High-tech standout Broadcom has sold over one billion GNSS chipsets worldwide, leveraging all major global satellite navigation constellations and the full range of GNSS features, including Galileo’s dual-frequency and innovative BOC modulation capabilities.

Broadcom is doing things that no other company in the world is doing. Positioned right at the leading edge of GNSS-for smartphone technologies, the company boasts an extensive portfolio of standalone GNSS receiver chips and combination GNSS receiver and sensor-hub, or location-hub, chips.

Already an industry leader in multiconstellation GNSS technologies, present in the latest, top-of-the-line handsets, Broadcom positively threw down the gauntlet last year with the inauguration of its new BCM47755 chipset, the first dual-frequency GNSS chipset for smartphones. The unveiling, which came in the fall of 2017, led European GNSS Agency (GSA) Director Carlo des Dorides to suggest we might see a fully functional dual-frequency smartphone as early as summer 2018.

Obviously Multi-Constellation

Broadcom has always been a firm believer in the cumulative value of each additional GNSS constellation. Speaking from his office in Madrid, Broadcom’s Associate Director for GNSS Product Marketing Manuel del Castillo told us, “We initially supported GPS and then progressively added all of the other major GNSS constellations – GLONASS, BeiDou and Galileo.

“We treat the different GNSS signals in an equivalent way, in terms of acquisition and tracking, and interchangeability in subsequent fixes. Our motivation has always been accuracy and yield improvements in challenging urban environments, where our customers have constantly pushed us to keep improving.”

The first Broadcom multi-constellation chip, adding GLONASS to GPS, came in 2011. “After that, in 2013, we added BeiDou,” del Castillo said. “And in 2014, two years before Galileo Initial Services were announced, Broadcom added Galileo.

“Our Galileo chip features a multipurpose, sensor-hub and sensor-fusion software for use in smartphones and tablets, as well as ‘system-on-chip’ architecture, so we can meet the challenge of always-on location with very low power. And of course our users benefit not only from the additional Galileo satellites, but from the new BOC modulation, which itself improves accuracy.”

In each case, del Castillo said, whether it was adding GLONASS, BeiDou or Galileo, similar processes were involved, including understanding the ICD, discussing implementation and testing initial prototypes, developing the B0 revision and carrying out receiver tests.

“In the case of Galileo, however,” he said, “there is a lot more support and clarity compared to previous constellations, in particular BeiDou. And Galileo is easier in terms of the RF part than GLONASS or BeiDou. On the other hand, Galileo’s baseband is more complex due to the longer codes, secondary codes and BOC modulation.”

Dual Frequency at Last

Last year, Broadcom launched the chip that changed everything. “The BCM47755 in includes support for dual frequency in both GPS L1 and L5, and Galileo E1 and E5,” said del Castillo.

Until now, mobile positioning and navigation devices have been powered by single-frequency GNSS receivers. The expanded availability of L1/E1 and L5/E5 frequencies, thanks especially to Europe’s Galileo constellation, now means reduced multipath and ionospheric interference, improving positioning in urban and other environments.

“We are working with a number of handset vendors, Samsung in particular, to bring multi-constellation and multi-frequency capabilities to your next smartphone,” said del Castillo. “The BCM47755 provides a high level of accuracy with minimal power consumption and footprint, and is capable of enabling an entirely new set of high-precision LBS applications.” These include lane-level vehicle navigation, advanced gaming apps, mobile augmented reality, car-hailing applications, driving assistance for cars, drone guidance, and many others that have yet to be imagined.

Del Castillo said the decision to go dual-frequency was an easy one to make. “Once we were sure that the cost increase was going to be tolerable and the technical benefits would be far greater than the added cost, we forged ahead quickly. Of course we were moving into unexplored territory for a mass-market chip. The development process was extensive and it involved a substantial chip revision.

“A major challenge was always going to be the ten-fold increase in complexity of L5/E5 signals, and the fine-tuning of the internal phase delays between L1/E1 signals and L5/ E5 signals. Also, there was the need for additional memory and computational resources, while reducing power consumption. And we had to do all of this with a minimal cost impact for our customers.”

“Bringing in Galileo and, most importantly, implementing the dualfrequency support has absolutely paid off,” del Castillo said. “The benefits for our customers start, of course, with the performance improvement, delivering reliable sub-meter accuracy even in difficult environments. But for Broadcom itself, supporting Galileo in E1 and in E5 means we are now clear leaders in the innovation race, and we are benefiting from joint actions with the GSA and working in partnership with ESA.”

Not Standing Still

Competitors beware; the gang at Broadcom shows no signs of resting on their laurels. In another example of innovative thinking, the company has recently partnered with Google to host some location-based applications within its GNSS chipsets. Activity recognition is one such application that uses smartphone sensors to determine what the user is doing, i.e. walking, biking, driving, etc. The application can run on the smartphone’s own applications processor, which uses quite a bit of power, or, now, it can be pushed down to a low-power processor in an onboard Broadcom GNSS chip.

Del Castillo said Broadcom is intent on staying smart, harnessing its multiconstellation and multi-frequency GNSS technologies to maintain and increase its advantage over all competitors. “We see further refinements coming in our E5 implementation, ultimately allowing us to deliver even higher accuracy for newer and even more exciting applications. As long as we keep moving the innovation bar higher and higher, we believe we can stay a step ahead of the competition, and the ultimate winner will be the GNSS user.”

The post GNSS Opinion Leaders – Manuel Del Castillo appeared first on Inside GNSS - Global Navigation Satellite Systems Engineering, Policy, and Design.

The goal of the new PNT effort, the Army said in a Feb. 27 special announcement, is to “provide position, velocity, and time data with the aid of non-GPS augmentation to the dismounted and mounted Soldier in GPS degraded or denied environments.”

The announcement on the Fed Biz Opps website was issued “for planning purposes” under solicitation number: Army_Program_Manager_Positioning_Navigation_and_Timing

To accomplish its goal the service’s PNT office said it is considering using the Consortium for Command, Control, and Communications in Cyberspace (C5) to pursue prototype A-PNT systems. C5 has an agreement that enables the Army to fund research using Other Transaction Authority (OTA), a mechanism created as a work-around for cumbersome procurement rules so that program managers can identify and fund a project in months instead of years. Department of Defense (DoD) managers can make OTA agreements valued at up to $50 million without additional approvals. Program managers can enter contracts for up to $250 million with a supporting determination from the department’s senior procurement executive according to a May 2017 article in the legal journal The Government Contractor.

“OTA is basically an alternative to the Federal Acquisition Regulation and it’s designed to make it easier for so-called non traditional defense contractors – think Silicon Valley – to make it easier for startup companies, cutting-edge, small technology companies that historically have had no interest in working with the government, Defense Department or otherwise,” said Charlie McBride, president of the Consortium Management Group, which is the management organization that owns and operates C5.

The way it works, he said, is that a program manager determines a need and comes to C5 with their requirements and a rough budget. C5 puts that information into the form of a request for white papers, which is then sent to all its members; they typically have several weeks to reply. Those organizations that are interested and can meet, or nearly meet, the requirements submit a white paper describing their approach. The government manager evaluates the papers, choosing the organization(s) he or she wants to work with. “There’s a process that takes over then that leads to an actual agreement between the government and C5 on behalf of our member,” said McBride. That last step, he said, takes 90 days. The project can also be expanded and even enter into production under OTA rules.

Organizations apply for membership online, McBride said. The C5 staff talks with applicants and check the applicant’s website, but approval is generally granted quickly “unless there’s something egregiously wrong.” Membership information is available at www.c5technologies.org/ though the organization was having website issues as SIGNALS went to press.

This full article can be read online at http://insidegnss.com/node/5811

The post Army Looks to Speed Assured PNT appeared first on Inside GNSS - Global Navigation Satellite Systems Engineering, Policy, and Design.

The Next Generation Operational Control System (OCX) being developed by Raytheon is essential for the Air Force to be able to fully utilize the capabilities of the GPS III and GPS IIIF satellites. The program is five to six years behind and, according to a March 2017 report from the Government Accountability Office, surged in cost from $3.6 billion in November 2012 to $5.5 billion in September 2016 – a 53.2 percent increase in fiscal year 2017 dollars. The December report by MITRE, obtained and described by Bloomberg News, put the cost at $6.1 billion.

The OCX contract was originally valued at slightly more than $1.5 billion with options, when it was awarded to Raytheon in 2010.

“That program is a perfect example of the problem we have in the Air Force,” Rogers (R-Alabama), chairman of the House Armed Services Committee Subcommittee on Strategic Forces, said at the forum in late February. “We’ve had that black hole for money for years and it’s still not able to be executed. And show me the person responsible for that. You can’t. It’s all those committees, they’re all pointing at each other. ‘Oh it’s not me; it’s them. It’s not me; it’s them.’ In the meantime we still don’t have the capability and we don’t see an end in sight, and this thing has been incredibly over budget.”

MITRE had reportedly suggested abandoning OCX in favor of upgrading the Lockheed Martin ground system currently being used. When asked if he wanted to see the Air Force make more changes to try to reform that program – or to have it try to stabilize and execute on the current program of record – Rogers indicated a need to push forward.

“I think we need to get it right,” he told a CSIS forum on the FY19 space budget. “I don’t think we should just stop with what it can do right now. But I don’t see that happening. I’ve been so disappointed in that program.”

Rogers shared the stage with Rep. Jim Cooper (D-Tennessee), his committee’s ranking member. The Strategic National Security Space: FY19 Budget Forum was held February 28 in Washington, D.C.

“Properly understood,” said Cooper, “the Constitution makes Congress a board of directors. We shouldn’t micromanage. We should look at the big policy decisions and then let capable services implement them. But this is a situation that really is a nightmare. To have satellite capability and no ground communications for over a decade and really no hope in sight – and with no accountability. This would never stand for a second in the corporate world. And the corporate world is not perfect but at least there tends to be accountability.”

Cooper, who did not suggest during the forum that OCX should be changed or dropped, agrees with Rogers that there should be a separate Space Corps as a way to elevate and speed military space programs and increase accountability.

This full article can be read online at http://insidegnss.com/node/5812

The post Key House Chairman Sticks with GPS OCX Despite Issues, MITRE Criticism appeared first on Inside GNSS - Global Navigation Satellite Systems Engineering, Policy, and Design.

1. Pipeline Inspecting Blimp

South San Francisco, California

Mothership Aeronautics is using autonomous solar-powered blimps to disrupt long-distance aerial data collection. The TerraSoar Aerial Intelligence Airship features a powerful combination of lighter-than-air lifting and high efficiency solar cells that enable higher endurance and longer-range flights than possible with drones, while also operating heavy payloads like LiDAR.

The name, TerraSoar, stems from a family of prehistoric flying reptiles. The Pterosaurs were the first vertebrates known to take to the clouds just as the TerraSoar is the first blimp-type drone to hit the drone inspection market and Mothership Aeronautics’ first Aerial Intelligence Craft.

2. Remote Control Fireboats

<em?>Vancouver, British Columbia, Canada

To address the evolving safety and security needs of modern ports, Vancouver-based naval architects and marine engineers Robert Allan Ltd., and international marine technology specialist Kongsberg Maritime are collaborating on the development of a radically new remotely-operated fireboat that will allow first responders to attack dangerous port fires more aggressively and safer than ever before.

The Kongsberg Maritime control & communications system will feature a robust high-bandwidth, low latency wireless link to a semi-portable RALamander operator console that can be located on a manned fireboat, or other vessel of opportunity such as a tug boat or pilot boat. In common with other KONGSBERG autonomous control systems, the architecture of RALamander’s control system will leave the door open to a range of autonomy levels, which are configurable or future-upgradable to suit the operator or port’s evolving needs.

3. Cereal-Box Satellite

Aalborg, Denmark

The European Space Agency’s first mission of the year, launched in February, was the GomX-4B, the ESA’s most advanced technology-tester yet, featuring a hyperspectral camera and tiny thrusters to maneuver thousands of kilometers from its near-twin to try out their radio link.

These CubeSats are built around standard 10×10 centimeter units by GomSpace in Denmark. As “six-unit” CubeSats they are as big as cereal boxes – but double the size of their predecessor GomX-3, released from the International Space Station in 2015.

“ESA is harnessing CubeSats as a fast, cheap method of testing promising European technologies in orbit,” comments Roger Walker, heading ESA’s technology CubeSat efforts. “Unlike GomX-3, GomX-4B will change its orbit using cold-gas thrusters, opening up the prospect of rapidly deploying future constellations and maintaining their separations, and flying nanosatellites in formations to perform new types of measurements from space.”

Image credits
1. Mothership blimp from the company: Mothership Aeronautics
2. Fireboat from Robert Allan Ltd.
3. Cereal-box satellite from the European Space Agency

The post GNSS Hotspots – Pipeline Inspecting Blimp, Remote Control Fireboats, Cereal-Box Satellite and More appeared first on Inside GNSS - Global Navigation Satellite Systems Engineering, Policy, and Design.

An end-to-end simulation is presented using the Positioning and Integrity Performance Evaluator (PIPE). The simulation includes side by side comparison of a vehicle path in the presence of spoofing as evaluated by the authors’ Generalized Pseudo Bayesian 1 (GPB1) algorithm and a snapshot least squares algorithm. Every field application has its own operational conditions and resulting requirements with respect to accuracy, availability, and continuity for systems that provide position, velocity, and time (PVT) measurements. For example, Global Satellite Navigation Systems (GNSS) found application for certain approach procedures in commercial aviation due to the advent of Satellite Based Augmentation Systems (SBAS) that provide additional integrity information to GNSS receivers. GNSS reception environments for aircraft are nominally clear-sky conditions and the combination of GNSS, SBAS, and advanced processing techniques like Receiver Autonomous Integrity Monitoring (RAIM) yields the required compliance for probabilities of detecting malfunctions and times-to-alert.

Providing absolute, three-dimensional positions on Earth together with well-proven and cost-efficient receiver technologies predestines GNSS equipment to enter the emerging markets of land-based users such as autonomous vehicles and the railway industry. However, since the operational conditions of these applications differ dramatically from those present in the aviation world, PVT measurements with integrity based on GNSS as previously developed cannot be translated directly to land-based users.

This article presents the scope and an application example of the Integrity for Navigation Land Users (INLU) research study as part of the European Space Agency’s Technology Research Program (see Additional Resources, J. Wendel et alia (2016a)) that develops techniques to provide PVT solutions to land-based users within defined integrity bounds.

The environments in which landbased users move impose reception limitations on GNSS receivers. In cities, the GNSS signals are frequently blocked and reflected by buildings, structures, and other traffic, for example. The same is true for railway applications. Here, GNSS blockages arise from features ranging from shunting yards over railway stations to rides through cuttings and tunnels.

To be able to create new satellite navigation technologies that provide PVT information with integrity and solutions for high-precision positioning even under difficult conditions, a well-suited end-to-end simulation tool is required. End-to-end means that realistic signals from complete satellite constellations can be generated which take, for example, transmitter and receiver antenna characteristics, atmospheric and local wave propagation mechanisms, receiver dynamics, and other distorting effects into account.

Often such end-to-end simulations are performed using a radio frequency constellation simulator (RFCS) and hardware GNSS receivers. A number of commercial off-the-shelf solutions are available for each of these items. However, for the development of said algorithms, it is more efficient to be able to use pure software or hybrid hardware/ software solutions. This allows for a much higher frequency for the development-integration-test cycles of a project.

To respond to INLU’s requirements and to meet today’s needs of GNSS endto- end simulations, Airbus Defense and Space has been developing the Positioning and Integrity Performance Evaluator (PIPE) as a research activity since 2012. Beyond INLU, PIPE found applications for the development of novel tracking (F. M. Schubert et alia (2014a); F. M. Schubert et alia (2015)) and channel model research (I. Gulie et alia; F. M Schubert et alia (2016)), to name a few examples.

The article’s organization follows the high-level procedure of getting results from an end-to-end simulation run using PIPE. We first describe the scenario definition in terms of user trajectory, satellite constellation, and propagation channel conditions. Next, the configuration of the simulated receiver is reported with methods for acquisition and tracking of GNSS signals and computation of the PVT solutions. An example is then given that reports a filter bank approach developed during the INLU project. It is used for mitigating the influence a spoofer has on the receiver’s performance.

Scenario Generation

Figure 1 reports the general structure of end-to-end GNSS simulations that can be performed using PIPE. PIPE consists of three groups of programs: tools for digital signal processing (DSP), a GNSS scenario and constellation simulator, and a GNSS receiver. PIPE’s DSP provides classical signal chain simulations consisting of a signal source, one or multiple signal processors, and a signal sink. Among other programs, PIPE includes GNSS signal generators, filters, up- and down-converters, and interference signal generators.

PIPE’s scenario and constellation simulator programs are able to generate user trajectories, satellite positions for given times and orbits, and the propagation channel response based on multipath components. As this group of scenario-related programs differs from the DPS programs, they are called SNIPE tools — short for Scenarios for Navigation and Integrity Performance Evaluation. INLU also requires the possibility to process real-world signals. Signal sources for the PIPE receiver can be sampled signals as sensed by antennas, the PIPE software GNSS signal generator, or samples recorded from an RFCS’s output.

PIPE accommodates INLU’s diverse requirements by a modular approach of signal chain simulations: Certain software elements of the chain can be replaced by hardware components. Additionally, interfaces to front-end sampling and replay devices are available.

The following subsections describe the creation of the user’s trajectory, the simulation of a satellite constellation as well as the processing of the response stemming from a propagation channel model.

Trajectory Generation

The first step in the scenario generation process is the creation of the user’s trajectory. This trajectory serves as input for the generation of GNSS observations, as well as further sensor data like odometers, inertial sensors, baro-altimeters, and magnetometers. The challenge hereby is to produce consistent dynamics data. When the accelerations and angular rates provided by the trajectory generator are used for the generation of inertial sensor data assuming an ideal inertial measurement unit, the output of an ideally initialized strapdown algorithm must match the original trajectory exactly, even after longer simulation times. This is achieved in the trajectory generator first by producing desired positions and attitudes over time. Then, a strapdown algorithm together with a flight control-like algorithm is applied. From the small offsets between desired positions and attitudes and the strapdown state, accelerations and angular rates are generated, which are provided to the strapdown algorithm in the next epoch and drive these offsets to zero.

Constellation Simulation

From the trajectory generation, the GNSS antenna positions and velocities also can be calculated at equidistant points in time. For each of these points, the constellation simulation needs to calculate the corresponding satellite positions and velocities based on RINEX files. The trajectory time scale defines the times of reception of the satellite signals. In order to calculate the satellite positions and velocities, the times of transmission at the satellite need to be determined. This is achieved by approximating the satellite orbit in a time interval of T=0.1 seconds by a straight line, which is accurate to the sub-millimeter level. Denoting the time of reception with t0, the satellite position pS at the time of transmission t0–t can be expressed as

It must now be considered that the ECEF frame is rotating. The ephemeris describes the satellite position in the ECEF frame at the point in time for which a satellite position is calculated. In order to express the satellite position at t0–T in the ECEF frame that is valid at t0, the rotation of the ECEF frame in the time interval T needs to be considered:

In the following, pS(t0–T) always refers to the satellite position at t0–T, expressed in the ECEF frame at t0. In order to determine the time of flight of the satellite signal t, the equation relating the range between satellite position at time of transmission, pS(t0–t), and antenna position at time of reception, pA(t0), is used:

Here, c denotes the speed of light. Inserting the linear approximation of the satellite orbit and squaring the equation leads to

This is a quadratic equation that can be solved for the time of flight, t. Consequently, the satellite position at time of transmission, pS(t0–T), in coordinates of the ECEF frame at time of reception, t0, is obtained. The carrier phase, code phase, and Doppler measurements obtained at t0 can then be generated using appropriate error models.

Propagation Channel Models

The reception conditions for land-based users are impacted to a large extent by multipath propagation caused by objects and structures in the receiver’s vicinity. Moreover, in urban areas, signals are often blocked and diffracted by buildings and trees. State-of-the-art multipath propagation models reproduce these effects and generate channel impulse responses (CIR) at a given rate dependent on the receiver’s dynamics. CIRs contain components that reflect the complex amplitude and delay of line-of-sight as well as multipath components.

PIPE’s interface to a channel response generated by a multipath propagation model is given by the Channel Data Exchange format (CDX) (see CDX – Channel Data Exchange Library in Additional Resources). The PIPE GNSS signal generator can read the channel impulse response for every time step from a CDX file and generate the corresponding multipath components with their respective delays and complex amplitudes in the output signal. This allows for the usage of various channel models for INLU. Within the project, the model recommended by ITU-R P.681 for urban environments was used. Additionally, this model was extended for railway applications during a research activity (I. Gulie et alia). Figure 2 shows the visualization of a common railway scenery.

The INLU project also requires the generation of RF signals by hardware constellation simulators using the mentioned channel models. As these models produce more multipath components than a common hardware constellation simulator can re-produce, a component count reduction step is required. In INLU, a method based on F. M. Schubert et alia (2014b) is applied.

Land User Receiver Prototype

The PIPE GNSS Receiver used to run the INLU simulations is not only able to process samples, bit-true mode in INLU terms, it has also a semi-analytic operation mode that calculates correlation results based on pre-computed autocorrelation functions of GNSS signals. The semi-analytic mode runs multiple times faster than the bit-true mode and allows for simulation of long runs. The scenario input data as well as the receiver’s implementations for items such as tracking loops and PVT computation are identical for both modes.

Tracking Algorithms

In addition to standard tracking methods like Early-Minus-Late and Bump Jumping, the PIPE receiver offers a number of state-of the-art tracking techniques, such as the Double Delta Correlator, Kalman filter-based methods, Double Estimator, and Maximum Likelihood-based methods, such as Multipath Estimating DLL and Vision Correlator. All these methods are scrutinized during the INLU project. For the following example, the Astrium Correlator (AC) is applied which offers robustness against locks to false side peaks while tracking binary offset carrier (BOC) signals (F. M. Schubert et alia (2014a)). For signal tracking, the AC uses a BOC’s signal subcarrier to exploit the higher accuracy that can be achieved when tracking such signals to the legacy binary phase shift keying (BPSK) signals. At the same time, the AC checks if it tracks the BOC signal’s central peak via the observation of the BPSK envelope of the signal. If a lock to a side peak is detected, the tracker is commanded to switch its tracking point toward the correct main peak.

Integrity Algorithms ARAIM Tailored to the Railway Environment

The main objective of INLU was the development of integrity algorithms tailored to the land user environment. The following integrity concept for railway users was developed as an INLU scenario.

Providing integrity for train position information is a major technical challenge due to the small integrity risk that is tolerated in railway applications. In the current implementation of the European Rail Traffic Management System (ERTMS), the train position is propagated using odometry, and corrected when a balise group is reached. A balise is a transponder that provides an absolute location reference to the on-board unit of the train, allowing the train to locate itself within a movement authority. GNSS positioning is being considered in the context of the ERTMS evolution for the realization of a virtual balise concept, where GNSS is used for the detection of virtual balises. The approach of virtualizing the balise transmission system aims to reduce the cost of trackside infrastructure associated with the installation and maintenance of physical balises, while minimizing changes to the existing system and maximizing interoperability. Consequentially, requirements on the integrity of position information provided by the GNSS receiver are very stringent.

The integrity concept within INLU can be seen as a step towards a realization of the virtual balise function. A block diagram of this concept is shown in Figure 3. The integrity algorithm processes pseudorange measurements from a GNSS receiver, measurements from odometry, and exploits information contained in a map database. It consists of two major building blocks, a module for pseudorange measurement rejection, and the integrity module that calculates the projection level.

The measurement rejection module consists of a Kalman filter that integrates the odometer measurements with the track database and the pseudoranges from the GNSS receiver. The a priori position available from this filter is used to calculate the Mahalanobis distance of each pseudorange, and upon excess of a pre-defined threshold for the Mahalano-bis distance, the respective pseudorange is rejected.

The integrity module calculating the projection levels uses the pseudoranges that passed the Mahalanobis distance check in the measurement rejection module as well as information from the track database. In consequence, the position solution provided by the integrity module is independent of the odometry, in the sense that the odometer readings do not enter in the position calculation. For projection level calculation, a solution separation approach is used. In solution separation RAIM, the spread of subset position solutions with respect to the full set position solution is assessed. Subset position solutions are obtained by excluding subsets of satellites from the position calculation. Thus, the number of satellites that are excluded must reflect the number of simultaneous faults that need to be considered. The probability that a higher number of simultaneous faults occurs is very low, but might still be included in the integrity risk budgeting.

The major difference between a conventional solution separation RAIM outlined above and the proposed integrity concept is that from the pseudoranges which have passed the measurement rejection module, a three dimensional position solution is not calculated, but rather a GNSSbased odometer distance results. Using this GNSS-based odometer distance, a three-dimensional position solution can then be obtained from the track database. Consequently, for the calculation of protection levels, the spread of GNSS-based odometer distances is assessed, not the spread of position solutions. Obviously, calculating a GNSSbased odometer distance instead of a three dimensional position reduces the number of unknowns from four to two, which means that with two pseudoranges only, a solution can be obtained. More details of this integrity algorithm are given by J. Wendel et alia (2016b).

Filter Bank for GNSS and INS Integrity

Within the INLU project, integrate navigation systems were also addressed, in which the software GNSS receiver is combined with inertial sensors in loose, tight, and ultra-tight coupling architectures. Such architectures do not allow for the calculation of protection levels using ARAIM algorithms because these require that full set and subset position solutions are calculated using snapshot least squares.

A variety of techniques can be found in literature which aim at the provision of integrity for integrated navigation systems. Batch processing approaches re-formulate the GNSS/INS data fusion as a least squares problem, which then allows us to apply RAIM or ARAIM techniques (M. Joerger and B. Pervan). Another option is to use a filter bank. Each elemental filter in the filter bank is robust with respect to a specific fault.

In the simplest case, an elemental filter does not process the measurements of a specific satellite. In case the measurements of this satellite are faulty, this filter is not affected. This also avoids the need for a pseudorange fault model which is an advantage because such a model is in general rarely available. Examples of filter bank approaches can be found in (M. Brenner; J. Diesel and S. Luu), with the basic concept illustrated in Figure 4 with three filters only. Obviously,
a real GNSS/INS filter bank contains many more filters. For the single fault case, the number of elemental filters matches the number of satellites from which measurements are available, with possibly an additional elemental filter assuming no faults are added. Each of the elemental filters propagates its state and covariance matrix forward in time using the measurements provided by an inertial measurement unit (IMU).

When pseudorange and delta range measurements become available, they are processed by each elemental filter except for those the elemental filter assumes to be faulty. Hereby, the model probabilities are also updated. For each elemental filter, the model probabilities represent the likelihood that the assumptions of the filter are correct, i.e., that the satellites that the elemental filter assumes to be faulty actually are faulty. This update is based on the ability of the filter to predict measurements. The better the elemental filter pseudorange and delta range predictions match the actually available measurements, the higher the model probability.

After the measurement processing, a mixing step is executed. In this mixing step, each elemental filter is reinitialized with a new state estimate and covariance matrix, which are calculated from the model probabilities, state estimates, and covariance matrices of all elemental filters. It is important to note that for the most widely used filter bank, i.e., the Interacting Multiple Model (IMM) filter bank, all elemental filters are initialized differently. Therefore, state and covariance of each elemental filter must be propagated separately, even if all elemental filters assume the same system model.

The main drawback of such a filter bank is the huge computational load. In most integrated navigation systems, most of the computational cost is spent in the propagation step. The reason for this is that the state and covariance matrices of the navigation filter must be propagated with a reasonable update rate in order to cope with the vehicle’s dynamics. For example, in a GNSS/INS system, several propagation steps are performed (for example, every 5 milliseconds when the inertial sensors provide measurements) before one measurement step takes place, i.e., every second when a typical GNSS receiver provides measurements. With the values given in this example, 200 propagation steps are performed before one measurement step is performed.

Within the INLU project, a GNSS/INS integrity algorithm based on a Generalized Pseudo-Bayesian 1 (GPB1) filter bank was developed. The only difference between an IMM and a GPB1 filter bank is the mixing step. For the GPB1 filter bank, the mixing step initializes all elemental filters identically. As all the elemental filters have the same system model — namely the error propagation equations of inertial navigation plus additional states to estimate the inertial sensor biases — the use of a GPB1 filter bank allows INLU to perform propagation steps with one elemental filter only, instead of with each elemental filter of the filter bank. Then, when GNSS measurements become available, all elemental filters are initialized with the propagated state and covariance of this first elemental filter before each elemental filter processes the measurements. This approach avoids, to a large extent, the increase in processing complexity typically connected to filter bank integrity algorithms.

Application Example: Spoofing Scenario

The previously introduced technique is demonstrated with a scenario employing a spoofer using a hardware constellation simulator. The scenario simulates a short drive of a land vehicle. The multipath environment was generated according to the ITU-R P.681 channel model; nominal Galileo and GPS constellations were assumed. Additionally, an ideal spoofing of one Galileo and one GPS Open Service signal was simulated. From a certain point in time onwards, a ramp on the pseudoranges of the respective satellites was generated, starting from zero. Using this RFCS scenario, baseband samples were recorded and then post-processed with the PIPE Receiver. Additionally, artificial inertial sensor data simulating a medium-grade MEMS was generated.

The results of the post processing of the recorded baseband samples are shown in Figure 5. The trajectory starts in the left upper corner and follows the road that is visible in the Google Earth picture. Two different PVT solvers were used: a snapshot least squares solver that produced the position fixes indicated by the red markers, and the GNSS/INS filter bank approach described in the previous section, indicated by the blue markers. Obviously, the snapshot least squares position fixes, for which no attempt to counter spoofing was made, walks off the road. In comparison, the GPB1-based GNSS/INS integrity algorithm proves to be robust with respect to the simulated spoofing threat.

Conclusion

To further grow the application of GNSS receivers for landbased applications, the integrity of the computed PVT solutions must be ensured for this user community in a comparable fashion as is done for the aviation industry. In doing so, the integrity risks of the respective applications like vehicles in cities and railways have to be accounted for. The list of nominal and non-nominal threats is comprehensive, ranging from false locks during tracking over unintentional interference and intentional counterfeiting of GNSS signals, i.e., spoofing.

During the INLU project, various state-of-the-art and promising candidates of future receiver techniques are studied to understand how they can contribute to robust PVT results for land-based users. This comprises not only tracking and RAIM algorithms but also integration of GNSS with inertial measurements using, for example, filter banks.

A GNSS end-to-end simulator needs to have a high versatility to be able to accommodate the manifold requirements that such a project imposes on it. Key factors for the achieved flexibility of the PIPE and SNIPE tools are:

• partitioning into units that serve single tasks with clear inter-module and input data interfaces,
• abstraction layers that allow for the addition of tracking and PVT computation methods to the receiver with minimum effort and the ability to operate these in bit-true and semianalytic modes.

The effectiveness of the chosen approach was demonstrated in a scenario where a traditional GNSS receiver was misled by a spoofer while the implemented filter bank approach corrected for the error transparently.

A separate article will report on the entirety of INLU’s results, i.e., the conclusions drawn from results gained from a number of stochastic scenarios.

The post Navigation Integrity for Land Users Robust Positioning in Challenging Environments appeared first on Inside GNSS - Global Navigation Satellite Systems Engineering, Policy, and Design.

Among the hot topics in Munich was the role of GNSS in the burgeoning field of automated and unmanned systems. Michael Baus, Program Director at Robert Bosch GmbH, described his company’s fresh approach to precise point positioning (PPP) for autonomous road transport.

“Our new vehicle motion and positioning sensor (VMPS) will hit the road in 2020 and will be one of the first GNSS inertial positioning systems using correction data for highly automated driving,” he said.

“The VMPS will output a safe position, velocity, attitude and time, using a multi-frequency, multiconstellation GNSS system and our sophisticated fusion and integrity algorithms. The system will use wheel spin sensors as well as automotive-grade, safe inertial sensors to bridge GNSS outages.”

Yes, that’s right, he said “safe” two times. And then he said it again, this time with a “not”: “You all know GNSS alone cannot be safe; we cannot control the infrastructure. So, we are using a correction service consisting of a dense network of reference stations spaced about 250 kilometers apart, and we’re starting in Europe, North America, going to China, Japan, South Korea and the rest of the automotive countries.” Processing centers, he explained, will calculate orbit and atmosphere corrections.

Data is transmitted via geostationary satellites as well as over cloud connection to the vehicle, Baus explained, and the VMPS uses it to calculate the output signals and the all-important integrity information.

“We are working in partnership with Trimble for safe and precise positioning, and with u-blox for automotive safe positioning, and, together with Mitsubishi Electronics and Geo++, we are a stakeholder of the Sapcorda joint venture, whose target is worldwide safe and precise correction data.”

SBAS for All?

Another way to better and surer precision is through space-based augmentation systems (SBAS), such as the GPS Wide Area Augmentation System (WAAS) in the United States or EGNOS in Europe. These regional satellite systems provide GNSS correction data, giving increased accuracy and integrity for key “safety-of-life” applications, including civil aviation.

“Regions like the U.S., Europe, Russia or China may be able to afford to build their own SBAS systems,” said Miguel Romay, Executive Director of GNSS Aerospace at Madrid-based GMV, “but what if you are in a plane flying from Washington to Rio de Janeiro? In the U.S. you have SBAS, but in Brazil you don’t.”

Romay was in Munich to talk about GMV’s brand new “magic” user terminal, which uses SBAS capabilities onboard an Inmarsat GEO satellite already in orbit to provide GNSS correction over a wide area. “We thought there was a possibility to develop a company that provides SBAS services to different countries,” he said. “So, they don’t have to develop their own space-based system.

“We are cooperating with Lockheed Martin in the U.S., and we have a project in the Australia/New Zealand region where we are already transmitting through Inmarsat.” Specifically, GMV is providing the processing facilities in charge of the augmentation system, Lockheed Martin is doing GEO satellite signal uplink, and Inmarsat is responsible for the SBAS payload on the 4F1 satellite.

“Having multiple constellations allows us to move in the direction of a global, not regional, SBAS,” Romay said. “We have an operational test bed. All the infrastructure was completed in October of last year, and since then we have been testing it in different fields; maritime, aviation, mining.”

Romay showed us a “demo” receiver, the key components of which, he said, can actually fit inside a smartphone. “I believe this is the first dual-frequency, multi-constellation SBAS receiver, giving PPP with integrity. Using this in Australia, you can receive the signal through the SBAS satellite, and if you are outside Australia you use the internet to receive the SBAS and PPP corrections.”

GMV intends to offer two different products: a lower-cost, single-frequency, PPP receiver, equipped with the u-blox M8 chip and delivering 40-50 centimeter accuracy with integrity; and a higher-performance, NovAtelequipped receiver using GPS, Galileo and GLONASS and capable of achieving around 5 centimeter accuracy with integrity.

“Many people at this conference are talking about PPP with integrity,” Romay said, “so, here it is, we have a solution that we are ready to bring to market, and it can be integrated with other technologies and sensors.”

Galileo CS Reverberations

Things do appear to be moving on the PPP front. For its part, the European Commission (EC) continued in Munich to defend its recent decision to provide for free a high-accuracy GNSS service, originally conceived as one component of a Galileo fee-based commercial service (CS).

On the industrial side, where a number of companies have already developed and are delivering high-precision positioning for a fee, some wondered out loud whether the Commission’s move will undercut their own business. Of these, President of Hexagon Positioning Intelligence Michael Ritter was among the more outspoken. He expressed some confusion as to what “high-accuracy” actually means in the context of the new Galileo service.

“We’ve heard different definitions of what that is – 20 centimeter accuracy? Higher? What convergence time?” he asked. “Because what gives you 20 centimeter in one minute gives you one centimeter in five minutes, so there’s a lot of ambiguous vocabulary being used right at the moment.

Ritter pointed out that existing markets of survey, mapping, agriculture and offshore are already well served by existing European PPP and RTK correction networks and services, and that Hexagon Positioning Intelligence has been providing correction services for three decades. While industry is already working to solve the next challenges of GNSS correction services, he said, the provision of free services will remove or severely diminish the revenue source that industry relies on to reinvest in research and innovation for the autonomous future.

About the American companies in the room, he said, “Most of their employees working on PPP are actually in Europe as well. We all need that money to feed our R&D chain and that ’s why of course we are opposed to that free service.

“The reality is the only way we can get to the kind of functional safety and integrity we need, that kind of authentication, that kind of accuracy in a short time, which involves not a few but thousands and thousands of reference stations, we need to finance that, or we all sit back and wait 15 years for the government to do it. I’d prefer the EC and GSA would use these funds to speed up the Galileo authentication service.”

Commission Come-Back

In response, European Commission Galileo Commercial Service Manager Ignacio Fernandez-Hernandez said, “We believe that in the long term high accuracy is becoming cheaper and will eventually be free.” Hence, the Commission’s desire to get in front of this inevitable wave.

“Some are asking how Galileo high accuracy will affect the industry. But what we are proposing is not comparable to an end-to-end guaranteed service. What we intend to offer stays at the signal level. This relates only to providing better satellite information, better atmospheric information, which is what satnav providers have been doing for the last decade.”

Fernandez reminded participants of the steady trend towards increased accuracy even among open GNSS services available to the general public, and then he said, “You, the industry, are building partnerships and you are evolving your business models. With
Galileo high accuracy, maybe some existing services may not be as relevant as before, but there will be other new services.

“Our perception is that companies are ready to integrate high accuracy with other services, so there is still room
for innovation.”

Ritter returned to the question of what exactly Galileo intends to offer: “I don’t think the GSA [European GNSS Agency] has made any statement on what the convergence time is going to be, so 20 centimeters in 30 seconds? 20 centimeters in a much faster time? Between Trimble, Fugro, Deere and Hexagon and others, we invest a lot of money into fast convergence and high accuracy, so it’s actually an answer I would like to hear.”

“Just to be accurate, convergence is not a service that will be provided by Galileo,” Fernandez said. “Convergence depends on the user algorithms and the information that these algorithms process, so we are providing part of this information, but we are not investing in developing convergence algorithms or developing end-user solutions. These will 100% remain the business of companies providing high accuracy services.”

As for the stated 20 centimeter accuracy level, Fernandez revealed that that figure actually comes from a “higher level”, which we take to mean a more political level.

“We are talking about a very highlevel text,” he said, “which is a Commission ‘Decision’. We have this benchmark of 20 centimeters that expresses the willingness to relax accuracy compared to the service as it was originally defined as a payable service. This is in order to interfere less with existing markets and serve some end-consumer applications. But we are just at the start of the process of defining the service. The statistical characterization, the user environments, the baseline algorithms, all that is still to come.”

Authentic GNSS

While some participants were making a splash, others were offering cash. “If you have any brilliant ideas we have some funding for you,” said GSA Market Development Officer Reinhard Blasi. Then, having peaked everyone’s interest, he proceeded to discuss a number of other items of interest to the GSA, and to everyone else of course, such as market perspectives on authentication and the trend towards authentication as a priority for safetycritical applications.

For the Galileo program, authentication comes in or will come in three different forms. First, there is Open Service Navigation Message Authentication (OSNMA), which will provide a basic level of authentication and some anti-spoofing protection. Blasi said, “This signal will be disseminated over the E1 frequency and is available for singlefrequency users. It is aimed at consumer users and offered for free, already prototyped and under testing.”

A much higher level of authentication will be provided by the robust and secure, limited-access Galileo Public Regulated Service (PRS).

In between those two will be the “new” Galileo Commercial Service, comprising a data authentication signal that will be access-controlled and provided for a fee, based on the spreading code encryption of E6C plus some ancillary data in E6B/ E1B, including OSNMA.

Interesting to note is that even just within the services provided by one GNSS constellation, Galileo, there will be multiple authentication options available, with the choice falling to end users of determining their own vulnerabilities and security needs. But how, someone asked, will users know which level of protection they really need?

Authentic Corrections

Hexagon Positioning Intelligence Vice President of Innovation Sandy Kennedy said she sees military-like GNSS requirements, both in terms of accuracy and security, moving into commercial areas. “There is an increased risk to all established GNSS applications, and with new applications such as autonomous vehicles and verified location for financial transactions coming on quickly, pseudorange-based positioning is not sufficient. Authentic correction sources are now as important as authentic GNSS signals.”

We simply cannot defend against all kinds of attacks, Kennedy said, and those attacks will evolve. So GNSS providers will need to be vigilant, proactive and quickly reactive as new information presents.

To end-users, she said, “So, do you want a sports car or anarmored car? A person who decides to buy a sports car values certain things; they want the engine, they want the speed, the maneuverability. Somebody who has an armored car, it’s still a vehicle, but they value the security. They’ve probably given something up in terms of driving performance, but it’s because safety is their priority.”

It’s the same with GNSS receivers, she suggested. “Which of the parameters are most important for a particular user and why do they choose one or the other? We’ve heard about having a very quick time to first fix, when you drive out of a parking garage and want a solution instantly. Well you can get that but you’re going to have to wait some time before that’s an authenticated or verified solution – or you might not be able to run your RTK (real-time kinematic) solution at faster than 20 hertz, or we might not be able to track every single satellite in view, and so on.”

Are users ultimately going to have to choose a level of vulnerability they are willing to accept? “It really depends on your perception of risk,” Kennedy said. “What we are working on is what level of protection can be provided by default in the background without causing any compromises to performance. How much CPU (central processing unit) can we be using to run a bunch of interconstellation and intersignal checks and that sort of thing.

“Right now what we are seeing is customers that can articulate very clear needs in terms of accuracy levels in this many seconds, but we don’t have that kind of specific definition by customers in terms of interference detection.”

And why is that?, someone interjects. “I think people just don’t know.” Bingo. So there’s an education process that still needs to take place. Kennedy borrows from an old Canadian TV comedy sketch: “GNSS users, you need to know what you need to be afraid of!”

Money, Honey

But really, what about that funding? Between 2014 and 2020, the GSA’s Reinhard “The Candyman” Blasi said, altogether 111.5 million Euro (approximately $138 million) will have been invested by the GSA, at the behest of the EC. This is mainly for Horizon 2020 Fundamental Elements projects, aimed at developing the likes of EGNOS and Galileo-enabled chipsets, receivers and user terminals.

“We want to build on innovative services and differentiators,” he said, “while addressing real user needs and maximizing the benefits for EU citizens.”

One of the key stated aims of all H2020 funding has been to increase industrial competitiveness within the EU, which means the program prioritizes EU participation, but parties from third countries can get funding under certain conditions.

Upcoming 2018 and 2019 grants, not yet awarded, will go to projects aimed at developing Galileo-based timing receivers, CS user terminals and enhanced GNSS user terminals. So, there you have it. Any brilliant ideas? Look for Reinhard Blasi.

The post Munich Hot Licks: Satellite Navigation 2018 appeared first on Inside GNSS - Global Navigation Satellite Systems Engineering, Policy, and Design.

It is important, therefore, to ensure satellite orbit calculations are as accurate as possible. As discussed in this article, Earth rotation plays a key role in this regard but surprisingly few references on orbit calculation actually mention its affect explicitly or how to compensate for it. Don’t fret, however, the correction is certainly applied or positioning accuracy would be much worse than is currently attained.

Reference Frames

Earth rotation is important because of the choice of reference system in which orbital calculations are performed. In particular, GNSS orbits — either from the broadcast orbital models or precise post-mission estimation — are parameterized in an
Earth-Centered Earth-Fixed (ECEF) coordinate frame such as the WGS84 reference frame used for GPS.

A common definition of an ECEF frame is one whose z-axis is the rotational axis of the Earth (pointing north), whose x-axis is in the equatorial plane and includes the median passing through Greenwich, and the y-axis completes the frame (typically in a right-handed sense). By definition,such a frame rotates with the Earth and is thus time-varying in inertial space with a period of 24 hours.

In the context of satellite position computations, this means that satellite locations can be computed at any given time, in an ECEF coordinate frame that is valid at that same time.

An easy way to visualize this point is to consider an ideal geostationary satellite whose position relative to the Earth does not change over time — orbital parameters or orbital files would always yield the same coordinates for the satellite.

Effect of Earth Rotation

So where does Earth rotation enter the picture? Well, precisely from the fact that the time at which a satellite transmits a signal, and the time a receiver receives that signal differs. Between the time of transmission (tt) and the time of reception (tr) — roughly 70 milliseconds (give or take few milliseconds) for medium-Earth orbiting (MEO) satellites — the Earth has rotated by ωe . (tr – tt), where ωe is the rotation rate of the Earth.

To illustrate the effect of this, we return to our idealized geostationary satellite. We further consider a user located directly below the satellite. Figure 1 shows this situation looking down on the north pole. To simplify later discussions, we consider this figure to apply at the time of signal transmission.

Since the orbital radius of a geostationary satellite is known (approximately 42,164 kilometers) and the radius of the Earth is known (approximately 6,371 kilometers) the separation of the user and satellite at any given instant is constant and can be easily computed.

Now consider Figure 2, which shows the same figure but also includes the location of the user and satellite at time of signal reception. Because of Earth rotation, the signal travels the path denoted by the blue line, which is obviously longer than the instantaneous separation of the satellite and user. This is the path in inertial space (ignoring the Earth’s orbit around the sun for simplicity).

The problem, however, is that because orbits are parameterized in an ECEF frame, the computed position of the satellite will still be directly above the user. This leads to a situation where the true signal path and the computed signal path differ. Unless accounted for, this difference will manifest as a ranging error in the receiver’s position engine, which computes the difference of the measured and predicted signal paths (i.e., ranges). The magnitude of the position error depends on the number and distribution of satellites, as well as user latitude. As an example, in Calgary, Canada, ignoring Earth rotation results in a shift in the estimated user position of about 20 meters, primarily in the east/west direction.

Before moving on, although we used the example of a geostationary satellite, the exact same effect applies to non-geostationary orbits as well. The main difference is that the satellite positions in Figures 1 and 2 would not necessarily be directly above the user, and the distance between the user and satellites, projected into the equatorial plane (which is shown in Figures 1 and 2), will vary with time as satellites move along their orbits. The good news is that regardless of the orbit, the method of compensation is the same.

Simple Solution

To remove the discrepancy between the measured and computed signal paths, we need to compute the ECEF position of the satellite at the time of transition in the ECEF frame at the time of signal reception. Fortunately, this is easily accomplished by realizing that the two coordinate frames are related by a rotation about the z-axis.

Mathematically, we can write

where is a position vector at the subscripted time (or frame), and R3 (ωe . (tr – tt)) is the rotation matrix about the z-axis by the angle subtended by the Earth rotated during signal propagation.

Applying the transformation in (1) yields the position of the yellow satellite in Figure 2, which allows for the proper computation of the (orange) user position.

The astute reader might be wondering how the propagation time is computed. This can be found by iterating to a solution: first, assume an initial distance between the user and satellite (e.g., 70 milliseconds); then compute the satellite position using this assumed distance (for Earth rotation compensation); use the approximate user position to re-compute the range to the satellite; and finally use this range to compute the satellite position.

The accuracy of the user position in the iteration is not typically a problem. The reason is because, even with a position error of 10 kilometers, the worstcase propagation time error would be 33.3 μs (i.e., 10 km / 3e8 m/s). Multiplying this by Earth rotation rate (~7.3e-5 rad/s) yields an angular error of about 2.4 nanoradians. Even over an orbital radius of 26,000 kilometers (assuming a MEO orbit), the orbital error is less than a decimeter. Then, of course, after the first epoch, the position error is typically several orders of magnitude smaller making the effect of user position error negligible.

Summary

This article has shown why Earth rotation needs to be accounted for when computing satellite coordinates for GNSS applications. The compensation is simple but crucial steps for obtaining the highest possible positioning accuracies.

The post How does Earth’s rotation affect GNSS orbit computations? appeared first on Inside GNSS - Global Navigation Satellite Systems Engineering, Policy, and Design.

Most federal agencies rely on GPS to support their missions using it for things like mapping, earthquake prediction, animal studies, environmental assessments and emerging capabilities like positive train control, which can prevent accidents. There are also new industries such as drones and driverless cars that require accurate, reliable GPS and the majority of the nation’s critical infrastructure — everything from the internet and cell towers to the power grid and the stock exchanges — needs its exceptionally accurate timing data. Most of that infrastructure is reliant on, or becoming reliant on, GPS.

The Ligado issue, according to sources, is pitting the agencies needing GPS against federal spectrum managers on the hunt for bandwidth to address surging demand from commercial users. Those sources also say that the Air Force has been crystal clear on its opposition to Ligado’s proposal but the top echelons of the Defense Department, specifically the Office of the Chief Information Officer, has not. DoD’s pro-GPS stance was essential to putting the original network plan on hold in 2012. That plan was proposed by Ligado’s predecessor LightSquared. They filed bankruptcy after the plan was set aside, emerging in 2015 with a modified proposal and, soon thereafter, a new name.

As of press time the issue was expected to come up at the March 22 meeting of the federal government’s top coordinating and policy body for navigation and timing — the National Executive Committee for Space-Based Positioning, Navigation, and Timing (the PNT EXCOM). Earlier meetings had to be canceled so this was to be the first EXCOM of the Trump administration with some officials perhaps weighing the issue for the first time.

Sources tell Inside GNSS that the members of the EXCOM seemed poised to take the next step — which would likely be the launch of an interagency team to develop an official position. That position would then be sent to the National Telecommunications and Information Administration (NTIA), which represents government interests in spectrum decisions. The NTIA would then submit the letter along with its own recommendations to the FCC. All of this could be completed by the end of April 2018, said one expert familiar with the issue.

Gap Analysis

With so much at stake a virtual nor’easter of new reports and filings hit Washington in early March. One of the most important is an interagency technical analysis of five interference studies that was released just a week ahead of the meeting.

Called the Gap Analysis, the assessment was done by the EXCOM’s panel of technical experts, the National Space-Based PNT Systems Engineering Forum (NPEF). The interagency group mirrors the EXCOM in that it draws its expertise from more than a dozen agencies, can add experts as needed, and is chaired by the Departments of Defense and Transportation.

The NPEF was asked to evaluate the methodologies used in the five tests, which all looked at interference issues raised by having a terrestrial LTE network operating in the frequencies neighboring the GPS L1 signal. The NPEF was told to base its work on the recommendations of the EXCOM’s other panel of top satellite navigation experts, the Space-Based PNT Advisory Board (PNTAB). )

That’s important because the PNTAB measures interference using the internationally accepted 1 dB degradation Interference Protection Criterion (IPC) — that is a one-decibel (1 dB) degradation in C/N0, the carrier-tonoise power density ratio. Ligado has been seeking to redefine the yardstick to a more favorable method: a change in positioning and timing accuracy.

1) Federal Communication Commission (FCC)-mandated Technical Working Group (TWG)
2) National Space-Based PNT Systems Engineering Forum (NPEF)
3) Department of Transportation (DoT) Adjacent Band Compatibility (ABC)
4) Roberson and Associates (RAA)
5) National Advanced Spectrum and Communications Test Network (NASCTN)

The NPEF also was to determine if there were any unanswered questions or untested conditions that would hinder the GPS community from determining the “maximum aggregate power level of out-of-band transmissions to ensure that the existing and evolving uses of space-based PNT services are not affected.”

Tests 1 and 2, which were done in 2011, showed there was debilitating interference to GPS receivers from the telecom network as it was designed at the time by Ligado’s predecessor firm, LightSquared. The results of these tests contributed to an FCC decision to set the LightSquared plan aside.

The final report on the third test, the Adjacent Band Compatibility Assessment (ABC study), has yet to be released. There were preliminary results presented in 2017 and the NPEF appears to have had access to either these or the final report.

“While the exact values vary by receiver category and LTE network architecture and the resulting aggregate power,” wrote the Gap Analysis authors, “the test results indicate that the maximum tolerable EIRP of interference sources in the frequency bands adjacent to GPS are in the milliwatt or microwatt range.”

The NPEF found there is sufficient data to assess the risk of using frequencies near the GPS band for a groundbased communications network. While there were also some research gaps, that is questions that would be valuable to have answered, the study team also concluded that the data from the three government tests — when used in combination — was sufficient and appropriate to determine the maximum tolerable aggregate power level of transmissions in the band adjacent to GPS L1.

As for the Ligado-sponsored tests, the NPEF acknowledged the effort that went into them but determined that their scope and framework were insufficient when evaluated against the minimum criteria. Also noted in the back of the report was the fact that the studies had limited to no input from the public or from GPS experts.

Garmin Weighs In

Ligado has been going to great lengths to put a positive spin on its tests and its settlements with different GPS receiver
manufacturers — suggesting, for example, that the deals with the five receiver firms indicate that the interference problems have really been addressed.

Garmin, it seems, disagrees.

In a letter filed March 19, the GPS receiver manufacturer stated that, while it “has long supported the domestic development of new broadband service;” it believes that “broadband development generally should not come at the expense or harm to the nation’s well functioning, innovative, and economically important global positioning service (GPS).”

The company said that in its legal settlement with Ligado it agreed not to object to proposals regarding its noncertified aviation and general location / navigation lines of business — provided that certain technical parameters were met. The firm, however, had reserved the right to comment on issues related to certified aviation and its deal with Ligado did not constitute an endorsement by Garmin of Ligado’s proposal.

Garmin also said that the two companies had not reached an agreement about the 1 dB Interference criterion as
an appropriate metric to evaluate interference. Moreover, in previous filings, Garmin has made clear that it supports the 1 dB standard and disagrees with Ligado’s approach.

The NPEF underscored that the 1 dB standard is the right approach. Going forward they strongly recommended that decisions impacting the GPS radio frequency environment be informed by data from tests aligned with the PNTAB’s set of minimum criteria,
including the 1 dB standard, and that full consideration be given to the potential operational, scientific, and economic impacts.

Taking Out the Spin

The NPEF assessment goes a long way toward sorting out the various tests and judging their usefulness. The NASCTN test for example — the second one to be sponsored by Ligado — was actually a reasonable effort to design a test methodology, and was never intended to be an interference test.

Ligado, however, has repeatedly claimed that the NASCTN data was supportive of their assertion that GPS receivers could exist along side Ligado’s revised network.

When pressed by the PNT Advisory Board during their November 2017 meeting Ligado’s Executive Vice President and Chief Legal Officer Valerie Green said that she agreed that interference testing was not the “specific intent” of the NASCTN effort.

“They did in fact develop a test methodology which, I happen to think is great and they did in fact do it (the testing),” said Green. “And as a result of that there was, in fact, data, which our engineers interpreted. That’s what engineers do with data.” It was that interpretation, she indicated, that supported the assertion that GPS operators and Ligado could coexist.

Where’s the ABC Test?

One of the real questions raised by the NPEF study, however, is the whereabouts of the ABC Assessment. The research is completed and, as of March 2017, it was supposed to be through interagency coordination and be published by May 2017.

Now, nearly a year later, experts following the issue are wondering if it is being “slow-rolled” to keep it from being factored into the Ligado spectrum decision.

“The Adjacent Band Compatibility Report is expected to be posted to the USDOT Office of Positioning, Navigation and Timing and Spectrum Management website in the coming weeks,” said a DoT spokesperson. That could put it outside the window of consideration if the EXCOM, NTIA and FCC complete their parts in a month as suggested.

The filing of the NPEF test into the docket may help prevent the ABC Assessment from being ignored. The Resilient Navigation and Timing Foundation posted the report to the FCC docket — forcing it to be addressed by the FCC as it weighs its decision.

Also seemingly missing is mention of the Air Force testing of military receivers, which was being done in conjunction with the ABC research. Though the results may be classified, the fact that such a report is being done is public knowledge, so it seems odd that it does not come up.

“The Air Force will have a separate report on the sensitivities of the military receivers,” an Air Force representative told the March 30, 2017 public workshop on the ABC Assessment. “It will be classified at the appropriate level but it will be available to federal government stakeholders that need that information,” he said, noting that he had already been tasked by GPS Program Director Col. Stephen Whitney.

Patent Emerges

In addition to the NPEF Gap Analysis the RNT Foundation also located and posted to the docket a patent that
appears to support an accusation of fraud made in a $2 billion lawsuit over how one of Ligado’s predecessor firms handled data on interference.

Phil Falcone’s Harbinger Capital sued Apollo Global Management and others for concealing the results of tests done in
2001 that showed that voice and data signals in the band neighboring GPS would cause interference to GPS receivers.

In the patent application submitted by Gary Churan, the director of systems analysis at Mobile Satellite Ventures
LP (or MSV), Churan described testing using frequencies in the 1459-1559 MHz band.

Inside GNSS asked a signal expert to look at the application to see if it did indeed reflect interference to GPS receivers from telecom signals in the band neighboring GPS.

The testing described in the patent missed some subtleties, said the expert, who asked to speak anonymously, but “it was a good first effort and it showed some (interference) problems.”

A spokesperson for Apollo Global Management in a written statementmade shortly after the lawsuit was filedsaid: “We believe the suit lacks merit and we intend to defend ourselves vigorously.”

Technology has changed since that testing was done and Ligado has lowered its planned signal power and said it no longer intends to use the problematic frequencies from 1545 to 1559 MHz — but that’s not really the issue, suggested satellite industry expert Tim Farrar.

“I think if the allegations made that the original license was based on fraudulent conduct, and the FCC decides to look into that issue, then it will be very hard, while that was ongoing, for them to grant a modification to the license,” said Farrar, who has closely followed the LightSquared/Ligado debate.

Interestingly, in February, the suit was put on hold until the end of the year. The preliminary conference is now scheduled for January 2, 2019.

Such a long stay is unusual said Farrar. “You can get extensions of a few weeks but you don’t get extensions of nine months. So it would imply that something has happened.”

The real question, Farrar said, is why Harbinger would agree to delay its case?

Other experts following the Ligado- GPS issue suggested the delay could be connected to Ligado — specifically that the case was put on hold so as not to undermine prospects for approval.

Farrar agreed that was a possibility — but said there were many other possibilities.

“For example they (Apollo) could have showed Harbinger what they planned to file, which might have made all sorts of accusations against what Harbinger did too,” he said, underscoring that the idea was strictly speculation on his part. “They (Harbinger) might have thought better of it.”

The post Ligado Decision May Be At Hand appeared first on Inside GNSS - Global Navigation Satellite Systems Engineering, Policy, and Design.

In Android Release N (“Nougat”), Google introduced APIs giving access to GNSS raw measurements from your phone. Now Google has publicly released Analysis Tools to process and analyze these measurements.

Android now powers more than two billion devices, and Android phones are made by many different manufacturers. The analysis tools enable manufacturers to see details of how the GNSS receivers are performing in different phone designs. Also, with the tools publicly available app developers, researchers, and educators can use them to develop, build and teach.

This article shows what these tools do, and how they reveal details of receiver and signal behavior that are not possible to observe without raw measurements.

We begin with the basic operation of the tools, then describe the raw measurements, derived data, interactive controls, and custom parameters. Additionally, we’ll work through three examples, showing how you can use the tools for easy, but fairly sophisticated, analysis. Next, we’ll cover report generation, and on-phone analysis. Then we’ll describe some specifics of dual frequency analysis, including the mission planner. Finally, we’ll describe where to download the tools, open source code on Github, frequently asked questions and how to provide feedback.

Basic Operation

The basic operation of the tools is: users log raw GNSS Measurements with their phones, and analyze on their desktops.

The desktop tools run on Windows, Linux and Mac OS.
The RF column (Figure 1) shows:
a) The strongest four satellites from each constellation
b) The time plot of C/N0 of all satellites
c) The skyplot of satellite positions
The Clocks column shows:
a) The pseudoranges
b) The offset frequency of the receiver clock. This is computed using a reference position — either:
i) Automatically computed mean position
ii) User entered lat,lon,alt
iii) NMEA file with truth reference PVT
This is the first major benefit of raw measurements:
you can see the receiver clock behavior to better than 1ppb (part per billion) precision. This is a really important thing to see when you are making a phone — because any heat source near the reference oscillator inside the phone may cause the clock error to ramp rapidly. And the best way to see this, in the actual phone, is to use the raw measurements like this.
c) The offset of the standby clock that keeps time if the receiver duty cycles the primary oscillator
The Measurements column shows:
a) The weighted least squares (WLS) position results obtained from the raw and smoothed pseudoranges
b) The residual errors of each pseudorange, raw and smoothed
c) The residual errors of each pseudorange-rate (-Doppler) measurement
This is the second major benefit of raw measurements: you can see the errors of each measurement, and this allows significant insight into the signal environment and receiver behavior.

Raw Measurements and Derived Data

The Android API describes truly raw measurements. One of the first things you might notice when you examine the API (see Table 1) is that there are no pseudorange measurements. This is because pseudorange is not a raw measurement, it is derived from received satellite time (ReceivedSvTimeNanos). This is something of a paradigm shift for readers brought up on survey receivers that output pseudoranges. But remember that the primary purpose of the GNSS Raw Measurements API is to observe (and thus improve) the operation of the GNSS receiver, and that is why the APIs describe the fundamental raw values. It is our hope and intent that developers will create apps that build on these measurements, providing many derivations of the raw measurements. Indeed this has already started with apps to do PPP and generate RINEX data from your phone (see Additional Resources at the end of this article).

How do you get pseudoranges from these values? The analysis tools will do it for you, and if you want to do it yourself, see the open-source code. But as a summary:

pseudorange = (tRx- tTx)*c,
tTx = ReceivedSvTimeNanos [ns],
tRx = (TimeNanos +
TimeOffsetNanos) –
(FullBiasNanos+BiasNanos)
– weekNumberNs [ns],
where
weekNumberNs =604800e9
*floor(-FullBiasNanos/604800e9)

This summary is correct for GPS when time of week is known (State = STATE_TOW_DECODED or STATE_TOW_KNOWN). For other constellations and/or other states you must take care of details such as modulo milliseconds, and system time offsets. This is beyond the scope of this article, but these details are handled by the analysis tools and the resulting pseudoranges are available in the derived data.

The Desktop Analysis Tools compute smoothed pseudoranges as follows.

For intervals where the hardware clock is continuous, the smoothed pseudorange for a particular satellite signal is the least squares solution x to the matrix equation:

Wy = WAx
where
y = [column vector of raw pseudoranges
column vector of prr],
prr is the measured pseudorange-rate or, if available, the
change in carrier phase divided by Δt,

Δt is the time interval between measurements
W is a diagonal matrix, with Wii = 1/σ(yi)

That is, the smoothed pseudorange is the minimum variance linear estimator of the true pseudorange, given the measurements and variances of pseudorange (pr) and pseudorange-rate (prr), and zero-mean uncorrelated errors. In plain English, this is the best estimate we can get by post-processing all the available information.

Derived Data

Once you have processed a log file with the desktop tools, you can save all the derived data to a comma-separated text file. The derived data file includes: satellite azimuth and elevation, raw pseudorange, smoothed pseudorange, and the residual errors of the raw and smoothed pseudoranges (residual errors derived from the known reference positions). The derived data also contains the receiver clock bias and frequency error. From this file you can regenerate all the line plots and the skyplot produced by the Desktop Analysis Tools.

The tools provide interactive controls, and custom parameters. We’ll introduce these now and then show three examples of how to use them for analysis.

For greater control you can set custom parameters by including a text file: CustomParam.txt in the same directory as your log file. In this file you can declare the satellite(s) to be used for computing the clock errors (Figure 2).

Following are three examples that illustrate how to use the interactive controls.

Example 1: Measurement error vs C/NO.

The tool lets you select any subset of the visible satellites. In this case we’ve chosen three with different C/N0: Strong, Medium, Weak (Example 1). You can read the computed raw pseudorange errors directly from the plot, including mean and standard deviation. You can clearly see two things:
1) the relationship of C/N0 to pseudorange errors
2) the increased filtering that the receiver does to track the weaker signals, combined with the fading caused by multipath, especially on weak signals — note the increasing time constants of the errors.

Example 2: Iono, and Tropo analysis

In this example we process data collected at a known point, under open sky, away from any buildings or reflecting surfaces. We disable Iono and Tropo models by unchecking the respective boxes on the control panel, and we set the Reference PVT to the known true position of the receiver (Example 2).

The pseudorange error plots will now include the sum of Ionosphere, Troposphere, and Orbit errors. For GPS and Galileo, the typical orbit error in the broadcast ephemeris are now less than 0.5 meters (see Additional References), so the rest of the pseudorange error will be dominated by Iono+Tropo delay.

There is one thing left to do: specify which satellite to use for computing the clock error. Ordinarily the analysis program will automatically select several satellites for this. But, in order to observe relative error of iono and tropo we select a single reference satellite for computing the receiver clock. The pseudorange error on this satellite will be zero, by definition, and the errors on the remaining satellites will be the difference between their errors and the reference satellite. We choose the highest satellite, and specify it in a text file CustomParam.txt:

You analyze measurement errors from a moving receiver by supplying an NMEA truth reference file. The Analysis Tool reads the NMEA file and extracts the GGA messages (for 3D position) and RMC messages (for velocity). With this information the Tool can compute clock offsets and frequency, and measurement errors for pseudoranges and pseudorange-rate.

As the car drives into the city from bottom-right towards top-left, we see a sudden and dramatic change in C/N0 for the low satellite, and at the same time, a large increase in pseudorange error (Figure 3). While the error on the other satellites remains small.

So, the question is: what happened to this satellite GPS PRN 22?

Multipath — yes, but not classic multipath (with line-of-sight and non-line-of-sight signals overlapping), rather, what we see here are pure reflections. How do we know that? Because we can see from the skyplot, and the Google Earth image that this satellite is completely blocked by the building to the south of the car; combined with the C/N0 and PR error plots we can see that the signal that is tracked is something reflected from across the street.

As these examples show, in just a few minutes you can do fairly sophisticated analysis at a level that, until now, was inaccessible to anyone but the chip manufacturers themselves.

Having done this we run the analysis, and, in this example, pick five satellites with increasing elevations:

For these five satellites we observe the expected trend of greater atmospheric delay with decreasing elevations. The difference between the delay for the highest satellite (PRN 32, at 80°) and the lowest (PRN 24, at 7°) is approximately sixteen meters. In general there will be other errors (such as noise, and multipath) and at times they may be larger than the atmospheric delays. So not all signals will be this well behaved, but you can quite easily see the expected trend in most datasets.

Example 3: Urban Multipath/Reflection analysis

You can observe the individual effect of multipath and/or signal reflections on each measurement. As an example, here is a drive test in San Francisco — you can see the San Francisco Giants baseball stadium in the image.

The green dots here show the truth reference obtained from a GNSS inertial navigation system. We will analyze the GNSS measurements from a smartphone chip.

On-Phone Analysis

You can also do some basic analysis directly on the phone, using the Gnss-Logger App. This app was developed to:

• Exercise and validate the applicationfacing Android GNSS APIs
• Compute real-time (least-squares) position from raw measurements
• Collect data for offline analysis

The Settings View also shows important device related info such as the GNSS HW Year, Android Platform Number, Android Api Level and GnssLogger version. Log View in which the switched ON GNSS outputs will be shown (Figure 4). This view also adds the capability to log the GNSS data for offline processing and as well to send the logged files via email or other sharing options.

The offset between the two positions is also shown (Figure 5). As the device location is generally filtered, offsets of 5-10 meters may occur between device location and WLS (weighted least squares) location; even in open sky. With even higher offsets expected as the device moves to urban areas, or indoors.

Map view where one would see on Google Map the position reported by the device vs the weighted least square position (Figure 5, bottom).

The Signal Strength View is shown in Figure 6. The Residual Error plot is shown in Figure 7, using ground truth location entered by the user. If the ground truth-location is unknown, the average reported device location can be used as ground-truth. The window over which the ground-truth-location is averaged varies based on the user activity (e.g. standing, walking, running and driving).

Multi-Frequency, Multi-Constellation

The tools support multi-constellation (GPS, GLONASS, Galileo, BeiDou and QZSS) and multi-frequency. Specifically, with the advent of L1L5 receivers for smartphones, v2.6.0.0 of the tools includes features specific to analyzing the quality of L5/E5 measurements.

If L1/E1 and L5/E5 measurements are present, then the bar chart of C/N0 shows both L1/E1 and L5/E5 signals, as well as the mean difference between L1 and L5. One of the challenges of implementing L5 in a phone is the L5 antenna, and the tools show the presence and magnitude of any RF losses. Also, the pseudorange error plots show the group delay between signals on different frequencies.

The skyplot and control panel also show which satellites signals have been tracked (e.g. L1,L5, or E1,E5a, etc).

Mission Planner

The tools include a Mission Planning feature, showing which satellites will be visible from any location at any time. In particular, the Mission Planner highlights satellites with L5/E5 signals, so you can plan multi-frequency testing at appropriate times.

For example, Figure 8 shows the Mission Planner skyplot as viewed from Singapore for a 12-hour period. The GEO and IGSO satellites of BeiDou and QZSS have been selected from the control panel.

Receiver Test Report

The tools provide automatic test reports of receivers. Click “Make Report” to automatically create the test report. The report evaluates the API implementation, Received Signal, Clock behavior, and Measurement accuracy. In each case it will report PASS or FAIL based on the performance against known good benchmarks. This test report (Figure 9) is primarily meant for the phone manufacturers to use as they iterate on the design and implementation of a new phone.

Compare Log Files

You can load multiple files and do sideby-side comparison of C/N0. The “Compare” tab lets you load several different log files to compare against each other.

The [Plot C/N0] button then produces side-by-side plots of the strongest satellites from each log file (Figure 10).

To Download the Tools and Open-Sourced Code

The compiled tools run on Windows, Mac and Linux. They have been publicly released by Google, and are available free for download at: https://g.co/GNSSTools

Open-sourced Java code is available for the GnssLogger app, and open-sourced Matlab code is available for the GPS-only part of the desktop analysis. The point of this open-source code is to help developers create their own apps, and also to provide a template for how certain values are computed (such as pseudoranges, discussed above). We will evolve the analysis tools in response to user requests, but we do not intend to open source all the code for the desktop tools beyond what is already available.

Manufacturers

In Example 3, the green dots showing the truth reference are obtained from a NovAtel SPAN system from NovAtel Inc., Calgary, Alberta, Canada. The dual frequency measurements are obtained from the BCM47755 chip from Broadcom Corp., Irvine, CA. The GnssTools smartphone app, and GNSS Analysis desktop tools are from Google Inc., Mountainview, CA.

The post GNSS Analysis Tools from Google appeared first on Inside GNSS - Global Navigation Satellite Systems Engineering, Policy, and Design.