Compression: Technology, Need Continue to Grow
The amount of available bandwidth is the only part of the broadcasting sector not growing. How is the latest in compression technology keeping revenues flowing and what lies ahead for broadcasters?
The broadcasting industry needs more bandwidth to meet growing consumer demand for services. Until that becomes available, efficient use of existing bandwidth is becoming more and more critical. "Bandwidth is finite," says David Szelag, vice president of technical operations for Globecast America. "Content is becoming increasingly and readily available on so many different mediums: mobile, 3G, cellular, Wi-Fi, and WiMax. These are the areas that drive consumer convenience, and that’s where we’re going to see the greatest impact of compression. The better compression technologies we have, the better chance we have of succeeding. Compression also opens a lot of opportunities for new markets. The question is how far can we push that envelope and what can we get through the smallest pipe with no perceptible loss of quality?"
DVB-S2 modulation and MPEG-4 advanced video coding (AVC) (H.264) compression have become a necessity for optimizing satellite links used for video broadcasting and point-to-multipoint networks. But while the need is great, adoption of the latest compression technologies is hampered by factors such as legacy networks with plenty of life remaining and the cost of new equipment. "Today, we want to send more, do it in a different way and not pay any more," says Daniel Enns, senior vice president, strategic marketing and business development, for Comtech EF Data. "Demand for bandwidth is infinite and would grow and grow if no cost growth was involved. Bandwidth is in a perpetual spiral of needing more. In order to give you that ability to do differentiated services, we have to do things such as optimizing operational expense, extending the experience to you so you stay with me but don’t lose money."
"There is a significant need for compression today just as there was 10 to 15 years ago when MPEG-2 technology first came on the scene," says Lisa Hobbs, vice president, business development, satellite and broadcast for Tandberg Television. "What we are seeing is an explosion in HD (high-definition) services broadcast over satellite. Clearly that takes up a lot of bandwidth, so the more compression you can get without impacting video quality the better. So we are seeing a move toward MPEG-4. It provides better compression efficiencies than MPEG-2, anywhere from 30 percent to 50 percent bit rate gain depending on video content."
NSR projects that the market for compression equipment sales will reach more than $8.2 billion within five years. According to "MPEG-4 and DVB-S2: Assessing Implementation Schedules for Advanced Video Compression and Satellite Modulation," released in January, the market for DVB-S2 and MPEG-4 equipment will experience strong demand, driven by new system deployments and migration scenarios for five satellite applications: direct-to-home (DTH) and satellite free-to-air (FTA), satellite broadband and IP trunking, headend in the sky, video distribution and contribution, and digital media distribution. Demand for equipment such as MPEG-4 set-top boxes, DVB-S2 receivers, satellite broadband indoor units and integrated receiver-decoders will drive revenue growth.
"In terms of real need, specifically video, HD is the only driver I see," says Hobbs. "If look at SD (standard-definition) channels, there just haven’t been nearly as many new SD channels launched as a move toward HD, because everyone knows HD is the wave of the future. There are roughly the same number of SD channels as in years past, but we are getting HD versions of those channels, so more overall channels are being offered."
Motorola Inc. is active in a number of different segments, delivering services for players such as HBO and Showtime as well trying to capture business from DTH operators, says Neil Braydon, product marketing manager for Motorola’s MPEG-4 Encoding Solutions Group. "Certainly the new compression technology has been key in this delivery of HD. Without the new compression technology, the delivery of HD would have been difficult. It’s such a bandwidth hog. With MPEG-2, delivery would have been limited. We really needed AVC to come along to help enable widespread delivery, and we are seeing HD is very much mainstream now."
Compression also is in demand in the data delivery sector, and an area that is thriving is WAN (wide area network) optimization, says Enns. "WAN optimization is a very booming market today. There is not a network appliance or network product manufacturer that doesn’t have something on WAN optimization. All users want more. They seldom want to pay more but want a better experience. From the perspective of a service provider, we are trying to figure out how to do this without paying more. Any piece of equipment is valued by an offset in operating expenses. How much capital expenditure will I put in to get what kind of operating expense benefit?
"On a satellite, this is about power bandwidth or equivalent of power bandwidth. If I can use same power bandwidth and give you twice the throughput, I’d be a hero," says Enns. "Then the service provider could lower what they charge or bring in twice as many users. To that effect, we’ve been using forward error correction and higher modulation as getting more for satellite bandwidth is what a satellite modem’s entire premise of life is. We’re pushing the envelope until we have no more way to get to nirvana."
"The biggest need for compression is in applications that don’t have decent bandwidth. Prime examples are newsgathering, emergency response/assessment or special forces operations."
— Hildeman, Streambox
Compression advancements also will be key in new broadcasting areas such as IPTV. "All of them have one thing in common, they are interested in compression efficiency. We are doing our best to continue to drive efficiency to able to deliver video at the lowest bit rate. It’s amazing how quickly the bit rate has come down since using AVC technology," Braydon says. "It has become a necessity as the amount of HD delivery has increased. Operators have a lot of commercial pressure to deliver a lot of content, and AVC technology has helped quite a bit."
Gal Garniek, assistant vice president of marketing, North America for Scopus Video Networks, sees the digitalization process opening up more opportunities for technology providers. "In every location in the food chain there are so many opportunities. If some kind of opportunity is filled, then another area opens up. Take the U.S. cable industry. Around 2003-2004, the old digital transition ended and most [multiple system operators] moved all of their signals to the digital domain. That led to a lot of encoders, statistical multiplexing and so forth. "Around 2005, we see how business slowed down, and then on another front, DVB-S, EchoStar and DirecTV, H.264 HD suddenly opened. Tandberg and Harmonic started selling a lot of encoders to this area. Satellite goes in waves. The first deployment in 2005-2006 and then the second in 2006-2007 became huge competition for cable, so cable had to add more. So now operators were adding 50 to 100 channels to their facility, which again is an opportunity for encoders."
Making the Transition
While there are concrete reasons and obvious benefits for companies that make the transition to the latest compression technology, not many are doing it all at once, as many have made considerable investment in their current infrastructures. This means the suppliers have to offer equipment that can work with MPEG-2 and MPEG-4 at the same time. "It’s more than just compression. What we are seeing is the importance of having the products at the edge. It allows customers to take the AVC HD signal off satellite and transcode it to MPEG-2 to deliver into the existing infrastructure," says Braydon. "Our existing customers have such a big deployment of MPEG-2 set-top boxes and cannot jump easily to AVC anytime soon, so we still need to do delivery of MPEG-2 to their customers. HBO is trying to maximize use of their satellite transponders. Then when the signal is received at the cable headends it has to be delivered as MPEG-2. That’s where our transcoding [integrated receiver decoders] have played an important role."
The market expects SD channels to become extinct, but no one has a good idea on when this will happen, so until that day, technology providers need to make sure the equipment can work in all formats. "We are talking to customers who are going all HD but still need SD at receive locations, so they need to downconvert and create SD channels from HD channels," says Hobbs. "Everybody on the distribution side wants to move to an all-HD distribution platform, but the receive equipment has to handle HD and recreate SD, and we believe that will be true for many years to come. We are beginning to reach the point that if broadcasters want to add one or two HD channels they can find the capacity to do that. They maybe even can squeeze that out of existing platforms, but if broadcasters want to add many more, they generally have to go to an MPEG-4 platform. When HBO announced they were launching HD channels they talked about a total of 26, half at the end of 2007. The 13 new HD channels would take a significant amount of bandwidth in MPEG-2, so they really needed an efficient compression system to handle that, so they went with MPEG-4."
The deployment of HD has not yet met industry expectations, but officials across the board remain sure that adoption of the technology along the distribution chain will increase dramatically within the next few years. "We are starting to see quite a bit of talk about 1080p and 1080p60," says Braydon. "Some of the satellite players already are delivering 1080p24 content, which is good for movies but not for other uses where you need 1080p60. The compression technology is virtually there today. The major holdup in getting 1080p60 is the infrastructure. You need all these cameras and encoders. It’s a big, long, wide delivery chain. I see some niche players delivering this in the next year, but it will be quite a while before 1080p60 becomes mainstream.
The investments made by cable players in MPEG-2 infrastructure could give IPTV providers an advantage in terms of efficient use of bandwidth, as the new players are able to adopt the new technology more quickly, says Braydon. "The cable players have been slow to deploy AVC set-top boxes. It’s just a function of how quickly the MPEG-2 boxes will reach the end of life, and they have a significant life," he says. "The IPTV players are greenfields and have deployed AVC straight out of the box. There is no legacy to deal with. The can take advantage of the AVC capacity and that allows them to maximize their use of infrastructure. They need this to compete with the cable players and deliver multiple streams simultaneously. They can deliver multiple services as well as Internet, and the low bit rates of AVC technology is allowing them to do that. The cable players originally thought the IPTV players were not much of a threat, but it’s turning out to be a threat, and the cable players are taking notice."
Along with traditional broadcasting, compression technologies are enabling growth in other areas where delivery of time sensitive information is crucial such as: newsgathering, emergency response and assessment, military, and telemedicine as well as any application for which video is delivered over low-data-rate satellite connections and where last-mile networks are inferior, says Bob Hildeman, chairman and CEO of Streambox. "The biggest need for compression is in applications that don’t have decent bandwidth. Prime examples are newsgathering, emergency response/assessment or special forces operations," he says. "For each of these applications, video is often streamed from remote areas or areas where networks other than satellite networks are unavailable."
A demand for more HD news is driving a new group of users to look at upgrading their compression technology, says Damon Semprebon, director of product management for Comtech TV. "That tends to be MPEG-4 and DVB-S2, with the objective being that it makes the transition easy because they can keep the same satellite plan. If they are running 4-megabit slots in the same slot they used to do MPEG-2 DVB-S SD, they can now do MPEG-4 HD DVB-S2 with fairly comparable quality. This allows them to upgrade amplifiers, dishes, etc. It’s the same truck. This is an early-stage growth area, but clearly everyone is thinking about it and talking about it with the next year being a big growth year."
Another place seeing a lot of interest is multi-camera backhaul, which allows smaller organizations to provide more complete coverage of sports and other live special events, says Semprebon. "Today, for a football game, you might do MPEG-2 DVB-S right around 40 megabits in a full transponder. If you take MPEG-4, you can do 80 megabits on the same transponder, which allows you to do six or seven HD cameras or a few HD cameras and then SD cameras and backhaul the entire sporting event to the facility. You are not rolling out a big production truck, just cameras and a small truck."
Comtech helped a local news station in Denver produce coverage of the Democratic National Convention in August, and the compression technology enabled the station to use three HD cameras and three SD cameras to cover the convention floor and the station’s anchors. Rather than putting a switching facility and backhauling a single channel, Comtech’s technology allowed the station to run all six cameras robotically."
The fundamental driving force throughout the TV industry is cutting the workforce, especially replacing highly trained engineers with technology that can be operated by a single person or even by automation, says Semprebon. "Engineers are expensive, and the movement is to make trucks simpler to operate or remotely operable so you don’t need a satellite engineer, just a cameraman to drive the truck and push two or three buttons. This has great resonance within the broadcasting community, which is looking to trim cost, and the biggest cost is expensive people. People are starting to do this, but it’s a new investment in infrastructure. If you are looking at a new way of doing business you have to redo your facility, but you only have to do it once. Then the economics are there for years."
Scopus also sees a market developing in upgrading satellite newsgathering trucks and other mobile units to handle HD feeds. "Customers like to see HD, but the current bandwidth doesn’t allow it," says Garniek. "The reason is the transition from 16:9 SD, which can be seen on HD screens, and the gradual transition because of customer requirements on one hand and bandwidth constraints on the other hand. The next step is to move to H.264 compression in terms of newsgathering so you can transmit true HD back to the studios on the same kind of limited bandwidth. Growth is a combination of compression and transmission, taking the H.264 stream or MPEG-2 stream and taking advantage of the evolving technology. Instead of 8-bit at H.264, you can now do 6-bit or 5-bit, depending on how compression advances. Another element is transmission. It can be very limited on microwave transmission, but if you utilize an IP backbone you are less constrained and don’t have to wait until compression gets down to 5 megabits. You can stream at 7 over IP.
Compression technology continues to advance at a rapid pace, and this evolution is critical to keep up with the expected demand from broadcasters who will need to make sure they can meet the demands of end users. "We have made a lot of progress with AVC compression in the last five years," says Braydon. "We are still in a fairly steep part of the improvement curve. We have seen improvements of 20 percent per year in bit rate for the same quality. That will flatten out, but we are still several years away. MPEG-2 took 10 years to go flat. With AVC we are still seeing quite a lot of improvement. The main players in the business are working very competitively to outdo each other, and we’re focusing on making in compression technology and will see substantial gains for the next few years."
Hobbs sees potential gains of 10 percent or more on bit rate efficiency or video quality with MPEG-2. "There are not huge amounts of improvements to be made, but there are some if someone wants to replace their encoding infrastructure if one wants to replace tens of millions of set-top boxes in the field," she says. "MPEG-4 is very much a new technology. It has only been in commercial use for four or five years. If you think back to when MPEG-2 started, it was making rapid improvements in its first five or 10 years. So we are expecting to see the same thing in MPEG-4. We are on second-generation MPEG-4 encoders, and there was marked improvement. As we move to third, fourth or fifth generation we will continue to see substantial improvement in MPEG-4, certainly for the next five to 10 years," she says.
"I don’t think we are at the limits, but I think advancement are going to dramatically slow down," says Szelag. "Now that we are in the digital age and going from MPEG-2 to MPEG-4, I think it’s going to be a number of years before we see any great leaps. Now the big advances are going to be in the delivery mechanisms and getting the content out." Hildeman also identifies equipment as the biggest limitation to continued compression advances. "I think the limit is in the hardware on which software-based encoders such as Streambox are deployed. The better and faster the processors and chips are, the higher the quality of video and the faster the transmission," he says.
Comtech is focusing its efforts on the physical layer of the technology, says Enns, who sees a business in replacing hardware that already is becoming obsolete after three or four years. "We are approaching the best we can do in terms of the physical layer. That’s why so much attention is given to the link layer and the application layer to see what can be done in other areas to provide better experiences," he says.
"From the compression standpoint, it’s going to keep pushing and pushing and pushing the human eye to see how much of original content can we throw away until a majority of viewers around the world are going to see that we threw too much away, then they’re going to pull it back," says Szelag. "Costs are not being driven up, but as the demand comes for greater quality in the sense of HD services, bandwidth is increased there, and there always is that fine line. What is the breaking point? How much bandwidth can I use and achieve the best perceptible quality that the customer wants without giving away the farm? Everyone always wants the biggest bang for the buck and the best price. We are all consumers at heart."