Several industry trends are coming together to give service providers an opportunity to increase revenues from digital video transport services. Over the next couple of years or so, television stations will be converting their broadcast signal format from traditional analog to digital TV, which will ultimately pave the way for mass consumer deployment of high-definition TV. Emerging applications such as pay-per-view events, interactive TV, videoconferencing, post-production for filmmaking, distance learning, telemedicine, and remote security monitoring are creating demand for digital video services.
Digital video is in line with the general trend in the consumer electronics industry, which is quickly moving to the digital equivalents of traditional analog devices, including the digital camera, digital camcorder, DVD player, and 3G wireless phone. It has different traffic characteristics than voice and data, and transporting all three across a single converged transport network can be challenging. Voice traffic maintains a constant rate and is sensitive to delay across the network. Data traffic runs at variable rates and is generally bursty in nature. Digital video traffic shares a number of voice and data characteristics, while consuming large amounts of bandwidth and requiring a deterministic (i.e., delay- and jitter-sensitive) network. In addition, video services often travel in one direction across the network (i.e., unidirectional), while voice and data services travel in both directions (i.e., bidirectional).
Previous attempts for video transmission across a MAN required a costly overlay network separate from the voice and data networks, because none of the current digital video interface standards were designed for WAN transport. As a result, initial approaches to digital video transport over SDH were problematic due to the inefficient manner that video was mapped into the standard SDH timeslots of STS-1, -3c, -12c, etc. Traditional solutions typically mapped digital video signals into ATM, which was then mapped into SDH. ATM was used to supply a virtual layer for providing finer bandwidth granularity to allow additional video channels to be transported over SDH. However, significant cost penalties were incurred due to overlay network complexity, additional capital equipment needed, and specialised training requirements.
The development of next-generation SDH technologies such as virtual concatenation (VC), link capacity adjustment scheme (LCAS), and generic framing procedure (GFP) gives service providers a new opportunity to market video services without replacing their metro SDH networks. Multiservice provisioning platforms (MSPPs) that support these technologies allow service providers to deliver digital broadband video, broadband data, video on demand (VoD), telephony, and other business services over converged networks.
Digital video requires enormous bandwidth. An uncompressed NTSC signal requires a transmission capacity of >200 Mbits/sec. That pales in comparison to HDTV, which requires raw bandwidth in excess of 1.5 Gbits/sec. Advanced video compression schemes are available that can compress digital video signals to reduce their bandwidth requirements, while maintaining entertainment quality. The Motion Picture Experts Group (MPEG) has developed MPEG-2, the most widely accepted compression method. MPEG-2 is an ISO/IEC-13818-1 standard and defines the syntax and semantics of bit streams in which digital video is multiplexed and transported.
The asynchronous serial interface (ASI) is a standard developed by the European Digital Video Broadcasting (DVB) Standards Association designed to provide simple transport and interconnection of MPEG-2 streams from different manufacturers' equipment. Equipment supporting this widely accepted standard includes MPEG-2 encoders, receivers, multiplexers, servers, and quadrature amplitude modulation devices.
The digital video industry is working to provide additional compression algorithms that will further reduce digital video bandwidth requirements and make it possible for video transmission over broadband access technologies such as DSL and cable-modem. MPEG-4 AVC, or ITU Recommendation H.264, a joint effort between the International Telecommunication Union-Telecommunication (ITU-T) and MPEG, improves compression efficiency to make it possible to deliver entertainment quality video at transmission speeds of <2 Mbits/sec.
While many next-generation technologies exist for optical transport, SDH still dominates public and to a large extent private networks. Recent activity within ANSI T1X1.5 and ITU-T committees has focused on enhancing SDH standards to create a better method for transporting video. The publishing of new mapping, switching, and interface standards gives service providers an opportunity to offer emerging digital video services without replacing their metro SDH networks. The new VC, LCAS, and GFP standards give equipment providers and subsequently service providers a cost-efficient solution for transporting digital video over existing reliable SDH networks without significant network-upgrade expenditures.
VC provides a finer granularity of SDH bandwidth mapping for transporting various video and data interfaces (STS-1-nv, -3c-nv). For example, a 270-Mbit/sec DVB-ASI video stream carried over an STS-1-5v using VC results in bandwidth efficiency of 89%. VC is an inverse multiplexing technique that enables an arbitrary number of low-order (VT1.5) or high-order (STS-1/-3c) SDH channels to be bundled into one virtual concatenation group (VCG). Only the source and the destination nodes need to be VC-capable since virtual concatenation is transparent to intermediate nodes. Thus, when carriers want to introduce VC support into their network, only the transport nodes at the add and drop points have to be upgraded to provide this function. VC enables a service provider to offer services that don't map neatly into fixed SDH timeslots without wasting bandwidth. That is especially important for Ethernet and video services, as shown in the Table.
LCAS provides a bandwidth on demand mechanism for capacity changes, enabling rapid service activation and upgrades without traffic interruption. It also allows auto removal and auto recovery of failed paths. LCAS provides a control mechanism and protocol that can dynamically increase or decrease the number of STS member links of the VCG to meet the bandwidth needs of an application, without affecting existing traffic. Signaling messages are exchanged in the overhead bytes of the synchronous payload envelope between the network elements. LCAS also provides a means of automatically removing member links that have experienced failure and recovering the member link that has cleared the failure condition. LCAS enables efficient service upgrades and improved response to network outages by eliminating the need for manual provisioning (see Figure 1). This capability helps service providers reduce the number of truck rolls, improve customer service, and lower operational costs.
GFP allows different types of traffic to be mapped into SDH tributaries or virtually concatenated pipes for transport across SDH networks. It is defined in ITU-T G.7041/Y.1303 for framing or mapping different protocol streams into virtually concatenated STS-1 or STS-3c payloads within a SDH frame. Compared with other framing procedures such as point-to-point protocol and X.86 that employ high-level data-link control, GFP has a more deterministic overhead and lower processing requirement.
Two types of GFP mapping formats exist: frame-mapped GFP (GFP-F) and transparent GFP (GFP-T). GFP-F is normally used to encapsulate packet/ frame-based protocols such as Ethernet and IP (see Figure 2). GFP-T is optimised for protocols that utilise 8B/10B line coding, including DVB-ASI,
Gigabit Ethernet, and major storage protocols. 8B/10B encoding provides an equal transition density of 1s and 0s in the data stream to ensure that timing can be recovered from the data (see Figure 3).
MSPPs that support VC, LCAS, and GFP provide a cost-effective way to consolidate voice, data, and video services onto the existing SDH-based network infrastructure. They combine the transport capabilities of multiple add/drop multiplexers, the switching/grooming capacity of a small digital-crossconnect system, and next-generation TDM, IP, video, and storage interfaces into compact platforms.
For unidirectional digital video broadcast, MSPPs support efficient transmission of compressed (DVB-ASI) video-signal formats directly over SDH links between hubs. Since VoD and near-VoD applications require a return path, digital video can be transported over Ethernet over SDH interfaces to leverage reliable and existing SDH networks. On-demand digital video signals are carried from the video server over the IP/Ethernet/SDH network into the remote hub for video distribution into homes.
In addition to point-to-point applications, broadcast applications for multinode video distribution via SDH are also fully supported. With the robust broadcast capabilities of SDH, adding another protocol layer such as IP to provide broadcast functionality is not necessary. SDH is widely used to broadcast video signals from a cable TV provider's headend location to remote distribution hubs. In this application, incoming video signals are duplicated to multiple output ports and sent to several distribution hubs. The drop and continue function of SDH is used to extract the video signal from the main ring and direct it to other feeder rings while keeping a copy of the video signal going around the main ring. Typical deployments use OC-12 (622-Mbit/sec) and OC-48 (2.5-Gbit/sec) rings for the feeder network, with an OC-48 or OC-192 (10-Gbit/sec) ring configured as the main ring. Both video transport modes are illustrated in Figure 4.
Figure 4. SDH supports both point-to-point and broadcast digital video links. For unidirectional digital video broadcasting, multiservice provisioning platforms support efficient transmission of compressed video-signal formats directly over SDH links between hubs. For broadcast video, incoming video signals are duplicated to multiple output ports and sent to several distribution hubs. The drop and continue function of SDH is used to extract the video signal from the main ring and direct it to other feeder rings while keeping a copy of the video signal going around the main ring.
SDH provides STS path-level protection for video signals to support uninterrupted transmission. The MSPP at the headend sends two copies of the video signal across diverse paths around the ring. If any MSPP on the ring detects a failure in the main path of the video signal, the MSPP will initiate a switch to the alternative path for protection. This switch is completed within 50 msec.
The development of next-generation SDH technologies gives service providers the option of offering video services without replacing their metro SDH networks. MSPPs added to these SDH infrastructures, allow service providers to create digital video transport networks that feature the same carrier class robustness that service providers—and their customers—expect.
- ITU-T G7041, Generic Framing Procedure (GFP), 2002.
- ISO/IEC DIS 13818, Information Technology - Generic Coding of Moving Pictures and Associated Audio Information Systems.
- ITU-T G.7042, Link Capacity Adjustment Scheme (LCAS) for Virtual Concatenation Signals.
- ITU-T G.707, Network Node Interface for the Synchronous Digital Hierarchy.
- ANSI T1.105.01-199x, Draft ANSI Standard, SONET Automatic Protection Switching.
- ANSI T1.105-1991, Digital Hierarchy - Optical Interface Rates and Formats Specifications.
- ETSI (CENELEC) EN 50083-9 - Cabled Distribution Systems for Television, Sound and Interactive Multimedia Signals, Part 9: Interactive for CATV/SMATV headends and similar professional equipment for DVB/MPEG-2/MPEG-4 transport streams (DVB Blue Book A0101), Annex B, Asynchronous Serial Interface, 1998.
William Yue is senior product planner and David Gutierrez is product manager at Fujitsu Network Communications (Richardson, TX). They can be reached at firstname.lastname@example.org or email@example.com.