Philip Hunter /
Originally featured on BroadcastEngineering.com
Company proposes third digital satellite standard
Israeli satellite technology company Novelsat is lobbying the Digital Video Broadcasting (DVB) group for the creation of a successor to the current second-generation DVB-S2 standard based on its own technology.
The company argues that with both uplink and downlink capacity coming under increasing strain, especially in Western Europe, through a combination of higher definition services and constraints over launch and positioning of new satellites, there is a pressing need for a new technology. However, until recently, at least this has met resistance from the industry not least because it was thought that anything more than a marginal improvement upon DVB-S2 was impossible. This was based on the observation that there was a fixed amount of spectrum and underlying noise that imposed a physical limit on the amount of data that could be transmitted. Indeed Shannon's Law, otherwise known as the Shannon-Hartley theorem, defines an upper limit on the amount of digital data that can be transmitted within a given spectral band in the presence of a specified amount of noise interference. But the founders of Novelsat realised that this limit was based on the assumption that the available spectrum was utilised as efficiently as possible under DSB-S2. They decided this was not the case and set out to prove it by first identifying the underlying mathematical techniques, then developing the algorithms, and finally incorporating these into modulators and modems. The equipment has now been tested on the AMOS3, AsiaSAT5, Intelsat 1805 and Eutelsat W3A satellites, and on average improves the bit rate achievable by 28 percent compared with DVB-S2 within the 7-9dB range normally used by broadacsters, according to Novelsat's co-chairman David Furstenburg.
Novelsat claims to have little competition at present largely because of the assumption that Shannon's Law implied that little improvement in spectral efficiency was possible beyond DVB-S2. For that reason, most of the R&D effort went instead into compression, which aims to encode more video per digital bit of data, rather than on getting more bits per megahertz of spectrum as Novelsat has been focusing on. It now seems that there is room for improvement on both sides, aiming to maximise the amount of video that can be delivered per megahertz of bandwidth. Indeed, Novelsat claims it is keeping some of its performance improvement in reserve for competitive reasons, just in case some other company comes up with a rival technology over the next few years.
While various techniques have contributed to Novelsat's modulation scheme, a major focus has been on reducing roll-off, which is the guard band at each end of a channel to protect interference. In DVB-S2, this guard band is 20 percent of the channel spectral width, which itself was an improvement over the 35 percent of the first-generation DVB-S. But Novelsat says that by exploiting the greater processing power available now compared with the time over a decade ago when DVB-S2 was developed, it has been able to reduce the roll off band to just 5 percent at each end of a channel. This accounts for much of the throughput improvement.
Indeed, the key lies in exploiting increased computational power, and for this reason Novelsat anticipates further improvements over the next decade and beyond. There may even be a DVB-S4 to follow.