FOCUS: Improving Digital Media Satellite Networks
by Keith Dunford
Since the early days of transcontinental, intercontinental and global networking of television content, great attention has been directed to maintaining the integrity and quality of video and audio content. Prior to the introduction of digital television in the late 1980s, satellite links were analog and subject to deterioration in signal-to-noise ratio and other distortions. This resulted in lowered quality of the video and audio signals delivered to consumers.
Analog systems required a large amount of transponder bandwidth, at first one 525/60 or 625/50 television signal in one 36MHz transponder. This was later improved to enable two such programs to be carried in the same bandwidth. With the advent of MPEG-1 digital compression, very little changed in bandwidth requirements, however deterioration in baseband signals was substantially eliminated. The advent of MPEG-2 in the early 1990s allowed greater compression ratios enabling up to six standard-definition (NTSC / PAL) programs to be carried in a 36MHz satellite transponder.
This virtual increase in transponder capacity and availability effectively lowered the cost of satellite links, delivered improved quality, and set the stage for affordable global, national and even local satellite news gathering.
With the introduction of Advanced Video Codec (AVC) technology and specifically MPEG-4 /H.264 standards, a further improvement in performance and cost savings have been made possible. Content encoded in AVC/H.264 can be carried over a satellite link requiring only 50 percent of the bandwidth used for the equivalent of MPEG2 encoded content. The standard has also afforded improvements in perceived video quality.
The use of high performance compression systems and digital satellite links has delivered the greatest benefit to Digital Satellite News Gathering (DSNG) operations, allowing real time, live news production to achieve reduced cost and improved performance. The transition from standard definition (SD) to high definition (HD) newsgathering has required new satellite transmission technologies that address the requirement of higher encoded data (bit-rates) of HD content that are at least twice that used for broadcast quality SD content. However the available channel bandwidth for DSNG is substantially the same, and in terrestrial microwave DENG, this has recently been reduced by 5MHz, from 17MHz to 12MHz per channel, in restructuring of the 2GHz Broadcast Auxiliary Services (BAS) band.
Improved compression systems such as AVC/H.264 have decreased the data rate required for broadcast standard HD and SD content. Improved channel coding technology has also increased the data throughput of satellite and microwave links. These distinctly different technologies, when combined in a satellite transmission system, have made possible the transmission of HD content in substantially the same bandwidth previously used for SD content.
The move from DVB-S to the newer DVB-S2 standard has increased the data carrying capacity of a satellite transponder channel by as much as 30 percent when using equivalent modulation schemes, as indicated in Table 1. The increased throughput of a DVB-S2 enabled channel is made possible by improved Forward Error Correction (FEC) codes such as inner code Low Density Parity Check (LDPC), concatenated with a Bose-Chaudhuri-Hocquenghem (BCH) outer code.
The new FEC codes provide an improved signal-to-noise ratio at the receiver (lower Eb/N0) for identical modulator constellation and FEC overhead compared to DVB-S. With the objective of maintaining a constant system Bit Error Rate (BER), the lower Eb/N0 values for DVB-S2 allow use of higher-order modulator constellations to achieve a BER equal to that for a DVB-S channel. This provides DSNG operations the option of using higher order modulation schemes such as 8PSK as opposed to widely used QPSK in DVB-S systems. This provides approximately twice the channel payload capacity, without increasing existing satellite channel bandwidth.
The significant improvement in channel coding and the use of higher-order constellations are increasing virtual channel payload capacity. DVB-S2, with 8PSK modulation and an FEC of 3/4 allows broadcast standard HD content (720P/60Fps or 1080i/30Fps) video, audio and metadata with multiplexed data rate in the order of 10Mbps to be transported within the same 5.5MHz satellite channel previously needed for broadcast standard definition content.
Fujitsu has incorporated both H.264 advanced video compression and DVB-S2 modulation technology into the high performance IP-9500e encoder and IP-9500d decoder system shown in Figure 1 above. This system, along with other improvements in digital satellite content contribution and distribution, will be demonstrated during the upcoming National Association of Broadcasters Convention at Las Vegas in April.
In addition to improving bandwidth utilization and network performance, satellite networking of television content many times requires several intermediate links to complete end-to-end delivery. This is referred to as concatenation and has been an area of significant attention in regard to maintaining video and audio integrity. In earlier analog systems, demodulation and re-modulation of the television signal at each en-route satellite terminal contributed to signal degradation particularly video signal-to-noise, which in many instances reduced the image at the point of delivery to less than that normally acceptable for broadcast television.
The addition of digital video compression in satellite networks has created new challenges in mitigating the effects of degradation caused by concatenation in the multiple encode-decode process associated with intermediate digital turn-around operations. One particular problem is that of maintaining the integrity of the 4:2:0 chroma component where an HD-SDI interconnect is used between concatenated decoders and encoders. This relates equally to both the widely used MPEG-2 and newer MPEG-4 (H.264 / AVC) codec standards.
HD-SDI is inherently a 4:2:2 interconnect which requires that a 4:2:0 stream be up-sampled and then down-sampled at each digital turn-around. This can, within as little as 4-5 concatenated links, result in visual blurring of the color image to the point where the content can be severely degraded as shown in Figure 3.
This has particular importance in Digital Electronic News Gathering (DENG) where as many as 10 concatenated operations can take place before final delivery to network affiliates and others. Video encoded by AVC/H.264 using 4:2:0 chroma sampling can provide significant operational and economic advantages, but can be subject to severe degradation due to concatenation of decoders and encoders in the chain that use SD/HD-SDI as the interconnect. Serial Digital Interface (SDI) is inherently a 4:2:2 interconnect which requires that a 4:2:0 stream be up-sampled and then down-sampled at each digital turn-around as shown in Figure 3. This can, within as little as 4-5 concatenated links result in visual blurring of the color image to the point where the quality and integrity of HD content can be severely compromised.
There is now a solution to this long standing problem, and soon to become available in the Fujitsu IP-9500 series of encoders and decoders. It uses a precise set of filters which process the 4:2:2 chroma component, down-sample it with a low pass filter and 2:1 vertical sampler to create a 4:2:0 output. At the receive end, or intermediate turn-around point, the 4:2:0 chroma component is accurately up-sampled to re-create a 4:2:2 signal as shown in Figure 3. The result is no degradation, allowing an encoded 4:2:0 stream to be carried efficiently over a concatenated satellite link.
In order to achieve non-degraded 4:2:0 chroma resolution throughout the concatenated chain it is necessary that all encoders and decoders in the link use identical filter sets. However, interoperability with other decoders has been taken into consideration, allowing a 4:2:0 encoded video stream to be decoded by other manufacturers decoders not fitted with the precision filter technology. Such decoders may, however, be subject to normal concatenated chroma degradation.
The non-degraded filter technology has been proven in tests demonstrating its ability to protect the integrity of the 4:2:0 chroma component in up to 16 concatenated encode and decode operations.
The combination of several new technologies and standards including; DVB-S2; AVC. / H.264; and Non-degraded 4:2:0 has the potential to significantly improve the performance, operation and quality of digital satellite networks, and at the same time delivering valuable cost savings.
About the author
Keith Dunford is currently a consultant for the Video Solution Group, Fujitsu Computer Products of America, Inc. Keith is Managing Partner of The Exam Group, a consulting company engaged in technology evaluation and integration for the Electronic Media and IT industries.