You are here: Home Page » Storage & Networking » Storage & Networking Archive » Raising the standard

Raising the standard

Sep 1, 2007 12:00 PM, BY TIM SHEPPARD

MPEG-4 AVC allows for a new generation of content delivery platforms.

The MPEG-4 family of audio and video coding standards and related technology has been a huge enabler in driving forward the growth of audio and video delivery across multiple platforms.

While the MPEG-4 standard actually contains 23 different parts, the part most widely referred to now in terms of content delivery is MPEG-4 Part 10 (ISO/IEC 14496-10) — also known as MPEG-4 AVC and H.264. The H.264 reference gives a clue to the standards lineage coming from the ITU Telecommunication Standardization Sector (ITU-T). The broad base of support from the Moving Pictures Expert Group and ITU during the standard's development by the Joint Video Taskforce has led to widespread acceptance and rapid adoption of the standard between vendors and operators following its initial introduction in 2003. In this article, MPEG-4 refers to MPEG-4 Part 10 because despite potential confusion with the prior MPEG-4 ASP standard, the newer standard is becoming the de facto meaning.

Interoperability

One of the major benefits of MPEG-4 is that it is an open standard that has been adopted by organizations across the globe, including — notably for TV applications — the DVB and SMPTE. The high level of interoperability that this provides — along with the extent of its use — has led to rapid development that may not have occurred if a range of proprietary encoding solutions had been developed in its place. Many products and devices now support MPEG-4.

Figure 1. The MPEG-4 AVC standard improves HD bandwidth efficiency. Hardware encoder performance lags behind softwarebased encoders.
Click image to enlarge.

Just like its forebear MPEG-2, MPEG-4 includes multiple toolkits and a range of complexity to allow developers to improve video and audio quality while reducing the necessary transmission bandwidth. There is huge potential to continue to improve MPEG-4 in terms of the implementation of encoders in particular, beyond the level of sophistication available today. The evolution and improvement of the standard is mimicking the development of MPEG-2 10 years ago but at a far greater pace, driven by the wider understanding and larger number of organizations that have adopted the new standard.

The fact that MPEG-4 performance upgrades can (on some makes of encoders) consist purely of software is a great asset for developers and solution providers who can offer ongoing performance improvements through field upgrades. The capital expenditures (CAPEX) benefits of investing in hardware that can support such upgrades are clear. (See Figure 1.)

Bandwidth

One of the main contributors to MPEG-4's success is the need to make the most efficient use of the available bandwidth for contribution, distribution or final transmission. For each delivery platform, however, the bandwidth scenario is slightly different.

In the case of satellite operators, many providers are trying to differentiate themselves from their competitors by providing high-quality HD services. Despite having a comparatively large amount of available bandwidth that can be increased by adding additional transponders, the satellite business model demands a relatively large number of HD channels to deliver more than any other broadcaster, providing competitive advantage and increased average revenue per user (ARPU).

Figure 2. A typical MPEG-2 GOP sequence
Click image to enlarge.

Using an MPEG-2 solution, running at between 15Mb/s to 20Mb/s only allows a maximum of two HD channels on a typical transponder, which does not make commercial sense. However, by using MPEG-4, the satellite operator can provide about six channels per transponder, with the likelihood of increasing to eight or 10 channels per transponder as MPEG-4 technology evolves in the near future and when combined with the DVB-S2 transmission standard.

IPTV operators also have a strong interest in making the best use of their available bandwidth, but for different reasons. Most IPTV operators use DSL or ADSL 2+ technology, which introduces bandwidth limitations when the exchange (DSLAM) is far away from the user's home. In order to provide acceptable reach for delivering SD and especially HD services, the operator needs to keep the bandwidth requirements as low as possible while maintaining an acceptable quality of service. By using MPEG-4, the operator can reduce the bit rate substantially while maintaining the same level of quality and significantly increasing the number of users that can receive the service without the need for additional network CAPEX. This can be the difference between a working business model and a failing one.

For DVB-T, most transmissions are in the range of 20Mb/s to 24Mb/s. Therefore, achieving HD using MPEG-2 at around 15Mb/s to 20Mb/s would involve allocating one HD service per channel, which is a highly inefficient use of the available spectrum. To offer multiple HD services, a reduction in bit rate is required, and implementation of MPEG-4 allows at least three HD services per channel, with the possibility of four in the near future.

Figure 3. A typical MPEG-4 hierarchical GOP sequence

There is also a driver for cable operators who, while theoretically have more bandwidth than IPTV and DVB-T, are increasingly using that bandwidth for complementary services such as data and VOD to provide competitive advantage and increased ARPU. Consequently, the available bandwidth for high-value services such as HD has been reduced, necessitating an improvement in the encoding technology being used.

MPEG-4 technology is not only applicable to final transmission, and many content owners (e.g. programmers) have adopted the standard for their contribution and primary distribution due to the benefits of lower bandwidth requirements, or higher quality for a similar bandwidth to MPEG-2. This attribute provides some much-needed flexibility for network operators who can now make a choice between minimizing bandwidth and maximizing quality to best service their particular business model.

While making the best use of the available bandwidth is still a key issue for many operators, some have introduced a self-imposed quality threshold of about 10Mb/s to maximize customer experience, especially for HD services. At the other end of the scale, operators with severely limited bandwidth such as IPTV over xDSL are delivering streams at below 5Mb/s.

In addition to the reduction in bit rate compared with MPEG-2, MPEG4 provides significant improvements in terms of video quality. The ringing and blocking artifacts sometimes visible with MPEG-2 encoding are considerably reduced or smoothed in a way that is far more pleasing to the human eye. With skilled use of the available toolsets, it is possible to remove the majority of the artifacts completely. The bit rate does, however, have a huge bearing on the video quality. Some of the more sophisticated techniques that can improve MPEG-4 video quality include: dynamic GOP, hierarchal GOP, support for high profile and single-slice architecture.