Italian broadcaster Mediaset selected a network based on DWDM technology to connect its Rome production centers. Image courtesy of Cisco Systems.
In a simpler time, video traversed systems on predictable (copper) circuits. Starting in the early days of monochrome systems, analog video exclusively used coaxial cable and connectors, which were intended for RF usage. Initially, many systems used UHF connectors, and beginning in the late 1970s, BNC connectors were increasingly used in manufactured products. The 75Ω nominal impedance was universal, and the bandwidth of under 6MHz made runs of more than 1000ft practical without amazing efforts to maintain equalization.
Complex systems were made practical by this universal interface. There was effectively no difference in the universal interface between systems in 525/30Hz and 625/50Hz countries. SMPTE and other standards organizations had a lot to do with establishing and maintaining effective standards for interchange based on these common elements, sometimes using MilSpec.
The transition to color somewhat modified the status quo, though not the physical layer of the interconnect. With the sensitivity of the subcarrier-based color systems of PAL and NTSC, it was important to establish tighter control over equalization, differential phase and differential gain on long circuits. In an effort to improve the quality of signals that were often degraded by long cables and microwave transmission, some manufacturers began to consider newly-developed fiber-optic technology to modulate video signals onto beams of light.
The first systems more than 20 years ago used strictly analog modulation and still required careful control of the same parameters to keep acceptable signal quality. Long fiber transmission suffers the same degradation as any other analog medium, including signal level caused by absorption of the light in the extended length of a circuit. Good modulators and demodulators had to be used to ensure the recovered signal was a good representation of the original. Audio was usually added as a modulated subcarrier above the visual spectrum. Some of this equipment is still in use today.
The advent of high-bandwidth HDTV signals at first seemed to require fiber interconnection for any reasonable interconnect distance. SMPTE worked on fiber interconnect standards for HDTV at the same time coax interconnections were developed. Manufacturers did an outstanding job of building interconnect hardware using SMPTE 292 transmitter and receiver chips for coax, making HDTV possible in reasonable facilities without fiber. But the march of technology in our industry to wider bandwidths continues to put increasing pressure on copper infrastructure and make fiber more attractive, for technical reasons at least.
Fiber interconnect has been ubiquitous in the IT and telephony industries. A major breakthrough was achieved in 1966 by Charles K. Kao at Standard Telecommunications Laboratories, when attenuation below 20dB per kilometer was first demonstrated. Koa correctly determined that the barrier to low attenuation was impurities in the glass itself. In the last 40 years, the capacity of fiber systems has grown to include deployments of Dense Wavelength Division Multiplexing (DWDM) fiber carrying multiple OC-192 rate (9.6Gb/s) signals. Using this approach, data rates currently extend to nearly 400Gb/s. In our industry, fiber is used for high bandwidth signals, such as SMPTE 292 (1.485Gb/s), as well as for a number of other important applications. These include:
Extension of L-band satellite signals from teleports to receivers some distance away.
Connections between two buildings where separate power sources could produce ground loops.
Connection for HDTV cameras in remote and studio venues.
ENG connections for remote cameras.
Trunking of multiple video signals between two sites, or within one site when cabling space is at a premium.
As a medium to connect stage boxes to audio mixers (and variants on that theme).
Each of these applications has a spectrum of variations. For instance, one manufacturer introduced a new type of patch panel last year with both optical and electrical connections. A signal can be patched using conventional coax patch cords, but the signals are extended to their destination over multiple fibers, saving considerable weight and space. This was initially developed for mobile units, but it is not hard to see how this might apply for large facilities needing to add HDTV infrastructure to crowded cable trays.
For optical interconnect to be ubiquitous in our industry, this kind of blurring of the distinction of what the interconnect is will need to proliferate. We are beginning to see digital routing switchers with fiber on the back plane, as well as a wide variety of other modular products from several manufacturers. As our industry embraces telecom technology more, we will see more penetration of their industry-standard solutions.
The issue, as in almost everything in life, is the cost. The cost of the components for electrical connections is still cheaper than fiber. This is partially offset by the low cost of the actual fiber itself, with multiple strand fiber costing about the same as single runs of coax. At one time, the cost of termination was considerably higher due to the labor needed to cleave and polish the ends. New termination technology has reduced the handicap in labor cost considerably, though not completely.
It is also true that the number of technicians and installers skilled in terminating fiber is much lower than those skilled in coax termination. Much progress has been made, and system integration firms now train technicians in fiber termination routinely. With connections to video servers and transmission links requiring fiber, this can only grow over time.
It is reasonable to expect photonic routing to show up in broadcast in the next few years, with no conversion back to electrical signals necessary for switching. Currently, that is pretty pricey, but that has a way of healing itself in our business. At one time, very little fiber was used at all. Look around you today, and you will see loads of it installed in broadcast-related applications.
John Luff is senior vice president of business development at AZCAR.
Send questions and comments to: firstname.lastname@example.org