Home >> May 2013 Edition >> Broadband-On-The-Move Satellites Take The Pole Position
Broadband-On-The-Move Satellites Take The Pole Position
by Dr. Rowan Gilmore Managing Director, EM Solutions Pty Ltd.

Defence forces, first responders, and emergency services and disaster recovery personnel all require reliable and trusted communications to operate in theaters in which infrastructure must often be transported. Backhaul of traffic from multiple devices—phones, cameras, computers, and equipment control systems—to a remote headquarters has, perhaps, become simpler and more integrated with IP-enabled devices and localized Wi-Fi.

Grapg1 However, the migration to IP has also accelerated the need to satisfy two other challenging core requirements that can still remain unmet—broadband backhaul capacity, and providing backhaul from a local management and command node that is itself on the move.

Although VHF and UHF radios have served this purpose for many years, such radios are, inherently, bandwidth limited. Microwave radios offer greater bandwidth, but they require line-of-sight to the horizon for backhaul. They also require stabilized antennas to maintain pointing back to base while the node is moving. Satellite, therefore, offers a solution that provides both bandwidth and the ability to stabilize an antenna that is pointed skywards, rather than towards the horizon, reducing obstruction while in motion.

In this article, we describe how EM Solutions (EMS) has developed and delivered a ground breaking on-the-move (OTM) terminal system for a government customer to assist in its future disaster recovery efforts. These terminals were fitted to mobile communications vehicles intended for rapid despatch to impacted areas of the country, and to provide fiber-like data speeds in off-road conditions while the vehicles are still in motion.

Using as a baseline product its existing Ka-band OTM terminal developed for the military and WGS satellites, EM Solutions added support for commercial Ka-band frequencies and provided an additional rotational axis to accommodate linearly polarized carrier signals. 

The new terminals were able to maintain their pointing accuracy by using closed-loop tracking data from monopulse tracking information derived from the satellite’s low power telemetry signal, supplemented with readings from the system’s built in mechanical gyroscope.

An integrated 40W Ka-band linearized block upconverter (BUC) provided sufficient uplink power for 155Mb/s data throughput from a 650mm reflector.

Newtec_ad_MSM0513 Within six months of receiving its contract, EM Solutions was able to successfully deliver and demonstrate terminals capable of meeting < 0.2 degree pointing error for off-road and marine conditions, providing exceptional performance and link availability limited only by line-of-sight to the satellite.

OTM Terminal System Specification
In the design of any satellite ground terminal, the link budget must be carefully calculated to determine the required gain (G/T) of its receiver and the effective radiated power (EIRP) of its transmitter. These two parameters, and the equivalent system parameters of the satellite itself, ultimately determine the mechanical and electrical characteristics of the terminal that can be developed.

Typically, the customer specifies the maximum possible terminal footprint that can be accommodated on their vehicle—this, in turn, determines the available antenna gain for a given choice of antenna. Next, with the available channel bandwidth and desired uplink data rate given as important system constraints, the link budget can be used to calculate the required terminal EIRP, and, in turn, the linear transmitter power required.

In the EMS system, to minimize the size of the radome and vehicle footprint, while still meeting the EIRP and spurious interference specification, a parabolic antenna of 650mm in diameter was selected. This resulted in an antenna gain of approximately 42.5dB (including the feed and radome losses) at 28GHz, the transmit frequency.

Working backwards from the link budget, the determination was made that a 40W (Psat) BUC was required to achieve the desired 155Mbps data rate. This is an exceptionally high uplink data rate to support from an OTM terminal, made possible only by the combination of high antenna gain and high transmit power together yielding an EIRP (saturated) of 58.5 dBW.

Of course, these are just the first system constraints. For any OTM terminal, there will be other constraints in addition to the data rate that must be met. These include achieving the correct polarization of the signal and pointing the terminal accurately to minimize interference with adjacent satellites.

In addition, the terminal must clearly survive as well as also be operable when undergoing severe vibration. This terminal was also required to support standard bent pipe and Asynchronous Transfer Mode (ATM) operation through the satellite, requiring band selection in the BUC and LNB to accommodate this.

RowanFig1 At first glance, a phased-array antenna might appear the ideal antenna candidate, as electronic steering is conceptually appealing to partially eliminate mechanical motion. Phased- array solutions can use a combination of mechanical steering for azimuth and electronic steering over a limited range of elevation angles.

However, there are a number of critical problems with a phased-array approach. First, phased-array antennas with a reasonable number of elements have low gain, particularly when pointed off-axis. High data rates can, therefore, be achieved only by operating at much higher power levels than with other antenna solutions.

Secondly, they generate significant energy outside the main lobe and often require waivers to achieve certification status, which need to limit the allowable unwanted radiation that is transmitted via sidelobes to satellites in adjacent slots.

Third, beam squint (i.e., the need to maintain the same pointing angle across the entire frequency band), limits the fractional bandwidth to 5 percent, even for scanned angles as small as 25 degrees.

Finally, at Ka-band, the transmit (28GHz) and receive frequencies (18GHz) are a half-decade apart in frequency, so separate arrays would be needed to cover each frequency—arrays that would need to be synchronized with each other to individually steer each to within a fraction of a degree.

It is the view of EM Solutions that any broadband phased-array antenna would, therefore, be burdened with unacceptable compromises to achieve a practically useful and operational system at Ka-band—in spite of the obvious advantage of lower height profile.

RowanFig2 The use of a single parabolic reflector antenna overcomes these limitations. A parabolic reflector and horn with combined waveguide feed for transmit and receive signals will achieve much higher gain, coverage of the entire bandwidth, automatic alignment of receive and transmit beams, and tight control of sidelobes. The disadvantages: The need to mechanically steer the reflector and the difficulty of designing a dual-band feed have been overcome through robust and precision systems engineering of the terminal.

OTM Terminal System Design
A core design consideration is how to acquire, and then track, the satellite. There are two broad approaches. The first, which is an open-loop approach, is to use the known position of the geostationary satellite in the sky to orient the antenna and to reorient the terminal given the terminal’s current heading.

However, pointing accuracy to within a fraction of a degree is quite difficult (and expensive) to achieve with an open-loop tracking system that relies solely on inertial measurement systems to steer the antenna. Mechanical tolerances, that can also vary with temperature, are also difficult to compensate for with this approach.

Furthermore, inertial measurement systems that rely on GPS measurements cannot account for Ka-band signal refraction through the radome and are, therefore, susceptible to radome variations and large offset errors that also depend on the angle of elevation.

GPS-assisted inertial navigation units (INUs) are also quite sensitive to multipath, so moving in the vicinity of buildings or trees could upset the INU, which would then result in pointing errors.  To make matters worse, it would be hard to even know that these pointing errors had occurred.

The second approach, based on closed-loop tracking, avoids all of these problems. With this approach, the satellite is tracked using its own transmissions. The terminal, for instance, could seek the orientation that maximizes the receiver signal, or satellite beacon signal, or some other derived signal, as in a “monopulse” system.

The signal beamwidth is rather broad at its 3dB points. Finding the center of its maximum using a traditional mechanically scanned approach involves pointing a conventional reflector antenna intentionally off center, to the beam edges where the signal strength begins to drop off rapidly.

RowanFig3 Two examples of mechanical scanning are conical scan and step-track. These are deployed in so-called on-the-hop terminals, which remain stationary in operation once deployed on location. However, a deliberate pointing-error must be introduced to verify the maximum. This reduces the gain and effective power. It also responds too slowly for rapid vehicle motion or acceleration.

For its terminal, EM Solutions uses a system known as monopulse tracking, which provides the most certainty and accuracy to the true direction. Monopulse systems are able to estimate the pointing-error without any mechanical scanning and without needing to deliberately mis-point the antenna.

Monopulse antennas generally use a dual feed: The main feed has a normal broad antenna pattern, while the secondary feed internally generates a pattern with a sharp notch along the bore-sight. By comparing the signals from the two feeds, the antenna can be precisely pointed without ever needing to deviate from the maximum signal strength.

This has two advantages: The link budget remains strong and availability is always maximized, as the pointing error is kept small; and power consumption is kept low, as the antenna is held stationary rather than intentionally swept in a constant scanning motion.

Other design considerations also had to be accounted for with a mechanically rotated system. These included:

• Friction, which causes the tracking mount to lose its pointing angle during vehicle motion, so the motors must apply torque to overcome the friction. Too much friction within the motor and bearings will result in the motors having to use more power to overcome the frictional force that will tend to shift the antenna the same way as the underlying motion of the vehicle. This can be overcome by using high quality bearings and design for a low inertial mass, balanced antenna system.

• Balance, which is a critical factor in tracking mount design. A balanced system, where the axis pass through the centre of mass, will have lower power consumption and improved system performance. In an unbalanced mount, linear acceleration of the vehicle will translate into rotational motion about the axes, forcing the motors to consume power just to maintain the original pointing angle. A well balanced design can avoid this effect.

• The “keyhole” effect, which occurs when the mount is required to track a satellite at elevation angles approaching 90 degrees from its base (i.e. looking straight up, as at the equator on a ship). In systems using a two-axis drive, to reduce overall vertical height, the antenna cannot directly pass through overhead as the elevation changes. This produces a blind region until the antenna can completely rotate around its azimuth axis and reacquire the satellite.

RowanFig4 EM Solutions’ design approach to overcome these issues permits the antenna to point directly overhead, by using a three-axis system, with a cross-elevation axis, to allow rotation throughout the entire sky. The additional degree of freedom in elevation allows direct overhead pointing with the base able to rotate to an optimal position, overcoming the keyhole problem. Adding the second elevation axis does slightly increase the cost and height of the terminal as more mechanical and control system design is required, however it will ultimately result in power savings and less wear on the azimuth axis. It vastly reduces the pointing error as well, because the cross-elevation axis allows the antenna to stay still, since it is only friction and residual imbalance that causes transfer of vehicle motion through to the antenna.  

The challenges do not end there. As the platform is itself travelling, the satellite beacon, used to estimate the pointing error, suffers Doppler shift. Uncertainty in the beacon frequency is consequently quite large. This is due to drift in the satellite’s own local oscillator as well as the Doppler shifts caused by vehicle motion. The frequency offset can be several hundred kHz, and the Doppler shift can change at a few kHz per second as the vehicle maneuvers. These frequency offsets are cancelled internally by the system controller.

Compared with the Wideband Global SATCOM (WGS) military satellites, the Ka-band signal in the satellite used for this particular disaster recovery application was linearly polarized, rather than circular. This is uncommon for Ka-band satellites and required a fourth rotational axis, that of the antenna feed itself, to be incorporated into the system design.

The function of this fourth axis is to rotate the antenna waveguide feed to match its polarization with that of the incoming signal. Of course, this varied as the vehicle twisted and turned, as well. An additional control loop was designed and fitted to monitor the polarization of the received signal, and this was used to drive a fourth motor fitted within the antenna and dedicated to rotating the waveguide feed about its axis.

RowanFig5 Outcomes—Broadband-On-The-Move
The decision to base the system around a parabolic reflector antenna rather than phased array was justified by measurements on its emitted power spectral density (ESD) made at EM Solutions’ outdoor antenna test range. These measurements show the power contained in any sidelobes generated by the antenna and must be maintained within tight limits set by certification authorities, such as ARSTRAT (for the WGS system), or in the present case, the ITU.

The results with and without the radome were essentially the same in terms of pointing and shape and indicated that the radome’s main effect was to attenuate the signal. Figure 4 on the following page shows the predicted and measured values for ESD, as well as the ITU and WGS specification limits, at an azimuthal (Φ) range varying about the central beam by +/- 30deg.

The ESD is a maximum when the terminal is operated at the maximum linear power and the minimum symbol rate, and is shown for the worst case condition. All measured and predicted ESD(&#966;) plots are referenced with respect to -12dBW/Hz at bore sight, corresponding to a symbol rate of 6Mbaud and linear power of 44dBm. The figure shows how closely the measured data follows the predicted results, subject to limitations on the step size used for Φ (+/-2 deg). The ESD falls below the ITU sidelobe mask as required.

Careful attention to system design as described here can result in many benefits that the end-user in the military, or first response situation, will almost certainly appreciate. These include:

• Maximum system availability—visibility of the satellite is an essential requirement that may not always be under the user’s control. However, the system designer can still optimize availability by minimizing signal loss due to mispointing, and by rapidly re-acquiring the signal after tracking has been lost and visibility regained. The former was achieved with best-in-class pointing accuracy, a result of using closed-loop beacon signal processing and tracking. Even in motion off-road, this system maintains pointing loss less than 0.2 degree, well within the antenna beamwidth, which in turn preserves the link budget to within 0.1dB of bore sight, and improves performance on marginal links.

Quickest re-acquire time—this is required after the satellite is obstructed from view and becomes visible again, to maximise system availability. We achieved it by incorporating an innovative gyro-lock mode in the control system that predicts satellite direction during the signal loss, and readies the unit for immediate operation after the satellite reappears into view.

Continuous coverage over all ranges of motion. The user does not want to be restricted from communications by the pitch or gradient of the road over which they are moving. EM Solutions uses a three-axis gimbal mount system to eliminate keyhole effect and annoying synch losses when the satellite is close to overhead, and where other systems need to rotate violently to maintain lock.

RowanFig6 • Reduced maintenance and power consumption. The power budget of systems used in many vehicles must be restricted to preserve battery life, while most stabilised platforms used in backhaul communications on the move are power hungry. The terminal described here uses high life, sealed brushless motors, and balances the BUC and waveguide feed behind the antenna. Its high moment of inertia is designed to be balanced to minimise internal movement of the antenna, which should remain stable and pointing to the satellite. This reduces both peak and average power consumption to just a few watts greater than the power consumed by the BUC itself. Mounting all the RF subsystems behind the antenna also avoids the use of an RF rotary joint and a failure-prone component.

• Cost effective. Users are frequently surprised that while quoted terminal costs can appear reasonable, the “add-on” costs of required accessory INU and GPS systems can prove very expensive. EM Solutions OTM terminals rely on closed-loop tracking of the satellite itself, and GPS and INU data are only secondary data points, used for initial location of the satellite window. As a result, much lower cost INU modules can be used —and these are embedded in the terminal itself.

• Complete end-end design, manufacture, and support. As described above in the section on system specification, the two critical design variables in an OTM system are the antenna size and BUC power. EM Solutions has control over both of these, since it manufactures and designs its own RF and control electronics. With full control over the system supply chain, it can customize its terminals and provide rapid turnaround.

• Configurable to end-user needs. EM Solutions terminals are able to track both circularly and linearly-polarized signals, so extension to X-band and Ku-band satellites is straightforward. Such terminals are now being designed. With a variety of terminal sizes, powers, and frequency bands, the customer can optimize their configuration to suit local conditions.

Maximum Flexibility
The broadband on-the-move system designed in this article was certified and licensed by the customer’s national telecommunications authority for use on the satellite in March 2013, less than nine months after EM Solutions received its contract to develop the system. It was successfully demonstrated to the government customer the same month, ready for deployment. Several terminals have already been delivered as part of an ongoing rollout program.

Broadband on-the-move satellite communications terminals now offer true mobility and high data throughputs to military users and first responders, even under the most demanding and severe on-the-move conditions.

RowanFig7 In considering the suitability of a terminal for a given application, the user must consider how the requirements below can be solved technically :

• Linear and angular acceleration for which the platform must compensate

• Mobile elevation and azimuthal range

• Acquisition time and recovery time following blockage

• Regulatory constraints on receiver tracking and transmit pointing-error, sidelobes, and appropriate allowances in the link budget

• DC power constraints

This article has described the design constraints and features of EM Solutions’ most recent OTM terminal that provides backhaul of any IP-based traffic, a complete integrated satellite communications solution that includes tracking antenna, satellite receiver and transmitter, closed-loop beacon tracking system, and a terminal management system. It also includes its own integrated GPS compass and INU (inertial navigation unit), unlike other systems where these can be expensive add-ons.

The system described uses a simple, single parabolic reflector and horn with combined waveguide feed for both transmit and receive signals to achieve higher gain, coverage of the entire bandwidth, automatic alignment of receive and transmit beams, and tighter control of sidelobes than any other equivalent antenna or phased-array system.

Gimbal-mounted with three axes to eliminate keyhole effect, and driven by a closed-loop tracking control system, the system achieves unsurpassed pointing accuracy and reacquisition time once locked, providing superb communications availability at previously unheard of data rates up to 155Mbps. Even while traveling at 120km/hour on the highway or 80km/hour off-road, connectivity is assured whenever the satellite is in line of sight.

In the future, X- and Ku-band capability will become available, based on swapping a custom-engineered, modular “kit” comprising feed, SSPA/BUC and LNB’s with the Ka-system on the OTM platform. This delivers maximum flexibility in the field for either military or commercial customers across three important satellite bands using a single base terminal.

GilmoreHead About the author
Dr. Rowan Gilmore is currently Managing Director of EM Solutions Pty Ltd. The company is recognised around the world for manufacturing technologically superior microwave modules and systems for next generation broadband communications. It offers differentiated products that embed its unique IP, and are available on demand. He was previously a Vice-President with SITA, the global IT and telecommunications service provider to more than 600 of the world’s airlines. When based in London, he was responsible for network installation, operations, and service delivery in more than forty countries across Europe. Prior to that, he was based in Atlanta and responsible for the SITA high-speed backbone network in the U.S., Canada, and Mexico, and for the implementation, design, and build of new customer networks. Before joining SITA, he was Manager, Advanced Global Networks, R&D, Telstra Corp, Sydney, which serviced customer telecommunications requirements outside Australia, and delivered offshore telecommunications services to Australian customers, via both fiber and satellite.