Ground Segment: Transformational Antennas I – End of the Parabolic Paradigm?

Thursday, 30 July, saw the continuation of the GVF Webinar Series – the sixth event, produced in association with the Satellite Evolution Group, discussing Ground Segment: Transformational Antennas I – End of the Parabolic Paradigm?. With registrations from 79 countries, and a constant flow of “live” questions from attendees, this was the busiest yet in the webinar series, supported by KenCast.

Considering the central question “Will the antenna transformation mean the end of the parabolic as the very icon of satellite communications?”, the End of the Parabolic Paradigm? – the first in a two-part examination of ‘Transformational Antennas’ – focused on the long awaited and keenly anticipated market arrival of new antenna technologies.

Recognising the the critical need for new antenna and terminal technologies such as FPAs to fully realise the potential of the combined parallels of LEO constellation technology roll-out, and increasing demand for satellite services to mobile platforms (such as aircraft, ships, and land vehicles), it was acknowledged that the parabolic antenna will not disappear from the industry or from our skylines. Parabolics will still feature in hubs/gateways and, as was commented by one panellist, fixed satellite/GEO-based services will not really require any of the new technologies to continue, based on the excellent performance of qualified and type approved parabolics.

Starting with the questions “Why all the buzz about FPAs?” and “What advantages do they have over parabolic antennas?”, the four panellists – Keith Edenfield, CTO, AvL Technologies; Bill Marks, Chief Strategy Officer, Kymeta; Alvaro Sanchez, CEO, Integrasys; and, Esat Sibay, COO & CFO, Alcan Systems – detailed their perspectives in relation to the various technology offerings either already available to the market, or near to development completion.

The dialogue, facilitated by Jeremy Rose of COMSYS as moderator, went on to explore the end-user applications for which FPAs are an imperative ground segment innovation because they will do what parabolics cannot; issues surrounding how the different flat panel technologies actually work, the nature of beam forming, and associated issues concerning prevention and mitigation of various forms of satellite interference; differences in link budgeting requirements between parabolics and FPAs and in the software tools required to address these differences; the nature of the metamaterials used in FPA manufacture; issues concerning power consumption; and, manufacturing scale and consequent unit costs. In a particularly interesting point, one panellist focused on distinguishing between just the antenna and the entire terminal, noting that scales achievable in antenna manufacture are not necessarily equalled by scale in, for example, manufacture of BUCs, and the real key to the success of FPA technologies will be in bringing the rest of the terminal ecosystem up to the same capacities.

You can listen again to the insights of these key industry innovators, or if you missed the webinar live you can still join in, by clicking on the button below. Here you can also access video recordings on the entire GVF Webinar Series.

You can register NOW for the second of these two Ground Segment webinars, taking place on 13 August, Transformational Antennas II – Will terminals realise the promised LEO Connectivity Revolution? to see the dialogue focus shift to determining if there is there a single FPA solution that is a best fit for all current, and potential future, markets, orbital technologies, and frequencies. Questions to be explored are what solutions are in development, which are near to market, and where does the newest, cutting edge, innovation take the industry and its ever growing numbers of users?

Q & A continued….

The following questions were posed during the webinar but there was insufficient time to respond during the one-hour duration of the event. Thank you to our audience for taking an active part by asking questions, and to our panellists for their time to answer them after the webinar ended…

1. Can the panel please comment on off-axis power constraints. i.e. with a wider beam how can you mitigate interference with adjacent satellites?

Keith Edenfield: The wider beam intersecting with the start angle of the applicable regulatory off-axis EIRP spectral density limits as well as the higher Co-pol and Cross-pol levels will have to meet those limits. Compliance is determined by analysis of measured antenna patterns over the range of electronic scan angles. The result is approval is granted based on maximum allowed input power spectral density to the antenna which leads to determining the modem spread factor and corresponding satellite transponder occupied bandwidth.

Bill Marks: Kymeta developed proprietary uplink power control algorithms that adaptively controls the BUC power based on scan and skew angle to the satellite to ensure that we comply with off-axis PSD limits throughout the entire scan volume of the antenna.

Alvaro Sanchez: There are solutions in the market to mitigate interference such as Adjacent Satellite, in a proactive way. Satmotion Pocket is one of them, Satmotion API a more automated and active way, and Alusat a remote way; they are all very effective.

2. Does Kymeta see their antennas as a ground segment competitor? Are the limitations currently faced by flat panel antennas something the team could explain will be overcome?

Bill Marks: The Kymeta terminal was designed from the ground up to be a satcom terminal for mobile terrestrial platforms. Existing limitations of ground segments for vehicles and vessels include poor reliability due to moving parts, high power consumption beyond what car batteries can provide, poor performance at extreme environmental conditions and high cost. The Kymeta u8 terminal addresses all these limitations. With no moving parts, our FPA is highly reliable. It consumes approximately 160W of steady state power which is easily supplied by cars. It handles extreme hot and cold conditions well and is the most affordable solution on the market. In addition, we offer the u8 with both purchase and financing options so that the technology is accessible to a wide range of customers.

3. Are the panellists saying that parabolics are not applicable for maritime or NGSO tracking, just because of maintenance?

Keith Edenfield: No

Bill Marks: There are significant advantages in terms of reliability and maintenance to having no moving parts. Mechanical systems with moving parts generally equate to lower reliability and field life. Parabolics have clearly become adopted in maritime, but the inertia of a parabolic terminal means that a single dish can’t slew fast enough to break lock with the descending NGSO satellite and establish lock with the ascending NGSO satellite while meeting latency requirements. So, as an NGSO ground station, parabolic technology necessitates the use of 2 antennas. This adds expense, power, weight, and complexity to the ground terminal solution. Not to mention the added risk of mechanical failure.

Alvaro Sanchez: Parabolic is the standard for maritime networks and it will remain very relevant in my opinion for very long time. There will be some applications where flat will be competing, however, parabolic will remain with the majority of the market. On the other hand, in NGSO scenarios, flat panel antennas will be more relevant as they have more advantages and less disadvantages than for GEOs.

4. The flat panel antennas seem way much smaller in size compared to the smallest parabolic antenna – aren’t these flat panel antennas more affected by weather changes?

Bill Marks: Flat panel antennas are affected by weather, but much differently than parabolics. FPAs can be significantly impacted by high temperatures and to a lesser degree by rain and snow accumulation on the flat surface (typically the radome) itself. Because most flat panel antennas are housed in an IP rated enclosure of sorts the antenna can get quite hot when considering high ambient temperatures compounded by solar load. FPAs that require a high-power draw run hot are far less tolerant of these conditions and often simply shut down. The Kymeta u8 requires low input power and hence does not overheat. We included reflecting coating on our radome to reject the solar load and we also incorporated a self-correcting system for high temperatures experienced by the antenna core. We have run our antennas in very hot desert conditions in the middle of the summer with only little performance degradation which is accounted for by our link budgets. The second issue is rain and snow accumulation on the flat surface of the antenna. We have introduced a very slight doming to our radome and combined with a hydrophobic coating, the Kymeta u8 avoids this weather condition challenge.

Alvaro Sanchez: Indeed, the gain lower, so the weather changes affects more, therefore an accurate link budget and calibration of the antenna is even more important than with parabolic antennas.

5. Can we get flat antennas cheap? Let’s say lower than 500 euros.

Bill Marks: Flat panel antennas can indeed be inexpensive when they reach volumes of consumer electronic products. Unfortunately, we’re not quite there yet. This is why we introduced the Kymeta Connect plan which allows customers to finance their u8 terminal and bundle it with their data service. Our bundles for the terminal, 1GB/month data plan and service and support start at $999 per month.

Esat Sibay: ALCAN has announced a price of EUR 1,500 for its enterprise antenna due out next year. A consumer antenna in the future below EUR 1000 is not inconceivable.

6. Another challenge for flat antennas other than mobility is adapting to classification for use in hazardous area especially in oil & gas platforms. Can the panel comment?

Bill Marks: This is indeed a challenge for phased array antennas and other technologies that consume significant power and consequently generate heat. Kymeta’s technology consumes significantly less power and operates at low voltage so the risk of electrical arcs/explosions in hazardous oil and gas environments is much lower.

7. I guess flat antennas have less gain than parabolics. For internet services, for example, will remote stations need bigger BUCs?

Keith Edenfield: All FPAs are not the same, however, in general the answer is yes when comparing equal physical aperture sizes.

Bill Marks: Flat panel antennas behave differently from dishes as they suffer from scan loss. Therefore, it is not meaningful to compare the diameter and of a dish and a diameter of a flat panel antenna. In each case, a link budget needs to be developed for the desired use case condition and the BUC size is determined based on the desired transmit throughput.

Alvaro Sanchez: Bigger BUC means more weight, more heat, more power, more interference and the forward link still have lower gain, so it is not the ideal solution. In our view the solution is to have an accurate calibrated BUC at all times, which it maximizes the antenna performance at all times; not losing a bit.

8. How can we handle areas with a low up/down angles to the satellite? At the moment additional tilt is necessary in these areas.

Keith Edenfield: To some extent conformal arrays can help but for full hemispherical coverage multiple FPAs will be required on the vehicle

Bill Marks: That is true for satellites in GEO orbit. In higher latitudes the look angle at the satellite is challenging and tilting may be required. However, as NGSO constellations become available this problem goes away. This is yet another reason why flat panel antennas are uniquely positioned to leverage the capacity of NGSO networks.

Esat Sibay: Most phase array technologies operate up to scan-angles of 60deg. Beyond it, side-lobes degrade performance. Some solutions could be possible to extend this scan-angle by adding convex shape to antenna. However. these are not targeted in the earlier year.

9. Question on power consumption: What is currently the typical power consumption to operate a flat panel antenna and what do you expect the consumption will be in the future?

Keith Edenfield: As mentioned on the Webinar all FPAs are not created equal. An Active (transmit power amplifiers distributed within the FPA) Electronically Scanned Antenna (ESA) will be relatively power hungry depending upon the EIRP required – currently around 600 W for a terminal with 50 dBW EIRP. Incremental improvements may be possible in the future. If Passive (active beam steering only) ESAs achieve what has been advertised by some, then there is the potential to require less than half of that power for an equivalent EIRP.

Bill Marks: Power consumption varies considerably depending on the technology used. The nominal power consumption of the Kymeta u8 is 160 Watts under steady state conditions. This is the power consumption for the terminal which includes the antenna, modem, BUC, and other components. Peak power can be as high as 600 Watts when the ambient temperature is the coldest (-40C). When the ambient temperature is colder than 10C the terminal turns on an internal heater. Note that analog phased arrays generally consume 600 to 800 Watts steady state and digital phased arrays consume > 1 Kilowatts of power.

10. Is there a formula relating flat panel antenna size and parabolic antennas?

Keith Edenfield: Aperture size and efficiency (including aperture illumination efficiency less antenna dissipative losses) allows direct comparison between parabolic and Passive ESAs. For comparing to an Active ESA comparing EIRP and G/T are the best parameters for comparison.

Bill Marks: There is a basic directivity formula that equates the diameter, efficiency, and gain of an aperture antenna like a parabolic or flat panel. However, this formula creates a relationship between flat panel antenna performance at broadside (perpendicular to the panel) and the equivalent parabolic. As discussed before, all flat panel antennas also experience scan loss which should also be accounted for.

11. Can you discuss the effect of widening beam width at high scan angles (low elevations) on adjacent satellite interface in GEO applications?

Keith Edenfield: The peak antenna gain decreases by at least cos(θ), relative sidelobe levels increase, and beam width increases approximately 1/cos(θ). As discussed in the response to Question 1, EIRP spectral density limits must be met to control ASI and will become more onerous as scan angle increases.

Bill Marks: Kymeta developed proprietary uplink power control algorithms that adaptively control the BUC power based on scan and skew angle to the satellite to ensure that we comply with off-axis PSD limits throughout the entire scan volume of the antenna.

Alvaro Sanchez: As the antennas are smaller, the beam is wider, and the effect of adjacent satellite interference in GEO is larger, as mentioned, we have active software technologies which allow to prevent and mitigate wider ASI effect for Parabolic and Flat Panel Antennas.

12. One of the major problems for GEO satellite operators is the antenna verification part since we can’t do the same test that we do for parabolic antennas. How are manufacturers going to address this issue? Is there a plan to start a type approval process with major operators like Eutelsat?

Keith Edenfield: Compliance is determined by analysis of measured antenna patterns over the range of electronic scan angles

Bill Marks: We work with several satellite operators to establish type approval for our antennas on their networks. While our flat panels do not work the same way as parabolics (no flat panels do), we work with each operator to ensure that our terminal meets the requirements of their network. Sometimes waivers and/or exemptions may be required depending on system level constraints.

Alvaro Sanchez: There are tools that enable a rapid test or out of the box and access certification, one of them is called Satmotion Pocket. It is not a type approval as it is more effective to test a particular antenna making sure it operates correctly interference free.

13. What types of issues are there with flat panel antenna used in a satellite and the earth transmit facility is also a flat panel.

Bill Marks: There should be no issues in this scenario. Many satellite antennas that generate spot beams are already flat panel phased arrays.

14. Flat Panel Antennas – what is the frequency band of operations?

Keith Edenfield: We are working on standard Ku and commercial Ka-band SatCom frequency bands

Bill Marks: That depends on each antenna. Our commercially available system is designed to operate in the Ku-band however our technology scales to Ka-band and into Q/V bands as well.

Esat Sibay: For our first product ALCAN is working on a Ka-band antenna.

15. How does the roll and pitch on a vessel influence a flat panel antenna. Does it help to have multiple antennas on a vessel/ship?

Keith Edenfield: See Question 8.

Bill Marks: Being electronically steered, our antennas compensate for roll and pitch of a vessel. In fact, we have tested our antennas on many vessels with excellent results. Kymeta’s proprietary tracking algorithm mitigates motion in Roll, Pitch, and Yaw, and can track through motion in those axes at up to 30 deg/sec. Our experience showed that yaw is a greater concern in maritime environment due to mast blockage of the satellite view and varying look angles. To address this issue, we have developed an optional accessory that supports a multi panel configuration. This accessory automatically selects the panel with the best line of sight view of the satellite.

Alvaro Sanchez: In the maritime scenario, two antennas will be required as there are blocking objects such as masts.

Esat Sibay: ALCAN’S phased array technology can switch beams in less than 20ms. This is below the around 75ms that is required to maintain the link with a satellite beam when on a moving body. Therefore one antenna will be enough to maintain the link if no obstruction of view; however if there is a physical obstruction such as a sail in one direction, more than one antenna may be needed to ensure direct line of sight at all times. This is no different from existing antennas used in maritime context.

16. How power hungry are ESA antennas?

Keith Edenfield: See Question 9

Bill Marks: Power consumption varies considerably depending on the technology used. The nominal power consumption of the Kymeta u8 is 160 Watts under steady state conditions. This is the power consumption for the terminal which includes the antenna, modem, BUC, and other components. Peak power can be as high as 600 Watts when the ambient temperature is the coldest (-40C). When the ambient temperature is colder than 10C the terminal turns on an internal heater. Note that analog phased arrays generally consume 600 to 800 Watts steady state and digital phased arrays consume > 1 Kilowatts of power.

Esat Sibay: ALCAN’S UNIQUE LC PHASED ARRAY TECHNOLOGY is designed to be low power. For the current LEO/MEO antenna the power of the antenna alone is expected to be 35W.

17. When will LC antennas be commercially available?

Bill Marks: Today! We announced commercial availability of our first-generation product, the Kymeta u7 terminal, in March 2017. Our next-generation product, the Kymeta u8 terminal was announced in March 2020 to launch in October 2020. To date, we have deployed hundreds of units around the world on a variety of vehicles and vessels in multiple markets.

Esat Sibay: First product is planned for Q4 2021.

18. How do you compare LC flat panel antenna with Starlink’s ”UFO on a stick”?

Keith Edenfield: I have not seen a data sheet for that, but Mr. Musk has Tweeted that it has motors. I know that the User Terminal operates at Ku-band. The UT antenna may be a fixed beam FPA that is mechanically steered, or it may employ hybrid mechanical steering and electronic beam forming.

Bill Marks: Based on public information provided by SpaceX and SpaceX CEO Elon Musk, the Starlink antenna appears to be a phased array with mechanical elevation steering. It is designed for fixed applications. Kymeta’s technology is electronically steered with no moving parts and is optimized for mobile platforms in challenging environments.

Esat Sibay: Official date-sheet of Startlink’s antenna has not yet been released. So difficult to compare. However, early reports talk about a phased array antenna (semiconductor based?) that also has an actuator/motor for additional steering. Would expect both the cost and power of this type of solution to be higher than ALCAN’s LC phased array antenna.

19. Can we assume that flat panel antennas are flexible for integrating with different make BUCs/ modems and how much form factor margin is being considered for such add-ons?

Keith Edenfield: Yes, for Passive FPAs (no distributed RF amplifiers in the FPA itself). Any commercially available BUCs and LNBs may be considered.

Bill Marks: Yes. We offer the u8 technology in several configurations: antenna, ODU and terminal. The Kymeta u8 antenna allows integrators to create mobile satellite terminals with their choice of RF chain components and satellite modems. We provide integrators with a guide to ensure proper integration of our technology. An ODU is a terminal that allows for easy integration with an external satcom modem. The terminal configuration is a turnkey solution that is a fully outdoor terminal (rated IP66) with a built in RF chain and integrated satcom modem card.

Esat Sibay: Yes, antennas can work with all BUC/Modem types. Form factor will need to be considered depending on requirements. For enterprise antenna, this may not be a priority. For mobility (i.e., maritime, land-mobile, aero) this could be a priority and thus will need to ensure appropriately integrated.

20. Isn’t the terminal design different for each satellite system? How does this help with scale?

Bill Marks: In architecting the u8 we have developed the core antenna to be as generic as possible. For example, our antennas are OpenAMIP compliant so they can work with most satcom modems. They work with DVB-S2 and DVB-S2X waveforms as well as other custom waveforms. Ours is a software defined antenna that allows us to create custom terminals for each satellite system while keeping the core antenna hardware common. This allows us to leverage economies of scale.

Alvaro Sanchez: There is a big difference in the efficiency of the antenna between HTS and non-HTS “birds” in flat panel antenna networks. The terminal design is made to HTS satellites and most of them are designed to work with different “birds”. If all satellites were HTS, scaling of the Flat Panel Antennas would be much easier, and many more antennas will be required.

Esat Sibay: Core Antenna design can be the same for each frequency (e.g., Ka-band). However, will need different modem that operates with the specific constellation.

21. If the antenna (excluding modem, BUC and LNB) costs 1,500 Euro, do you think LEO constellations will bring Mbps cost so low to be competitive?

Bill Marks: That is a question for SpaceX and other LEO capacity providers. We find that the cost of the terminal is independent from the cost of capacity but the two can be bundled into affordable monthly plans.

Alvaro Sanchez: In my personal view, I believe it is a great achievement to bring down the flat panel terminal to 1500USD in 2020, when it was more than 10 times more expensive in 2019. Now the deployment of these terminals needs to be successful and the operation needs to be satisfactory with the right added value tools and user experience. If this is achieved, it would be a big success.

Esat Sibay: Market analysts like NSR are predicting that the increase in capacity that the new LEO constellations will make available, will also drive down the bandwidth costs. There are some scenarios where this is expected to be competitive with fiber for enterprise and allow low cost for consumers comparable to other connectivity options.

22. What about ASI and the huge variation of G/T over frequency and El angle as the most important link budget (LB) parameter?

Keith Edenfield: Yes, it is very important at extreme scan angles but can be mitigated if the scan volume is limited to about +/45 degrees and/or the satellite network implements dynamic resource allocation.

Bill Marks: We work with our customers to develop a LB to meet their unique needs. We provide them with the needed parameters to create their own link budgets. Regarding ASI, we developed proprietary uplink power control algorithms that adaptively control the BUC power based on scan and skew angle to the satellite to ensure that we comply with off-axis PSD limits throughout the entire scan volume of the antenna.

Alvaro Sanchez: As mentioned in the panel, Link Budget (LB) is quite different with parabolics and flats, that is why we created the Beam Budget FP module to perform an accurate link budget in any kind of antenna.

23. These new antennas use metamaterials physical properties to form the beams. Is there a heavy impact of temperature on these properties?

Bill Marks: There is no impact of temperature on the metamaterial aspect of Kymeta’s technology, per say. There is, however, a temperature impact on the liquid crystal, which is what Kymeta is currently using as a tuneable dielectric material to reconfigure the array pattern and steer the beam. Kymeta has developed several patent-pending innovations that mitigate the temperature impacts on the liquid crystal. Kymeta’s u8 product will perform under environmental temperatures of 55 degrees Celsius under full solar load.

24. Is the Alcan price of 1500 Euro support Ku-band or Ka-band and it is only for fixed or mobility?

Esat Sibay: The first-generation product will be for Ka-band and will be a fixed enterprise antenna. This is due out in 2021. The second-generation product will be a mobility antenna and will target the maritime market. Expected time for this product is 2022/23. A Ku-band product is also under consideration, but no date has been fixed yet.

25. How difficult is multi-beam support through a single panel enabling seamless handovers between satellites and how far away is the industry from a commercial multi-beam panel?

Keith Edenfield: Our plan is to offer single beam AESA based terminals in late 2021 and we have an approach defined and plan follow with Dual beam variants in 2022

Bill Marks: There are multiple different ways to achieve seamless handovers between satellites. Performing inter-satellite handovers without packet loss (i.e., in a seamless manner) can be done with fast switching or with multiple beams. Our technology is already capable of fast switching because we don’t need to actually steer the beam: we can form a new beam pointing at the right location in a speed that supports a seamless handover. Our technology can also be adapted to form multiple simultaneous beams.

26. For land mobility, isn’t the beam agility of passive antennas a problem?

Keith Edenfield: It is for fixed beam FPAs that have no electronic scanning capability.

Bill Marks: Kymeta’s technology, while considered passive from the RF perspective, can perform beam tracking at up to 30 degrees per sec while meeting regulatory pointing requirements. This has proven more than sufficient in the hundreds of land-mobile field deployments Kymeta has conducted over the past 3 years.

Esat Sibay: No, ALCAN’S LC Phased Array antenna has a beam switching time of less than 20ms, which is more the sufficient to track the satellite signals without interruption while driving. See also answer to 15 above.

27. What views do you have about the Starlink terminal – “UFO on a stick”?

Bill Marks: Based on public information provided by SpaceX and SpaceX CEO Elon Musk, the Starlink antenna appears to be a phased array with mechanical elevation steering. It is designed for fixed applications. Kymeta’s technology is electronically steered with no moving parts and is optimized for mobile platforms in challenging environments.

Esat Sibay: Please see answer to 18.

28. Where are flat panel antennas used now?

Bill Marks: Flat panel use case depends heavily on the technology utilized. For phased arrays that consume significant power, Aero tends to be the focus because those are the only platforms that can supply the power required and support the large footprint and mass. For Kymeta, we see significant adoption across land-mobile sectors with SWaP-constrained vehicles, because of our low power consumption, tighter footprint due to integrated Rx and Tx, and no need for thermal dissipation/heatsinking. This includes VIP vehicles, buses, utility trucks, and small military and first responder vehicles. We have also deployed on super yachts, smaller maritime vessels, trains, and some fixed sites as well.