[Satellite News 07-26-11] Israeli satellite bandwidth optimization company Novelsat hopes its NS3T satellite transmission modulation technology will be considered a breakthrough improvement on the current second-generation DVB-S2 standard.

   The NS3T solution for modems and modulators, which was made public July 20, is based on signal processing techniques that apply a mix of FPGA/ASIC algorithms that normally cancel each other out. When combined, however, Novelsat says these algorithms provide the potential to supply unlimited computational power.

   Industry resistance to the idea that marginal improvement upon DVB-S2 was possible has long been based on fixed amounts of spectrum and physical limits on the amount of data that could be transmitted. NovelSat Co-Chairman David Furstenburg has led efforts to prove that NS3T can overcome these challenges — not only in the eyes of skeptical industry engineers, but to the Digital Video Broadcasting (DVB) organization itself.

   Satellite News spoke with Furstenburg about how Novelsat’s development efforts could take video delivery bandwidth efficiency to the next level.

 

 

Furstenburg: We started four years ago in a quest to improve the spectral efficiency of DVB-S2 and become the 3G of the satellite industry. Most industry experts and engineers have thought that it was not possible to have a new modulation for satellite transmission that was higher in efficiency than DVB-S2. The project appealed to me because I have a natural rebellious quality in me to defy the world. I attended SATELLITE shows in Washington, D.C. and sat in on panel sessions where both the mass market of high-definition and 3-D and the U.S. military were placing real challenges on the availability of satellite bandwidth to support their needs. This was even more encouraging to us as a company to take on this initiative.

 

Furstenburg: Most of our research and development effort went into compression, which aims to encode more video per digital bit of data, rather than getting more bits-per-megahertz of spectrum as we have been focusing on. Novelsat’s CEO and Co-founder Itzik Wulkan assembled a team of highly skilled engineers and we challenged them to come up with a solution that would not be trivial. At the same time, we also put together about six different teams of mathematicians and gave them a separate challenge to conquer the problem from a different direction. He gave both sets of teams a general directive to work under the assumption that there were no computational limitations.

   These teams studied the DVB-S2 standard and noticed two very important issues. One of them was that DVB-S2 is an excellent technology to use as a foundation for the basis of the work that they were doing. The second thing was that there was a general confusion in the market regarding physical limitations and computation power, which was creating constrains for maximum performance. Since DVB-S2 had been written in the early 2000s, the computational power at the time was much weaker than it is today. The available computational power multiplies every 18 months. We understood that if we increased the computational power and we implemented a complex mathematics scheme into the equation, we could solve the constraint challenges and improve efficiency.

   At the end of a two-year period, we elected most of the resulting technologies and solutions from our teams and we connected them as a bouquet, with each one of them denigrating the performance of the other solutions to some degree. When we assembled them together and fine-tuned them, we were slowly discovering an ability to get the needed improvement.

 

 

Furstenburg: We went into this process thinking we would build and sell a core system, but we were approached by the industry to build an engine around this core. It is like building a car around its engine. When we started to drive this car, we discovered that the demanding performance of our core NS3T required us to build a mechanism to support it. The bottom line is that we are providing performance that will always be better or equal to DVB-S2 by at least .4 decibels thanks to its robust mechanisms.

 

 

Furstenburg: NS3T behaves exactly as DVB-S2 would behave without any type of negative impact, except that you need more computational power to run our technology. The performance result is that users will get as clean of a signal as they would get with DVB-S2. Based on tests that we have done, we can improve or boost the capacity of a given satellite segment anywhere from 28 percent in typical working environments for broadcasters on a regular transponder, to between 62 percent and 78 percent for customers using wider 56 megahertz or 72 megahertz transponders over the best implementation of DVB-S2. This varies depending on the specific situation of the specific transponder or satellite.

 

 

Furstenburg: We are currently proceeding through 32 different trials over live satellite links with various companies on different sizes of transponders. The equipment has now been tested on the Amos 3, Asiasat 5, Intelsat 1805 and Eutelsat W3A satellites. Our focus during testing has been on reducing roll-off. By exploiting the greater processing power available now compared with the time more than a decade ago when DVB-S2 was developed, we have 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.

 

 

Furstenburg: We didn’t go directly to the DVB organization when we started. We started by talking to experts in the market. There was an intense skepticism and refusal to our ideas that was borderline religion, be we understood early on that nobody would believe us and that the only way this could happen was if we were able to prove NS3T’s capability and remove all doubts by showing it work. We knew our company couldn’t just provide algorithms for standardization, so our plan was to produce a modulator and a demodulator in the modem and to eventually transfer this technology into an ASIC version, which will happen early next year.

   Engineers are now coming to see what we have. We also feel that now is the right time for us to go to the DVB organization and try to have it standardized. To do this, we registered as members of the DVB organization and are working to become committee members. We were told that we would be invited for the next meeting of the committee as an observer and after that, I believe we will become a legitimate member. I hope this will drive NS3T into standardization.