System.NullReferenceException: Object reference not set to an instance of an object. at DotNetNuke.Framework.DefaultPage.OnLoad(EventArgs e) in e:\websites\\public_html\Default.aspx.cs:line 791 Future-proofing | TvTechnology


June 1, 2007

When purchasing a transmitter, broadcast engineers need to select a system that will be a long-term asset. This may be difficult, however, without knowing how long a station will broadcast in the analog format and without knowing which digital format the station will switch to.

For example, in the past broadcasters had to plan for one of the color encoding standards while still in a black-and-white world, or plan for stereo in NICAM or BTSC while transmitting mono. Once a broadcaster selects a coding standard, life becomes much easier.

Related Articles

What Tom Said: Tying Twitter’s Future to Broadcast TV

HEVC Audio: Based on the Past, Headed for the Future

Convergence—The Wave of the Future

Broadcasters used to specify power level, spurious mask, channel frequency raster, frequency stability, video standard, blanking level, sound system, white level, clamping method and many other features before a custom-made transmitter could be manufactured.

Early digital transmitters deployed in test networks continued with this habit. Eventually, broadcasters had to deal with DVB or ATSC, but according to the territory, once the modulation scheme was chosen, they could plan for the future.

Dual-mode transmitters allow TV broadcasters to prepare for the transition from analog to digital, but now rules of the game are changing. New standards are being developed at an accelerating pace. Alliances between silicon manufacturers and broadcasters have shown that a new standard can be brought to air within months, not decades.

Even more impressive, consumer goods undergo the same accelerated development cycle. Products hit the shelves of a superstore near you in industrial quantities even before a standard has been published. Coping with these new challenges is difficult.

Software-defined transmitters

A new generation of broadcasting equipment provides a solution. Software-defined transmitters offer engineers a system that is compatible with existing and future standards.

The term “software-defined radio” has been around for some time. Digital signal processor (DSP) manufacturers coined the term at the end of last century. They were eager to use a new acronym to push DSP sales into areas where it was not obvious they could deliver a real advantage. Software-defined radio is now paired with any equipment that involves digital signal processing and can be modified by loading a new code.

This flexibility was the key to the dual-mode equipment based on digital signal processing introduced four years ago. In a digitally processed dual-mode transmitter, a single digital modulator engine loads software to achieve different operation, as opposed to a transmitter with two separate modulation engines that are switched to cope with dual-standard requirements. While this solves the dual-mode requirement, it does not prepare a transmitter for future standards.

The inner workings

Software-defined transmitters are designed with two nested digital layers that achieve complete flexibility and provide for future-proof evolution.

The outer digital layer is an insulating layer. It takes care of interfacing the external world to the internal processing. The outer digital layer insulates the inner cores from external dependencies, such as clocks and network interfaces. This allows internal processing to run asynchronous from any external input or output data flow. (See Figure 1.)

Front-end processing performs network adaptation of input data. Features like hierarchical modulation, single frequency network (SFN) operation, transport stream remultiplexing and multiple feeds management all require complex network management at every transmitter.

Transmitter algorithm processing is a computationally intensive task. Data in modern, nested forward error correction schemes require multiple rate changes and complex interleaving schemes. The modulation process itself requires a large number of iterations with access to a large amount of data.

RF generation shapes what the spectrum analyzer shows. A spectrum analyzer needs sharp channel filters, linear and nonlinear precorrection filters and fast interpolation filters. Again, this will require important processing power.

The independence of the internal processing layer to external data flow is the key to universal performances. This guarantees that any algorithm can be built inside the silicon by programming, and, at the end, it also guarantees that the system is future-proof.

The only limit to algorithm implementation is the total processing power of the selected silicon. A high-end field-programmable gate array (FPGA) provides a reasonably priced package up to 320 DSP blocks, plus a large amount of logic resources with a total equivalent gate count between 2.5 million and 10 million. This translates to an equivalent processing power in the range of 120,000 million to 800,000 million instructions per second (MIPS), which is enough for any actual and future digital modulation scheme.

The two-layer approach allows the broadcaster to compute any modulation scheme independent of clocks. Input rate, intermediate word rates and output symbol rates are totally virtual. This permits any existing and future standard to run without dedicated oscillators.

At the physical level, the two layers are built inside a single silicon platform. This means that all interconnections are run at maximum speed with maximum protection from external noise. Power consumption is minimal, and users can get the maximum from a silicon platform.

Software-defined transmitters must not be confused with digital-ready transmitters. Software-defined transmitters are 100 percent digital transmitters — even when they are broadcasting in analog mode with existing analog standards — and can evolve to future standards. Digital-ready analog equipment means that all or part of the transmitter can be replaced by the manufacturer to achieve digital operation.

It is easy to understand the difference: A piece of software can be changed at a reasonable cost and loaded into a transmitter with minimal effort while a hardware change requires a serious retest to guarantee reasonable performance.

Features of a future-proof transmitter

Aside from being software-defined, a future-proof transmitter requires other features. Distribution networks in the near future will be a mix of different delivery media. Satellite, microwave and wired feeds will coexist, with multiple feeds providing redundancy. Therefore, a future-proof transmitter must have multiple inputs, including legacy ASI and the now ubiquitous Gigabit Ethernet. It must also support any protocol (such as TCP/IP, UDP, RTP and the Pro-MPEG COP3/SMPTE 2022). And it should be able to sync to any packet length, synchronization method or sync byte. The input processing must be able to remultiplex incoming signals to a new stream, with in-band and out-of-band management.

Spectrum is the most valuable commodity in the RF world. A network's frequency management plays a vital role in exploiting this resource to maximum. The transmitter RF component must be capable of centering a signal at any output frequency in the chosen band, typically the UHF band.

As SFNs become commonplace, they must be able to generate exact center frequencies, both with 1Hz increments and with fractional hertz increments. Channel bandwidth must be flexible as well, and the total bandwidth must be in excess of 40MHz to accommodate efficient precorrection patterns.

Symbol rate must be independent from RF frequency, and this is achieved in the inner layer. Nevertheless, both symbol rate and output frequency must be capable of being locked to an external reference or to a GPS signal for SFN application and for inherent stability.

Total cost of ownership

Remember, sound stability of a network transmitter will always reduce cost of ownership — another key element of a future-proof transmitter.

As more media (cell phones, mobile TV, traditional broadcasting and legacy carriers) share the same facility, real estate within sites becomes more expensive. Single FPGA digital software-defined transmitters, with all legacy interfaces built in, are much more compact than traditional transmitters. Therefore, they allow site cost savings.

The most powerful reason to move to a software-defined transmitter is the cost of the transmitter itself. Software-defined transmitters have a sound cost advantage compared with analog solutions with the same performances and the interfaces.

Can broadcasters really afford to buy something that is not future-proof? I don't think so.

Mike Bargauan is R&D director for MB International Telecom Labs.

Receive regular news and technology updates.
Sign up for our free newsletter here.