of legacy video signals can be performed by existing legacy equipment, while switching of high-speed native video signals is enabled on the photonic layer at lower cost, bridging optical signals and using two disjoint paths.

As most network operators are implementing or already have implemented flexible photonic networks, switching of the native video signals is ideally performed in the operator network. The operator can provide its media and broadcast customers with a programmability interface, which allows taking control of a virtualized network partition operated and maintained by the network operator as shown in Figure 3. Respective interfaces are specified by ONF (Open Networking Foundation). A recent demonstration has proven practicability of this approach. For more information, check out the joint BT/ScheduALL/ADVA demonstration we held at this year’s SDN and OpenFlow World Congress.

Looking further out in the future, native video signals might be mapped into standardized Ethernet frames. This will allow using Ethernet physical interfaces, which come at significantly lower cost compared to specialized video interfaces. Previously applied aggregation and adaptation layer will become replaced by an Ethernet aggregation layer, which can be implemented with emerging cost-efficient, high-capacity white box Ethernet switches as shown in Figure 4. Those switches are developed for highest-capacity data center connectivity demand and packed flows can be programmed through open interfaces in the same way as shown with the optical network with Figure 3.  This Ethernet aggregation layer might either be provided by the operator and the media company is self-providing Ethernet flows over this network. Alternatively, the media company could operate the Ethernet network and self-provision high-capacity wavelength services from an operator.

Picture5

To sum everything up, the above outlined network strategy shows a smooth migration of video contribution networks from SD to HD even being capable to meet high bandwidth demand of emerging UHDTV services. Presently applied aggregation technologies will not be able to scale in a cost efficient way. A direct mapping of high-capacity signals onto optical wavelengths combined with an integrated aggregation function simplifies the layering, reduces the network cost and provides better performance. The scalability of such network architecture is not restricted by bottlenecks of electrical switching layers.