Innovation Insights with Richard Langley

IN THE BEGINNING of the space age, there was only one space-based positioning technique: satellite Doppler. Shortly after the launch of the first satellite, Sputnik 1, on Oct. 4, 1957, it was realized that by using a receiver to measure the Doppler frequency shift of a satellite’s transmitted signals combined with knowledge of the satellite’s orbit, the position of the receiver could be determined.

The United States Navy used this concept to develop the Navy Navigation Satellite System, commonly known as Transit. Although its initial use was for positioning Polaris submarines, it was released for commercial use in July 1967. Transit was used worldwide for positioning and navigation until it was decommissioned at the end of 1996. We talked about Transit in the introduction to the article “Easy Peasy, Lemon Squeezy: Satellite Navigation Using Doppler and Partial Pseudorange Measurements” in this column’s October 2012 edition.

Next on the scene was very long baseline interferometry (VLBI). This was, and still is, a technique for high-resolution mapping of galactic and extragalactic radio sources such as quasars. It was invented by Canadian and American radio astronomers with the Canadians getting the first interference “fringes” on a transcontinental baseline on May 21, 1967. VLBI uses radio telescopes, separated by 100s or 1,000s of kilometers, to record signals on storage media (previously magnetic tape and subsequently disk-based systems) synchronized by atomic clocks, typically hydrogen masers. The recordings are played back and cross-correlated at a central facility to produce the observation data – essentially the difference in arrival times of the radio signals at the radio telescopes. It was apparent that VLBI measurements could also be used to precisely determine the vector baselines between pairs of radio telescopes eventually down to a few millimeters, so VLBI became an important geodetic technique, even measuring the drift of the continents in essentially real time. We featured an article on VLBI in this column in February 1996, “The Synergy of VLBI and GPS.”

Around the same time that VLBI was being developed, satellite laser ranging (SLR) made its debut. SLR works by precisely measuring the two-way travel time of laser pulses sent from telescopes on Earth to arrays of corner-cube reflectors on specially equipped satellites. The first experiments were conducted with Beacon Explorer A in 1964. Initial results had a range accuracy of about three meters. Since then, more than 100 satellites have been launched with SLR reflectors, including the GLONASS, Galileo, BeiDou and Quasi-Zenith navigation satellites, the Indian regional satellites and a couple of GPS satellites with more to come. Ranging precisions are now as good as a few millimeters. Laser ranging is also conducted using reflector arrays on the surface of the moon. Back in September 1994, we had an SLR article in this column, “Laser Ranging to GPS Satellites with Centimeter Accuracy.”

Skipping over GNSS, with which most of us are very familiar, then came Doppler Orbitography and Radio Positioning Integrated by Satellite (DORIS). DORIS was developed in France by a group of institutions led by the Centre National d’Études Spatiales. Rather than transmitting signals from satellites and measuring the Doppler shift at receivers on the ground, the system transmits signals from a global network of ground-based beacons, which are picked up by receivers on specially equipped satellites and the data is subsequently downloaded to Earth. The first such equipped satellite was SPOT-2, launched in January 1990. Since then, 18 more satellites with DORIS receivers on board have been launched to date. DORIS, along with the other techniques, was discussed in the online GPS World article, “NASA Helps Maintain International Terrestrial Frame with GNSS,” published in February 2016.

Like the global navigation satellite systems with the International GNSS Service, the other techniques have their coordinated services, too: the International VLBI Service for Geodesy and Astrometry (IVS), the International Laser Ranging Service (ILRS), and the International DORIS Service (IDS).

All of these techniques and services contribute to the refinement of the International Terrestrial Reference Frame (ITRF), on which all positioning activities on Earth eventually depend. Tying the contributions from the different services together involves accounting for any systematic differences, which are reduced in part by using positional data at collocated sites where two or more techniques are sited with the vector ties between the instruments carefully measured. The September 1996 edition of “Innovation” was on the IERS and was aptly titled “International Terrestrial Reference Frame.”

The ITRF will enter a new era with the European Space Agency’s Genesis mission. The mission’s satellite will carry instruments for all four space-geodetic techniques: GNSS, VLBI, SLR and DORIS. In this quarter’s “Innovation” column, a team of Genesis mission engineers and scientists introduce the mission, describe its components and outline its benefits. My well-thumbed copy of the Concise Oxford Dictionary of Current English has two definitions for the word “genesis.” The first, with a capital “G,” is the title of the first book of the Old Testament with its well-known first verse. The second is “Origin, mode of formation or generation” and comes from the Greek word genēs, meaning birth, born or produced. It is clearly a fitting name for ESA’s new mission.

<p>The post Innovation Insights: A history of techniques and services that contributed to the refinement of the ITRF first appeared on GPS World.</p>

Genesis satellite.

The combination of advanced technologies for precise orbit determination and timing, as well as the scientific exploitation of GNSS signals, opens major new opportunities for relevant, innovative in-orbit scientific experiments. These opportunities come in the fields of Earth sciences, including geodesy, geophysics and GNSS remote sensing of the atmosphere, land, ocean and ice, fundamental physics, astronomy and time metrology. They could extend some current operational applications such as precise orbit determination for geodesy and altimetry and GNSS radio occultation for meteorology and space weather.

To further enhance the benefits of combining space-based geodetic techniques, the European Space Agency (ESA) has established the Genesis mission. The mission will collocate on board a single well-calibrated satellite, the four space-based geodetic techniques: GNSS, very long baseline interferometry (VLBI), satellite laser ranging (SLR) and Doppler Orbitography and Radio-positioning Integrated by Satellite (DORIS). This first-time-ever collocation in space will establish precise and stable ties among these key techniques. The Genesis satellite will be a unique, dynamic space geodetic observatory, whose observations, combined with the measurements using geodetic collocation techniques stations on Earth, will contribute to a significant improvement of the International Terrestrial Reference Frame (ITRF).

The ITRF is recognized as the foundation for all space- and ground-based observations in Earth science and navigation, and therefore this mission will potentially have a major impact on several GNSS and Earth observation applications. It is a particular realization of the terrestrial reference system, and its history goes back to 1984 when the former Bureau International de l’Heure, which was then in charge of maintaining an accessible reference frame, established a frame using space-based geodetic techniques. The tradition was continued by the International Earth Rotation and Reference Systems Service (IERS) when it was established in 1987. The IERS has periodically updated the ITRF incorporating new systems, data sets and analysis procedures. The Genesis mission will help identify any systematic errors in the ITRF and thereby improve the accuracy and stability of the frame, particularly the origin and scale of the frame, which are the most critical parameters for scientific applications.

The Genesis mission was endorsed by the ESA Ministerial Council in November 2022. The mission will be executed under the responsibility of ESA’s Navigation Directorate as an element of the Future Navigation Program in cooperation with ESA’s Operations Directorate.

ESA performed an internal mission feasibility study (a so-called concurrent design facility) in March and April 2022. A team of more than 40 experts reviewed the mission objectives and the possible implementation, derived high-level mission requirements, assessed the necessary mission instruments and their technology readiness level and concluded that the mission is feasible and compatible with the Genesis-defined program boundaries.

GENESIS MISSION OBJECTIVES

The overall mission goal, as defined by the Global Geodetic Observing System (GGOS) initiative of the International Association of Geodesy, is to help achieve an ITRF accuracy of 1 millimeter with long-term stability of 0.1 millimeters per year, to be able to detect the smallest variations in the Earth system solid, fluid and gaseous components.

Figure 1: Genesis mission concept.

The improvements of the ITRF will impact and improve multiple geodetic and geophysical observables, as well as precise navigation and positioning, and strengthen the geodetic infrastructure, including the Galileo constellation, by reducing the systematic biases between different observing techniques.

Furthermore, the Genesis mission will allow us to improve the link between the ITRF and the International Celestial Reference Frame (ICRF) due to improvement in determining the Earth orientation parameters (EOPs). The ICRF is a realization of a quasi-inertial reference system defined by extragalactic radio sources, mostly quasars, billions of light years away. The positions of a set of globally distributed VLBI radio telescopes are determined using the difference in the arrival times of the signals at the different telescopes. The ICRF was established and is maintained through a cooperation between the International Astronomical Union and the IERS.

The ITRF and the ICRF are related through the EOPs, which include pole coordinates, the Earth’s rotation angle typically referred to as Universal Time (and the related length of day), and nutation angles and their rates.

GENESIS MISSION OVERVIEW

Figure 2 Genesis project organization.

The baseline orbit of the Genesis satellite will be circular, will have an altitude of about 6,000 kilometers and an inclination of about 95 degrees. The mass of the satellite will be on the order of 250 kilograms to 300 kilogramsg, and it will have very precise on-board metrology, through a single ultra-stable oscillator. An artist’s conception of the satellite in space is shown in the opening image. The launch is foreseen for 2028, and the baseline duration for operations is two years with an option for extension.

The Genesis mission architecture will consist of the Genesis satellite, a ground control segment constituted by a mission control center and a (network of) ground station(s), a data processing center (including a global GNSS sensor station network), a data archiving and distribution center, and the required ground infrastructure for the VLBI, SLR and DORIS campaigns (See FIGURE 1). The scope of the procurement for this mission is the Genesis satellite, the ground control segment, the launch service and two years of operations with the option for extension.

As previously mentioned, the satellite will be launched as the first with all four space-based geodetic techniques on board — namely GNSS, VLBI, SLR and DORIS:

• GNSS receiver. This will be a high-quality multi-constellation (Galileo and GPS) and multi-frequency space receiver. The GNSS observations will be of very high quality and will allow multi-GNSS integer ambiguity resolution for the carrier phase with a very high success rate. This instrument is crucial for the very precise orbit determination of the Genesis satellite.
• VLBI. This instrument will transmit radio signals compatible with receivers at each observing VLBI station. To eliminate the ionospheric dispersive delay along the paths to each station, different frequency bands will be used. The signals will also comply with the evolving observation procedures at all VLBI stations. The signals will be observed by all geodetic VLBI antennas, including the new VLBI Global Observing System (VGOS) fast slewing stations, in their standard geodetic receiver setups. The transmitter currently under development is designed to transmit at different frequencies between 2 GHz and 14 GHz, but also higher frequency bands can be considered. The present setup for regular VGOS observations use four 1-GHz-wide bands within the S, C and X frequency bands. The unit is designed to transmit both pseudo-noise and random noise. The random noise signal mimics the broader-band noise emitted by quasar radio sources routinely observed by VLBI, and hence can be processed essentially by the usual station and correlator software. VLBI observations of Genesis will enable VLBI stations to be accurately located within the ITRF consistently with the other geodetic techniques, enable a frame tie between the celestial frame and the dynamic reference frames of satellite orbits as well as a frame tie between the ITRF and the extremely accurate and stable ICRF.
• SLR. A passive SLR retro-reflector (LRR) will be attached to the satellite in such a way to ensure an adequate field of view when the satellite is in Earth-pointing mode. The SLR observable is the round-trip time of flight of a laser pulse between a ground station and the LLR. Currently, the ITRF long-term origin is defined by SLR, and this is the most accurate satellite technique in sensing the Earth’s center of mass.
• DORIS. Genesis will include a DORIS receiver instrument. DORIS is based on the principle of the Doppler effect between a network of transmitting terrestrial beacons and the on-board instrument. DORIS was first featured on the SPOT-2 satellite launched in 1990. Since then, DORIS receivers have been featured on multiple satellites. The integration of the DORIS receiver on Genesis, given the high-precision knowledge of the Genesis orbit, will benefit other space geodetic techniques from the global DORIS network distribution.

All active instruments will rely on a single high-precision compact frequency standard payload, termed the ultra-stable oscillator.

GENESIS PROJECT ORGANIZATION

The Genesis mission is being procured in an end-to-end approach, meaning that the industry prime is responsible for the development of the satellite, including the payload instruments, the launch services and the satellite operations. For this reason, the following approach has been applied: contract signature was in March 2024. Design, development, validation and acceptance will take place between 2024 and 2027, leading up to a planned launch in 2028.

The contract for Genesis amounts to € 76.6 million. A consortium of 14 entities led by OHB Italia S.p.A. has been tasked with developing, manufacturing, qualifying, calibrating, launching and operating the Genesis satellite, including all its payloads. The mission is supported by Italy, Belgium, France, Switzerland, Hungary and the United Kingdom.

Figure 3 Processing, archiving and distribution of Genesis data and products.

The overall project organization is outlined in FIGURE 2. The ESA Genesis project team, led by the project manager, will manage and coordinate the work of all interfaces among i) the industrial consortium, ii) ESA in its role of handling data processing, archiving and operating the distribution center, iii) the scientific community for whatever the necessary interface is required for the preparation of scientific exploitation and coherency between the project development and the scientific mission objectives.

For the data processing, exploitation, archiving and dissemination of data to the scientific community, the PROcessing, Archiving, exploitation and Dissemination Centre (PROAD) has been set up, (See FIGURE 3), using the European Space Operations Centre (ESOC) Navigation Support Office facilities and the GNSS Science Support Centre (GSSC) of the European Space Astronomy Centre (ESAC).

For the data processing required in advance of scientific exploitation of the data, the ESOC Navigation Support Office facilities will be used. The data processing includes the precise orbit determination for the GENESIS satellite.

Figure 4 Genesis science team.

Furthermore, after the processing performed by ESOC, ESAC’s GSSC will be used for data archiving and data distribution for scientific exploitation. The PROAD will be set up and coordinated internally in ESA.

The setup and coordination of the required ground infrastructure, VLBI and SLR campaigns, the DORIS network and so on, will be managed by ESA’s Genesis project team together with a Genesis science team (See FIGURE 4).

The science team will also support ESA’s Genesis project team as required in the reviews and follow-up activities, especially with respect to compliance with the mission objectives.

SUMMARY

The Genesis mission is a very challenging one, which has been made possible by the combined effort from the scientific community, ESA member states, industry and ESA itself. The success of Genesis will strongly depend on the interaction, cooperation and support of the international scientific community. The mission objectives of Genesis address core scientific as well as societal aspects. Above all, the Genesis mission is at the foundation level of all positioning and navigation.

ACKNOWLEDGEMENTS

This article has drawn, in part, on the multi-author paper “GENESIS: co-location of geodetic techniques in space,” Earth, Planets and Space (2023), Vol. 75, No. 5, https://doi.org/10.1186/s40623-022-01752-w

<p>The post Innovation: ESA’s multi-modal space mission to improve geodetic applications first appeared on GPS World.</p>

Figure 1: GPSJAM map on March 23, 2024. The map is based on GPS accuracy reports fron aircraft broadcast digital radio messages (ADS-B) over a 24-h period. A vast (uncolored) area on the globe is not covered because of no ADS-B report. Arcs are drawn over part of Europe and the eastern Mediterranean to help visualize the cone of the interference coming from potential jamming power sources. (Photo: GPSJAM.org)

GPS services are critical for real-time information on positioning, navigation and time (PNT). Because of the highly accurate and continuous PNT solution provided by GPS in all weather conditions, this multi-use technology has been adopted for civil applications including some for transportation, agriculture, aviation and emergency services. The increasing societal dependence on GPS has also created a set of security vulnerabilities for these applications.

GPS applications are vulnerable to signal interference, spoofing and degraded or denied services. Both intentional — jamming — and unintentional signal interference can cause inaccurate PNT and poor navigation performance. In addition, GPS service may be intentionally degraded or disrupted during military operations and system testing. Environments in which the GPS service is unavailable or severely degraded require alternative solutions for PNT.

Several countermeasures have been implemented to mitigate the vulnerabilities of GPS receiving systems including flex power operation and signal encryption/authorization initiated by the service provider and signal filters and adaptive antennas implemented by a user. For example, a new military signal, M-code, from GPS III satellites has an improved anti-jamming capability in both the L1 and L2 frequency bands. GPS III can broadcast signals using a high-gain directional antenna, in addition to a wide-angle, full Earth antenna, which produces a restricted area spot service by manipulating signal strength. Such flex power operations help improve GPS performance in the presence of jamming. The flex power operation is different from the so-called Selective Availability (SA), which was an intentional degradation of civilian GPS signal accuracy globally. SA operation was discontinued in May 2000, so GPS services are always available for civil applications worldwide.

However, the transparency and openness of GPS services for peaceful uses is facing a hard reality, as the balance between peacetime and wartime applications can quickly change due to geopolitical conflicts. Attacking and overcoming GPS vulnerabilities has become a fast-evolving battlefield in modern electronic warfare. Jamming and spoofing of GPS — and other GNSS — have increased substantially in the eastern Mediterranean, Baltic Sea, and Arctic regions since Russia’s invasion of Ukraine. This was officially documented by the European Union Aviation Safety Agency in Safety Information Bulletin 2022-02R1 issued in November 2023.

For example, on March 23 to 24, 2024, widespread GPS jamming occurred in Eastern Europe that impacted more than 1,600 aircraft over a period of two days and was widely reported by mass media. The source of this massive jamming event was thought to be in Russia’s Kaliningrad exclave between Lithuania and Poland. As shown in FIGURE 1, the panoramic cone in the jammed region appears to support this speculation. Similar events with large-scale jamming occurred on December 25 to 26, 2023; January 19, February 2, 12 and 14; and March 1 to 3, 13, 15 to 16 and 18, 2024, according to GPSJAM.org. In addition, Figure 1 reveals a wide area of jamming in the eastern Mediterranean where the Israel-Hamas and Israel-Hezbollah conflicts are taking place.

The increased level of GPS jamming has had a significant impact on global science observations over the conflict regions, including low-quality measurements for soil moisture, and atmospheric and ionospheric soundings, as reported in the March issue of GPS World. In addition, NASA has observed many more dropouts from in-orbit GPS receivers in recent years, which degraded the ephemeris information used for scientific data. As the pattern of apparent GPS jamming continues, alternative filtering of the spaceflight GPS data would be required to safeguard continuous science operations.

The study reported in this article aims to provide a global perspective on recent GPS jamming and degraded services. Since late 2019, commercial CubeSat constellations such as Spire have provided atmospheric measurements with much-needed global coverage and spatiotemporal sampling. The amount of GNSS data increased from about 7,000 observations per day in 2020 to about 20,000 per day in 2022 thanks to the observation demand from weather and climate research.

Table 1: Spire satellite groups and GPS satellites observed.

SPIRE CUBESAT CONSTELLATION
Spire Global has flown more than 100 Low Earth Multi-Use Receiver (LEMUR) CubeSats since 2014. These cost-effective CubeSats are used to track maritime, aviation and weather activity from space, and can be replenished at a relatively fast pace for a low-Earth orbit (LEO) constellation. TABLE 1 lists the Spire CubeSats used in this study, showing their orbital inclination, flight model ID and the tracked GPS pseudorandom noise (PRN) code ID. Spire receivers track GPS L1C/A and L2 signals for precise orbit determination (POD) and radio occultation (RO) measurements with 1-Hz and 50-Hz sampling respectively. These Spire POD and RO data collections — from November 2019 to the present — are part of a contract under NASA’s Commercial Smallsat Data Acquisition Program.

Figure 2: A schematic of the Spire 3-unit (10 x 10 x 34 cm) LEMUR CubeSat showing a zenith-view POD (precise orbit determination) and a limb-view RO (radio occultation) antenna for GPS measurements. There are usually two RO antennas on the fore-and-aft line with respect to the flight velocity, but only one POD antenna is mounted at the top. (Photo: Dong L. Wu)

The Spire LEMUR spacecraft have gone through several generations and expanded capabilities from atmospheric sounding with GPS radio occultation (GPS-RO) to GPS reflectometry (GPS-R) for soil moisture and ocean winds, grazing-angle reflectometry (GPS-GR) for sea ice and GPS-POD ionospheric sounding for space weather. In essence, however, as illustrated in FIGURE 2, these measurements are made available from two types of antennas on the satellites: a low-gain POD antenna and a high-gain RO antenna. As the measurement capability and performance improved, these antenna designs have become increasingly sophisticated and may differ substantially from satellite to satellite. Thus, it is imperative to characterize these antenna patterns carefully before comparing their signal amplitude or signal-to-noise ratio (SNR).

Figure 3: POD antenna patterns derived empirically from Spire FM124 and FM128 data as a function of elevation and azimuth angles. The elevation angle is defined as the angle of GPS line-of-sight (LOS) above the spacecraft horizon. The azimuth angle is defined as the difference between GPS LOS and spacecraft velocity azimuth angles with respect to the north. A 5° x 5° bin size was used in the averaging. Colors are the mean L1 SNR in arbitrary unit from two-month data aggregation. The antenna patterns of L1 and L2 signals are assumed to be same.

Instead of using ground calibration data, we employed an empirical method using the flight data to derive Spire’s POD antenna patterns. For each CubeSat, we aggregated a few months of POD data according to the GPS-POD link direction, in terms of elevation angle and azimuth angle with respect to the satellite flight direction. The averaging of such a large ensemble of measurements allows us to smooth out the fluctuations due to ionospheric scintillations and GPS service power variations. Two examples illustrated in FIGURE 3 show drastically different antenna patterns and designs between Spire FM124 and FM128. The larger antenna gain values at high positive elevation angles are expected for the commercial off-the-shelf (COTS) planar POD antenna pointing at zenith, whereas the gain at the bottom of negative elevation angle range is an added feature in this antenna design to enable a limb sounding of ionospheric electron density. To monitor the GPS service power, we use only the SNR measurements at positive elevation angles greater than 30° using the antenna pattern normalized by the mean values in this angular range.

We analyzed both Spire POD and RO SNR data. The POD SNR data are used to determine the GPS service power, while the RO SNR data are used to estimate the jamming power or jammer-to-noise ratio (JNR) from the surface. The POD SNR data at high elevation angles are mostly from free space but need to be corrected for the antenna pattern effect to measure the GPS signal strength accurately. Depending on the GPS-POD link direction, the antenna pattern can cause a large variation in the observed SNR (see Figure 3). For JNR detection, we use the RO SNR data from very low elevation angles (lower than 0°) with a straight-line height (HSL) of less than -140 kilometers. At this height, a tracked GPS satellite is completely obscured by Earth and the RO receiver is essentially measuring the receiver system’s noise. Thus, any enhanced “noise” would be considered as a jamming signal. The RO antenna pattern is less critical in this case because the locations of ground jamming sources are unknown and their signals are weak at spacecraft altitudes. Roughly speaking, the RO antennas tend to acquire signals within a horizontal field of view (FOV) of 60°, corresponding roughly to a swath of about 1,800 kilometers at the surface. Therefore, the RO JNR has a coarse spatial resolution and represents a collective emission from the ground sources.

DEGRADED SERVICE AND JAMMING
To monitor GPS service power, we normalized the Spire POD SNR with the empirically calculated antenna pattern for each CubeSat. The normalized SNR data are averaged to produce a global monthly map and then annual maps (see FIGURE 4). The L1 and L2 SNRs from Spire represent an average GPS power at an orbital altitude of about 530 kilometers. The normalized SNRs are geo-registered using the CubeSat location where the measurement was made.

Variations between different GPS satellites as well as between different Spire CubeSat altitudes are neglected in this study. Broadcasting powers from the GPS satellites may differ by a small (about 10%) amount between PRNs, which manifest themselves as a slightly inhomogeneous distribution in the maps (see Figure 4). The impacts of the Spire orbital altitude on the estimated GPS power are small, compared to the regional GPS power reduction seen over Europe.

Further improvements can be made to produce a more accurate estimate of the GPS power as well as a time series of power changes from individual PRNs.
There is a clear GPS power reduction in the L1 and L2 signals over several targeted regions. The reduction appears to differ between L1 and L2 bands during the 2020 to 2023 interval. The most prominent power reduction regions are Europe and the Middle East, where the L2 reduction started as early as 2020. Although the L1 reduction is present in this region, it deepened after 2021 and perhaps widened more in 2022 and 2023. The degraded services for a targeted region are consistent with the new capability of GPS III in operation.

Figure 4: Annual mean GPS L1 (top panel) and L2 (bottom panel) SNR distributions observed by Spire POD receivers for 2020-2023. (Photo: Dong L. Wu)

A relatively small GPS power reduction can be found in East China and Southeast Asia in the 2020 to 2023 period. The L1 power reduction in this region reveals a shift from Southeast Asia in 2019 to East China in 2023, whereas the L2 reduction appears to be concentrated in East China during these years. While geopolitical tensions in this region did not escalate to any wars, electronic warfare operations have been widely reported over the South China Sea, the East China Sea, and the Philippines since 2017.

Jamming detection from space is a more challenging task because of the generally weak JNR at the height of orbiting receivers. In addition, a wide antenna FOV of GPS receivers could yield less accurate geolocation of jamming sources. However, jamming detection has been made from several LEO satellites by various teams of scientists and engineers. By tracking the front-end noise of an RO receiver on the MetOp satellite, researchers were able to detect the elevated noise power originating from ground-based sources. Also, using the radio frequency spectra recorded with a nadir-viewing receiver on the International Space Station, investigators demonstrated the feasibility of detecting jamming and spoofing signals from the ground. It has been shown that the location of jamming/spoofing sources can be pinned down accurately with observations from two satellites. This technique laid the foundation for a new class of space intelligence missions such as the DEEP prototype, STRATOS and HawkEye-360. Studying the SNR perturbations in POD data from GRACE and COSMIC-1/2, it was possible to generate global maps of jamming hotspots from 2007 to 2016. Investigators have made use of the enhanced noise in delay Doppler maps of down-looking GPS-R receivers for jamming detection. Recently, we analyzed GPS-RO SNR measurements at the tangent heights obscured by Earth and reported the increased level of jamming in northern Africa, the Middle East and the eastern Mediterranean after 2018.

Compared to nadir-view techniques, jamming detection from limb views has some advantages and disadvantages. Disadvantages have been associated with the long path length between the source and the receiver, resulting in a potentially weak JNR and poor geolocation of jamming sources. On the other hand, if jammers chose to radiate the power horizontally for a wide areal impact, it would allow limb-view sensors to pick up the jamming power and identify the directionality of jamming sources by comparing the JNR observed from opposite look angles.

FIGURE 5 As in Fig. 4 but the L1 and L2 JNR are derived from the Spire RO SNR measurements at HSL < -140 km. A 3 V/V background noise is subtracted from the SNR measurements to obtain the JNR. In 2023 the white region inside the enhanced jamming is caused by the quality control (QC) of Spire data processing that excluded the RO data with a low free-space SNR. (Photo: Dong L. Wu)

Without any sophisticated data processing, in the study for this article, we simply averaged all the RO L1 SNR data at HSL less than 140 kilometers to extract the JNR power by subtracting a 3 V/V (volts per volt) noise from the average (see FIGURE 5). This approach is slightly different from our previous study, which normalized the JNR by the free-space SNR. Because the JNR signals were so strong in 2020 to 2023, no normalization is needed to enhance the jamming detection.

It is not surprising to see in Figure 5 that the worst GPS-jammed region appears in the Middle East and the surrounding area where geopolitical conflicts have broken out frequently in recent years. In the Syria and Libya civil wars, as well as in the Russia-Ukraine and Israel-Hamas conflicts, low-cost UAVs (commonly referred to as drones) and precision-guided munitions were widely used in attacks, which was a major incentive to deploy electronic warfare to jam GPS-guided weapons and operations.

As shown in Figure 5, the JNR power was mostly concentrated in the eastern Mediterranean, the Middle East and northern Africa in 2020 and 2021 but spread to northern Europe after Russia invaded Ukraine in 2022. More JNR power appears to be in the GPS L2 band compared to L1, likely because L2 is a weaker signal and easier to jam and degrade GPS performance in navigation applications.

The regions of GPS signal reduction and enhanced jamming are highly correlated in the Spire observations. The high correlation is expected for the increased use of militarized commercial drones and GPS-guided munitions in the conflict zones. In the Russia-Ukraine war, low-cost GPS-based commercial drones have been imported to the battlefield, as have jamming capabilities. Their modifications and tactical use are evolving rapidly as the conflict continues. A precursor of such massive use of low-cost drones was in the Libya civil war (2014 – 2020), in which thousands of airstrikes were reported. Most of these commercial-grade drones relied on GPS civilian signals for navigation. Therefore, denying or reducing the GPS civilian signal power can help degrade the performance of militarized commercial drones. On the other hand, jamming and spoofing GPS signals remains a cost-effective electronic warfare technique in these conflicts.

CONCLUSIONS
This article has provided an overview of global GPS jamming and service reduction between 2020 and 2023 using recent observations from the Spire constellation. The service power reduction and jamming power increases are highly correlated on a regional scale, showing that Europe and the Middle East have been most impacted by the ongoing geopolitical conflicts. The area of the impacted regions has widened significantly from 2021 to 2023 and spread to Northern Europe. The dual-use civil and military GPS technology and services are currently experiencing an unprecedented scale of electronic warfare attacks.

ACKNOWLEDGMENT
The work reported in this article was supported by NASA’s Commercial Smallsat Data Acquisition program.

<p>The post Innovation: Recent GPS jamming in regions of geopolitical conflict first appeared on GPS World.</p>

Figure 1: Bobcat-1, with communications antenna stowed (left) and deployed (right). Bobcat-1 measures approximately 10 x 10 x 30 centimeters. (All figures except FIGURE 3 provided by the authors)

Bobcat-1 was a three-unit CubeSat developed and built at Ohio University’s Avionics Engineering Center in Athens, Ohio, and was named after the university’s mascot. FIGURE 1 shows Bobcat-1 with and without its antenna deployed. The satellite was launched to the International Space Station in October 2020 (see FIGURE 2) and deployed into low-Earth orbit (LEO) the following month (see FIGURE 3). In April 2022, it deorbited and burned up in Earth’s atmosphere as planned, after a successful 17-month mission, lasting eight months longer than anticipated. The last signal decoded from Bobcat-1 was received only about 10 minutes before the satellite’s demise, from an altitude of about 109 kilometers, by an amateur radio operator (ZR6AIC) near Johannesburg, South Africa, associated with SatNOGS, a global network of amateur satellite-networked open ground stations.

The main mission of the Bobcat-1 CubeSat was to evaluate the feasibility of GNSS-to-GNSS time offset monitoring from LEO. One of the secondary mission objectives was GNSS spectrum monitoring.

In addition, Bobcat-1 also included a side-mission, hosting a software-defined GPS/Galileo receiver developed by the University of Padova and Qascom — an Italian engineering company providing security solutions in satellite navigation and space cybersecurity — to perform its in-space demonstration and testing. This receiver served as a prototype for the receiver soon to be launched on NASA’s Lunar GNSS Receiver Experiment (LuGRE) mission.

Communications and control of the satellite utilized the 70-centimeter amateur radio satellite band (435-438 MHz) at a typical data rate of 60 kilobits per second and were primarily conducted using a dedicated ground station on the roof of the engineering building at Ohio University (see FIGURE 4). In total, Bobcat-1 collected and downlinked more than 656 megabytes of data during its lifetime. Over the course of the mission, Bobcat-1’s firmware was updated in-orbit on six occasions, allowing for minor enhancements to the data collection system.

Figure 2: Bobcat-1 launches aboard the Cygnus NG-14 resupply mission to the International Space Station. (All figures except FIGURE 3 provided by the authors.)

BACKGROUNDS: GNSS-TO-GNSS TIME OFFSET

GNSS-to-GNSS time offsets — also referred to as GNSS inter-constellation time offsets, inter-system biases or XYTOs — are among the critical parameters for full GNSS interoperability. Users with poor GNSS visibility, such as high-altitude spacecraft, which operate above the GNSS constellations, often do not have enough satellites in view to enable an accurate solution and can experience high dilution of precision. These users could benefit from XYTO estimates provided externally, assuming their receiver-characteristic inter-system biases (ISBs) are calibrated.

To determine a user solution using measurements from a single GNSS constellation, one must solve for four unknown parameters: the user’s spatial coordinates and the receiver-to-system time offset. This means that a minimum of four satellites must be visible to solve for a user solution. If a user has sufficient visibility of satellites from different constellations, a multi-GNSS solution can be determined. However, when applying measurements from multiple constellations, an additional unknown is added for each constellation used. For example, for a user solution incorporating measurements from both GPS and Galileo, one needs to solve for five unknowns: the user’s spatial coordinates, the receiver-to-GPS time offset, and the receiver-to-Galileo time offset. Since each constellation’s time scale is independent of the others, the inter-system time offset between the time scales leads to a prominent bias in a multi-constellation solution. Inter-system time offsets between GPS, Galileo, GLONASS, and BeiDou are generally expected to range from 10 to 100 nanoseconds, resulting in 3 to 30 meters of possible positioning error.

System-to-system time offsets are currently estimated by extensive networks of ground stations, such as those used by the International GNSS Service Multi-GNSS Experiment (MGEX). In addition, GNSS service providers often broadcast XYTO estimates in their navigation messages.

Figure 3: Bobcat-1 is deployed into low-Earth orbit by the Nanoracks CubeSat Deployer alongside SPOC, a CubeSat developed by the University of Georgia. (Photo: NASA)

So, why would estimating XYTOs from LEO be of interest?

Low-Earth orbit enables high GNSS visibility. The approximately 90-minute orbital period allows for observations from nearly all GNSS satellites multiple times per day. This enables high visibility of multiple satellites from each constellation, in turn enabling high observability of constellation parameters such as XYTOs, leveraging satellite-characteristics errors. In addition, tropospheric errors are absent and multipath is limited and can be bounded based on the CubeSat’s dimensions and geometry. Exploiting measurements from LEO could provide additional measurements and independent monitoring of the XYTO estimates provided by ground networks.

However, to estimate system-characteristic XYTOs, the receiver-characteristic biases need to be calibrated. The target is to reach accuracy of approximately 1 nanosecond or possibly lower. Therefore, the error sources need to be evaluated, mitigated, or bounded.

Figure 4: Bobcat-1’s dedicated ground station on the roof of Stocker Center in Athens, Ohio. (All figures except FIGURE 3 provided by the authors.)

Although the ionospheric effects are lower in LEO than on Earth, they cannot be neglected. Therefore, dual-frequency ionospheric delay estimates must be applied. To do so, the receiver’s inter-frequency biases (IFBs), which can introduce errors on the order of nanoseconds, need to be calibrated, as well as the satellite differential code biases (DCBs), orbit and clock errors and receiver antenna group delay. An additional challenge introduced by the LEO environment is the wide range of temperatures to which the receiver is subjected. Over a single orbit, the receiver’s temperature can vary from approximately 0 to 50 degrees Celsius. The effects of these temperature variations cause fluctuations in the receiver’s IFBs, which need to be evaluated and calibrated. Pre-launch measurements in a controlled environment using a climate chamber and two receivers of the same make and model were used for calibration. We have detailed those measurements elsewhere.

The multipath error can be bounded, as a first approximation, to 10 centimeters (or about 0.3 nanoseconds in equivalent signal delay) due to the dimensions of the CubeSat. However, given the mount of the antenna is on one of the CubeSat’s two 10 × 10 centimeter faces, that upper bound is in practice much smaller and the multipath error is mostly negligible.

Finally, the last remaining major error sources to be calibrated are the receiver ISBs. The main goal, to demonstrate the feasibility of LEO-CubeSat-based monitoring of GNSS XYTOs, requires showing the stability (or the repeatability) of the receiver biases in orbit.

Table 1: Summary of data collections discussed in this article.

DATA COLLECTION

Bobcat-1’s primary payload was a NovAtel OEM719, a triple-frequency multi-GNSS receiver, enabling measurements on all frequencies from GPS, GLONASS, Galileo and BeiDou, as well as the regional navigation satellite systems (RNSSs) QZSS and NavIC. The measurements were collected and downloaded, for post-processing purposes.

Pseudorange and carrier-phase measurements, as well as carrier-to-noise-density ratio estimates, were collected, together with the receiver’s position and velocity estimates, and other parameters such as the temperature measured by the two sensors embedded in the receiver. In limited instances, power spectral density measurements and in-phase and quadrature (I/Q) component samples were collected to support the secondary mission, GNSS spectrum monitoring. The limited downlink capacity of the satellite constrained these measurements to short time intervals.

Figure 5: Number of observations recorded by Bobcat-1 from each GNSS constellation during a data collection started on February 27, 2022.

The goal of the mission is to estimate the XYTOs for all the GNSS constellations. However, in this article only the Galileo-to-GPS time offset (GGTO) is considered. The Galileo Performance Reports published by the European Union Agency for the Space Programme (EUSPA) provide information on the accuracy of the GGTO broadcast parameters, which are typically within approximately 3 nanoseconds of the true GGTO. Therefore, the broadcast GGTO provides a point of comparison and reference for Bobcat-1’s estimates.

A summary of the data collections considered in this work is provided in TABLE 1. These data collections are among the longest recorded by Bobcat-1. As an example, FIGURE 5 shows Bobcat-1’s data collection for February 27, 2022. It should be noticed that data collections were initiated from the control station at Ohio University when the CubeSat was in view, and each data collection would start only when the satellite’s battery voltage was above a defined threshold. The collection would stop safely if a minimum voltage threshold was reached. The data sets collected during the first months of the mission had durations limited to one to four hours, since the minimum battery voltage threshold was set conservatively. However, as the mission continued, data collections recorded in the last several months before deorbiting were configured with lower thresholds, enabling continuous data collections with durations of up to 24 hours. During the longer data collections, the sampling period was set to 20 seconds to reduce the total quantity of data stored and downlinked. The work described here focuses on a select few data collections that span a period of five months between September 28, 2021, and February 27, 2022.

Figure 6: Bobcat-1’s ground track during a data collection for XYTOs estimation held in February 2022, approximately 24-hours long. Note that the blue dots correspond to the positions (latitude and longitude) of Bobcat-1. The red stars indicate that even if the position was calculated thanks to a multi-frequency and multi-GNSS solution, GPS L1 C/A measurements were not available. Analysis of the carrier-to-noise-density ratio measurements and comparison with the available spectrum measurements showed that in correspondence to those positions interference was present.

The data contain multi-frequency measurements from all systems, with an average of 180 observations made per sample. The maximum number of observations at once was 217. While multi-frequency measurements were collected from all constellations, this analysis only uses single-frequency measurements from two constellations: GPS L1 C/A and Galileo E1C.

RESULTS

There are two simple approaches to calculating inter-constellation time offsets: one involves computing multiple single-constellation user solutions, and the other involves a single multi-constellation user solution. Each approach has slightly different effects in terms of error propagation. In the first approach, the XYTOs can be calculated by taking the difference of the independently calculated receiver-to-system time offsets. This method requires at least four satellites from each constellation to be visible. In the second approach, all the receiver-to-system time offsets for all constellations involved in the solution are solved simultaneously. This reduces the number of measurements required per-constellation, with the minimum number of measurements needed being equal to the number of unknown state variables.

Figure 7: Broadcast GGTO (red) compared to Bobcat-1 Galileo-GPS time offset estimate, before calibration (blue) and filtered estimate (black). The results are related to data collection 181, started on December 27, 2021, which lasted about 16 hours (more than 10 orbits). The estimates’ variations, on the order of ±5 nanoseconds, are mainly due to temperature effects during the orbit and here are simply represented with a moving average.

In general, the latter method improves the XYTOs’ solution availability since the receiver-to-system time offsets for each system can be calculated with even fewer than four measurements from each system. For each sample point, the user solution was determined using this method, and the GGTO estimate was calculated by taking the difference of the receiver-to-GPS time offset and the receiver-to-Galileo time offset. This method allows the XYTO to be estimated by the receiver even when visibility is degraded. For example, as shown in FIGURE 6, Bobcat-1’s data collections are affected by interference, mostly on GPS L1, in some regions. Points where interference was believed to be present are marked by red stars on Bobcat-1’s ground track shown in the figure, specifically denoting points where the number of tracked GPS L1 C/A signals drops below four. For each sample point, the user solution was determined using the method discussed above, and the GGTO estimate was calculated by taking the difference of the receiver-to-GPS time offset and the receiver-to-Galileo time offset.

Figure 8: Difference between Bobcat-1 estimate and GGTO. The residual is mainly an estimate of the receiver inter-system bias that even pre-calibration shows to be stable in orbit as shown in Table 2.

FIGURE 7 shows (in blue) the GGTO estimate using Bobcat-1 measurements (data collection 181, started on December 27, 2021, and lasted about 10 orbits). The plotted values are the estimate of the system-to-system bias (GGTO) from which the receiver-specific ISB (Galileo-to-GPS) has not yet been removed. The oscillations visible in the unfiltered GGTO estimates are the result of temperature effects on the receiver. They can be mitigated by applying the calibrations made during pre-launch climate chamber testing, though for this analysis the estimates are simply filtered using a moving average (shown in black in the figure). Note that the abrupt change in the broadcast GGTO about 21 hours after the collection start corresponds to the start of a new day in UTC time, when a new estimate of the broadcast GGTO parameters was provided.

In FIGURE 8, the difference between the Bobcat-1 estimate of the GGTO and the broadcast GGTO is plotted (raw, in blue, and filtered with a moving average, in black). This is an estimate of the Bobcat-1 receiver’s Galileo-to-GPS ISB, which needs to be stable and repeatable in orbit, to enable accurate estimates of the true GGTO. As Figure 8 indicates, the receiver ISB shows stability even before calibration, showing periodical variations mainly due to temperature changes over the orbit.

TABLE 2 summarizes some results over a five-month period. Only the longest data collections were considered, but the shorter ones are also under analysis to provide a longer and denser observation window. From the data in Table 2, the Bobcat-1 receiver’s mean Galileo-to-GPS ISB, estimated by comparison with the broadcast GGTO, shows a standard deviation, pre-calibration, of less than 1.5 nanoseconds over five months. Considering that the accuracy on the broadcast GGTO is expected to be ≤ 3 nanoseconds, this estimate of the receiver ISB shows that its stability over time may enable accurate XYTO monitoring from LEO.

Table 2: Bobcat-1 Galileo-to-GPS time offset vs broadcast GGTO, for different data collections over about five months. All figures in columns two through five are in nanoseconds.

The implementation of the receiver bias calibration, including the temperature effects, will refine this result. The final test will include assessing the performance of the calculated system XYTO, utilizing it in the solution of another receiver previously calibrated and at a known location.

CONCLUSIONS

Results of five 15+ hour data collections spanning a period of five months are compared. The difference between the broadcast GGTO and the GGTO estimate calculated using data from Bobcat-1 appears to be stable within 1.5 nanoseconds. Observing the in-orbit data and comparing it with the data collected previously in a controlled environment in the laboratory, a high correlation is observed between the bias change over time and the measured receiver temperature. The mitigation of this effect will enable stability of our receiver characteristic GGTO estimate to within 1 nanosecond. These experimental results suggest that a few multi-GNSS receivers in LEO could provide a method to monitor XYTOs in near real time, providing redundancy and diversity to the ground-network-based estimation system.

ACKNOWLEDGMENTS

The authors would like to acknowledge NASA’s Satellite Communication and Navigation Office (SCaN), NASA’s Glenn Research Center, NASA’s CubeSat Launch Initiative (CSLI), and Ohio University for funding the Bobcat-1 CubeSat mission. Additionally, we thank Kevin Croissant and Gregory Dahart, previous student members of the Bobcat-1 team, and Dr. Frank van Graas, Ohio University Professor Emeritus and former faculty member of the Bobcat-1 team.

This article is based on the paper “Receiver-specific GNSS Inter-system Bias in Low-Earth Orbit” presented at ION ITM 2023, the 2023 International Technical Meeting of the Institute of Navigation, Long Beach, California, January 23-26, 2023.

<p>The post GNSS timing measurements from a low-Earth orbiting satellite first appeared on GPS World.</p>

Innovation Insights with Richard Langley

This is an introduction to the February 2024 Innovation article, “GNSS Timing Measurements from a Low-Earth Orbiting Satellite.”

In 1999, professors Jordi Puig-Suari at California Polytechnic State University and Bob Twiggs at Stanford University proposed a design for a miniaturized satellite that would allow students to more easily develop the skills necessary for the design, construction, testing and operation of satellites in low-Earth orbit (LEO). These nanosatellites would be built using standardized modules with a useful volume of 10 × 10 × 10 centimeters (hence the designation cube satellite or CubeSat) with a maximum mass of 2 kilograms. Apparently, the inspiration for the design came from the plastic box used to display “Beanie Babies,” a line of small stuffed toys. While a CubeSat can be constructed using one module or unit, termed a 1U design, modules can be stacked together to form sizes of 2U, 3U and so on.

Initially just a suggested form factor, the design was widely adopted by nanosatellite developers and in 2017 the International Organization for Standardization published the ISO 17770:2017 standard to formally define the physical, mechanical, electrical and operational requirements of CubeSats.

While some CubeSats have been launched as secondary payloads on launch vehicles, many have been released into space having been first launched to the International Space Station (ISS) in a cargo resupply vehicle. For example, Nanoracks developed a CubeSat deployer that can house multiple CubeSats. Once on the ISS, the deployer is positioned so that when its forward-facing door is opened, a spring at the back of the deployer pushes the CubeSats into space.

As of January 1, 2024, 2,323 CubeSats have been launched according to a nanosatellite database. Some of these satellites demonstrated new space technologies while others were science investigation missions to study Earth’s atmosphere or space weather or astronomical objects or other satellites. Many of these CubeSats, if not most, have been built by universities from around the world. In fact, various space agencies have programs to support the development and launch of CubeSats by students, such as NASA’s CubeSat Launch Initiative and the Canadian Space Agency’s Canadian CubeSat Project (CCP). As most CubeSats go into LEO, a lot of them have already deorbited. However, while in space, they provided a wealth of data of various kinds and many of the accumulated datasets are still being mined for new results. A nice example of such a dataset is that provided by Bobcat-1, a 3U CubeSat developed by Ohio University. Its mission, in addition to training students in aerospace technologies, was primarily to assess the feasibility of monitoring the time offsets between different GNSS, but also GNSS spectrum monitoring and testing a software-defined GNSS receiver. In this quarter’s “Innovation” column, authors from the Bobcat-1 team discuss some of their work on Galileo-to-GPS system time offsets. Go Bobcats!

<p>The post Innovation Insights: What is a CubeSat? first appeared on GPS World.</p>

Innovation Insights with Richard Langley

This is an introduction to the November 2023 Innovation article, “Using GNSS Phase Reflectometry on Maui’s Haleakalā”

We’ve all seen the news reports of the terrible devastation and loss of life in the town of Lahaina on the island of Maui by a wildfire this past August. Those terrible reports jarringly contrasted with happy memories of visits to Hawaii and its paradise islands. I recalled my visit some years ago to Maui in particular. My wife and I traveled all around Maui, but we particularly enjoyed the drive up to the top of Mount Haleakalā.

Rising to just over 3,000 meters, Haleakalā is a large, active (though currently dormant) shield volcano that forms about 75% of Maui. Just below its summit there is a visitor center with informative panels describing the geology of the volcano and the flora and fauna to be found on its flanks. On the drive up, for example, you can see endangered nēnē, the Hawaiian Goose, and the threatened silversword plants, which only bloom once in their lifetimes. And the sunrise and sunset views from the summit are quite beautiful.

A few hundred meters away from the visitor center is the Haleakalā High Altitude Observatory Site — a complex informally known as “Science City.” The site accommodates various optical telescopes and other instruments, including among others the 4-meter-aperture Daniel K. Inouye Solar Telescope (the largest solar telescope in the world), a satellite laser ranging station, and the Maui Space Surveillance Complex, which consists of a suite of telescopes operated by the Department of Defense for satellite tracking.

Also at the site is an innovative system observing the ocean surface far below using the phase of GNSS signals. Not only receiving normal line-of-sight signals from satellites, this system also receives signals that are reflected by the ocean surface, a technique called GNSS reflectometry or GNSS-R. GNSS-R can be thought of as a bi-static radar, where the transmitters (the GNSS satellites) and the receiver are separated by a large distance. The receiver can be on Earth’s surface, on an aircraft or on a low-Earth-orbiting satellite. The reflected signals contain information about the surface from which they were reflected. Depending on the receiver’s location and with suitable data processing, parameters such as ground surface elevation and its variation, water level and tide height, sea state (wave height, wind speed and wind direction), soil moisture content, and even snow depth can be deduced.

Over the years, we have had a number of articles on GNSS-R in this column using different receiver platforms (April, 1999; October, 2007; October, 2009; September 2010; September 2014; and October, 2019). In this quarter’s “Innovation” column, we have an article by some members of the team who built and operate the GNSS-R system on the top of Haleakalā. They explain how the system works and some of the preliminary observations and results they have obtained. More science in paradise!

<p>The post Innovation Insights: Science in paradise first appeared on GPS World.</p>

Originally developed for navigation and timing applications, signals from global navigation satellite systems (GNSS) are now commonly used for geophysical remote sensing applications, including observation of Earth’s surface and atmosphere using near sea-level ground stations as well as mountaintop, airborne and spaceborne platforms. GNSS reflectometry (abbreviated GNSS-R), which is the technique of using reflected signals to measure properties of Earth’s surface, has been a growing area of research and application for GNSS remote sensing. Notably, the Cyclone Global Navigation Satellite System (CYGNSS) satellite mission produces delay-Doppler maps (DDMs) that are used to monitor ocean surface wind speeds during hurricanes. Meanwhile, terrestrial and airborne GNSS-R has been used to monitor soil moisture, snow depth and vegetation growth. One area of increasing interest is precision reflectometry using signal carrier-phase measurements. The first attempt to perform precision (phase) altimetry over sea ice using GPS reflectometry measurements from the low-Earth orbiting TechDemoSat-1 was reported by researchers in 2017. Subsequently, researchers demonstrated the use of reflections collected by a Spire satellite to perform altimetry over Hudson Bay and the Java Sea and how reflections off ice in the polar regions can be used to measure ionospheric total electron content over the polar caps. While these demonstrations of GNSS-R for precision carrier-phase-based reflectometry are promising, more work needs to be done to characterize when carrier-based altimetry is feasible and what challenges it faces.

To study the challenges associated with processing reflected and low-elevation-angle radio occultation signals, the University of Colorado (CU) Boulder Satellite Navigation and Sensing (SeNSe) Laboratory has deployed a GNSS data collection site on top of Mount Haleakalā on the island of Maui, Hawaii. Recent collection campaigns aim to use this site as a testbed for GNSS-R algorithms that utilize multi-frequency and multi-polarization measurements. Previously, we carried out delay map processing for left-hand circular (LHC) and right-hand circular (RHC) polarizations for L1 and L2 GPS signals. Those results validate the open-loop processing methodology and provide an initial assessment of the data quality. We observed that the received reflected signals show deep and rapid fading in amplitude. In the work reported in this article, we extend our assessment to triple-frequency GPS (L1CA, L2C, L5Q) signals and document our methodology for extraction of the signal carrier phase. Our initial results indicate that coherent signal phase extraction is challenging, and may not be feasible for this particular experiment setup. We discuss ways in which the experiment may be improved for the purpose of obtaining coherent ocean surface reflections in the future.

EXPERIMENT BACKGROUND

The current form of the CU SeNSe Lab Mount Haleakalā GNSS experiment was deployed in June 2020. It consists of a side-facing dual-polarization horn antenna, which is shown in the left panel of FIGURE 1, along with a zenith-facing reference antenna. The horizontally- and vertically-polarized wideband signals from the horn antenna are fed into front-end hardware and are combined using 90-degree phase combiners to form LHC and RHC polarized signals, which are then recorded by a set of Ettus Universal Software Radio Peripherals (USRPs). Meanwhile, the signal from the reference antenna is sent to a Septentrio PolaRxS receiver. The right panel in Figure 1 illustrates the system setup. Note that the Septentrio onboard oven-controlled crystal oscillator is used to drive the USRPs. This allows us to use the Septentrio outputs to estimate the receiver clock variations and use them in the receiver clock component of our open-loop models, which we discuss below.

Figure 1: The side-facing horn antenna in its radome enclosure (left panel) and the hardware block diagram of the data collection system (right panel). (All figures provided by the authors)

Each USRP can record up to four signals at two different mixdown frequencies, allowing for recording of both the RHC and LHC polarized signals on up to four different bands. The first USRP records the L1 and L2 bands with center frequencies at 1575.42 and 1227.6 MHz, respectively, at a bandwidth of 5 MHz. The second USRP records the L5 and E6/B3 bands at center frequencies of 1176.45 and 1271.25 MHz and at a 20 MHz bandwidth. TABLE 1 lists the IDs for each receive channel along with its corresponding band, polarization and sampling rate. Note that the recorded signals covering the E6 band also capture BeiDou B3 signals, but we restrict our analysis to GPS L1, L2 and L5 signals in this article. The samples from these USRPs are written to disk along with the Septentrio Binary Format (SBF) output of the PolaRxS receiver.

Table 1: Receiver IDs with corresponding band and polarization.

Starting in June 2021, periodic collections were taken for around one hour at a time, which is about the amount of time it takes for a GPS satellite to pass from an elevation angle of 0 degrees to one of more than 20 degrees. The collection times were adjusted to target the passes of satellites whose specular reflection point passed within the azimuthal range of the horn antenna, which faces roughly to the south and has a beam width of around 60 degrees. FIGURE 2 summarizes the available datasets from the first month of collections. The right-most panels of FIGURE 3 show examples of the specular track for GPS PRN 6 as it sets over the horizon on June 13, 2022, at around 12:00-13:00 UT. This is the pass on which we focus in this work, since PRN 6 transmits the L1CA, L2C and L5 signals and consistently had a specular point in our region of interest.

Figure 2: Available data during the first month of collections. The average significant wave height in the region south of Haleakalā is also plotted. Numbers near the bottom indicate the datasets analyzed for this article.

METHODOLOGY

Our processing method for open-loop tracking of the reflected GNSS signals is based on our previous work in which we produced DDMs and delay maps of the signal-to-noise ratio (SNR) measurements for multiple signal frequencies and received polarizations.

Pseudorange Model. We start by generating a model of the pseudorange for both the direct and reflected signal. The model only needs to be accurate down to the chip level, since we correlate across several chips of delay for the received signals. Setting a somewhat arbitrary accuracy requirement of 0.5 chips (equivalent to a delay of around 150 meters for L1CA/L2C or 15 meters for L5 signals), allows us to ignore path delays from the ionosphere and troposphere, which should only account for up to several meters of delay. The model has three terms that we estimate relative to GPS System Time (GPST): the receiver clock error, the satellite transmitter clock error and the geometric range. We use a surveyed position of the horn antenna along with International GNSS Service precise orbit and clock products for the transmitter clock error and positions. These allow us to compute the transmitter clock error and path delay for the direct signal. The reflected signal path delay can be found by computing the specular reflection point on the WGS84 ellipsoid and adding the distances from the transmitter to the specular point and the specular point to the receiver. The remaining term to estimate is the receiver clock error. Recall that our USRPs are driven by the Septentrio internal oscillator. Therefore, the clock error in Septentrio measurements is associated with variations in the reference oscillator for the USRPs. We utilize a geodetic detrending technique to estimate these clock variations and apply them to our pseudorange model. To construct the full receiver clock error, we estimate the time-alignment of the samples near the beginning of the collections to GPST by tracking one minute of a strong, mid-elevation-angle satellite and decoding its timing information. This provides us with an estimate of GPST at the start of the file, which we can use to construct a full estimate of the GPST at any sample in the file. Also, given our pseudorange model, we can find the received code phase and the Doppler frequency.

Figure 3: Example of delay maps from GPS PRN 6. The panels to the left show delay maps for the L1CA, L2C and L5 signals, both RHC and LHC polarizations. The bottom panel shows the corresponding elevation angle over the duration of the pass. The maps to the right show the specular point location during the pass, along with a contour of the WW3 model for significant wave height and swell-significant wave height.

Signal Correlation. Using the established code phase and Doppler models, we generate correlations for both reflected and direct signals. We correlate a reference signal over each 1-millisecond interval, and for sanity-checking purposes, we compute correlations over ± 3 chips at 0.5 chip spacing. This results in in-phase and quadrature (I/Q) correlation outputs every 1 millisecond. The left panels in Figure 3 show examples of the processed reflected signals for RHC and LHC polarization L1CA, L2C and L5Q signals from PRN 6 on June 13, 2021, at 12:00-13:00 UT. Note that as the satellite sets, at around 4 degrees elevation angle, the reflected signals merge with the stronger direct signal on the L1 and L2 signals. This happens later on L5 due to its higher bandwidth. We use the 0.0 chip bin to obtain I/Q outputs for carrier-phase processing for L1 and L2. For L5, we use the 0.0, -0.5, or -1.0 chip bin to account for model mismatch toward the end of the file.

Signal Fading and the WW3 Ocean Model. An eventual goal of the Haleakalā reflectometry experiment is to compare the characteristics of processed reflected signals with the ocean surface parameters near the specular point and glistening zone. To this end, we have incorporated data from the Hawaii regional WaveWatcher 3 (WW3) model. The model outputs information about wave height, direction and period due to both wind and swell, and has a resolution of around 5 kilometers. The data from this model is available in NetCDF format from several web services. The right panels of Figure 3 show contours of the wind- and swell-significant wave height in the South Haleakalā region. Meanwhile, note that the reflected signals (left panels) show high variability in the received power throughout the duration of the collection. While we hoped to be able to immediately observe a correlation between these wave parameters and the power fluctuations, it is clear that we need additional processing to tease out such a signal, and the changing satellite geometry will likely make this difficult to observe and validate. Even still, our results at the end of this article will show that there is likely some correlation between fading and wind parameters, though to what extent is unknown. Finally, note that the LHC polarizations (RX6, RX8, RX2) show much stronger reflected signals than the RHC polarizations. Since we are interested in processing the phase for the reflected signals, we report exclusively on the use of the LHC polarization signals in the rest of this article.

Carrier-Phase Processing. Once the correlations are performed, we take the I/Q correlations for both direct and reflected signals and process them to retrieve the cleaned reflected signal phase. The first series of steps in this process involve processing the direct signal to determine navigation / overlay symbol alignment and to estimate any residual phase fluctuations, which are mostly due to unmodeled receiver clock fluctuations. FIGURE 4 illustrates this process for the L1CA signal. The raw I/Q correlations are shown in the top panel. To these we apply a Costas phase-lock loop (PLL) to track the residual phase fluctuations without being sensitive to navigation / overlay symbol transitions. Next, we remove these residual phase fluctuations to obtain the detrended I/Q values.

Figure 4: The I/Q data cleaning process for the L1CA direct signal.

As shown in the second panel, these quadrature components of the detrended I/Q values are centered at zero while the in-phase component now shows the data bits / overlay symbols. We use the detrended I/Q values to estimate the navigation bit sequence on the L1CA and L2C signals. Likewise, we estimate the alignment of the Neumann-Hoffmann overlay sequence for the L5 signal. Finally, we wipe off the estimated data bits or overlay sequence to verify the procedure. The results of wiping off the estimated navigation bits for the L1CA signal are shown in the third panel of Figure 4.

Having obtained the residual phase fluctuations and navigation / overlay symbols for the direct signal, we next apply these to clean up the reflected signal. Specifically, we remove residual phase fluctuations from the raw reflected signal I/Q values and then wipe off the corresponding navigation bits or overlay code. FIGURE 5 shows an example of the reflected I/Q data before and after this procedure. The navigation bits are clearly removed, but the reflected signal still shows fairly significant fluctuations in the cleaned I/Q values. It is from these values that we hope to extract the residual reflected signal phase.

Figure 5: The reflected signal raw I/Q (top) and the I/Q after detrending and wiping off navigation bits for the L1CA signal.

Under coherent conditions, the phase of the clean reflected I/Q data should contain only the unmodeled effects, including any signature of ocean surface height variation. However, the effect of multipath due to the rough ocean surface causes fluctuations in the received signal amplitude and phase, and can additionally cause cycle slips when we unwrap the phase. To filter out these cycle slips, we apply our simultaneous cycle slip and noise filtering (SCANF) method, which is essentially just a Kalman filter PLL with an additional step that tries to estimate and remove cycle slips. The figures in the next section show the results of applying this entire procedure to the reflected signals. The black and blue lines show the phase before and after applying SCANF. The reflected signal I/Q SNR is also included for reference. Note how the jumps in the black line coincide with SNR fades, and the blue line effectively recreates the phase trend of the black line without these jumps. This is good qualitative evidence that the SCANF algorithm was effective.

RESULTS

FIGURES 6, 7, 8, 9, 10, and 11 show the reflected signal SNR and phase for GPS PRN 6 on 6 different days. Note that these days correspond to the marked days in Figure 2, from which we observe that the wind-significant wave height is relatively high on days 1, 5, and 6, moderate on days 2 and 3, and relatively low on day 4. We noticed that the SNR fluctuations on days 1, 5, and 6 are comparatively more frequent than on other days, which we believe may be a signature of the ocean surface conditions. A more detailed analysis of this result is a topic for our future work.

Figure 6: Reflected signal residual phase before (blue) and after (black) applying the SCANF filtering for the June 11, 2021 dataset. Amplitude and phase are shown in alternating panels for L1CA, L2C and L5 respectively.

Figure 7: Phase processing results for June 13, 2021.

Overall, we observe that the phase trend is not consistent across the three signals (L1CA, L2C, L5) for any of the days. With all the multipath signatures in the cleaned reflected signal, it was uncertain whether the extracted phase will be useful for applications such as ocean surface altimetry, and these qualitative results suggest that they probably will not be. However, season and hours of the day that were processed for our work discussed in this article are very limited. It is possible that processing more data will shed further insight onto whether the reflected signal phase is usable in this experiment.

Figure 8 Phase processing results for June 21, 2021.

Figure 9: Phase processing results for June 25, 2021.

ACKNOWLEDGMENTS

The Haleakalā data collection system has been established with support from the University of Hawaii Institute of Astronomy, and the Air Force Research Laboratory. The authors appreciate the assistance from Michael Maberry, Rob Ratkowski, Daniel O’Gara, Craig Foreman, Frank van Graas and Neeraj Pujara. This research is funded by a subaward from the National Oceanic and Atmospheric Administration through the University Corporation for Atmospheric Research to CU Boulder and with partial funding support from the NASA Commercial Smallsat Data Acquisition program.

This article is based on the paper “Initial Carrier Phase Processing for the Haleakala Mountaintop GNSS-R Experiment” presented at ION ITM 2023, the 2023 International Technical Meeting of the Institute of Navigation, Long Beach, California, Jan. 23–26, 2023.

Figure 10: Phase processing results for July 1, 2021.

Figure 11: Phase processing results for July 5, 2021.

BRIAN BREITSCH is a postdoctoral fellow at the University of Colorado (CU) Boulder, where he received his Ph.D. in aerospace engineering sciences.
JADE MORTON is a professor in the Ann and H.J. Smead Department of Aerospace Engineering Sciences and the director of the Colorado Center for Astrodynamics Research at CU Boulder.

<p>The post Using GNSS Phase Reflectometry on Maui’s Haleakalā first appeared on GPS World.</p>

Figure 1: Diagram of cis-lunar space, which includes the real GPS sidelobe data collected on an HEO space vehicle. (All figures provided by the authors)

As part of NASA’s increased interest in returning to the moon, the ability to acquire accurate, onboard navigation solutions will be indispensable for autonomous operations in cis-lunar space (see Figure 1). Artemis I recently made its weeks-long journey to the Moon, and spacecraft carrying components of the Lunar Gateway and Human Landing System are planned to follow suit. During launch and within the GNSS space service volume, space vehicles can depend on the robust navigation signals transmitted by GNSS constellations (GPS, GLONASS, BeiDou, and Galileo). However, beyond this region, NASA’s Deep Space Network (DSN) serves as the system to track and guide lunar spacecraft through the dark regions of cis-lunar space. Increasingly, development of a lunar navigation satellite system (LNSS) that relies on a low size, weight and power (SWaP) “smallSat” constellation is being discussed for various possible orbits such as low lunar orbit (LLO), near rectilinear halo orbit (NRHO) and elliptical frozen orbit (ELFO).

Figure 2: DPE 3D (left) and 2D (right) spatial correlogram shown on a 3D north-east grid.

We have implemented direct positioning estimation (or collective detection) techniques to make the most of the limited and weak GPS signals (see Figure 2) that have been employed in other GNSS-degraded environments such as urban canyons. The algorithm used in conventional GNSS positioning employs a two-step method. In the first step, the receiver acquires signals to get a coarse estimate of the received signal’s phase offset. In the second step, the receiver tracks the signals using a delay lock loop coupled with a phase or frequency lock loop. The second step enables the receiver to get fine measurements, ultimately used to obtain a navigation solution. In the scenario addressed in our work, where a vehicle is navigating beyond the GPS satellite constellation, the signals are weak and sparse, and a conventional GPS receiver may not be able to acquire or maintain a lock on a satellite’s sidelobe signals to form a position solution. For a well-parameterized region of interest (that is, having a priori knowledge of the vehicle orbital state through dynamic filtering), and if the user’s clock error is known within a microsecond, a direct positioning estimator (DPE) can be used to improve acquisition sensitivity and obtain better position solutions. DPE works by incorporating code/carrier tracking loops and navigation solutions into a single step. It uses a priori information about the GPS satellites, user location, and clocks to directly estimate a position solution from the received signal. The delay-Doppler correlograms are first computed individually for the satellites and are then mapped onto a grid of possible candidate locations to produce a multi-dimensional spatial correlogram. By combining all signals using a cost function to determine the spatial location with the most correlation between satellites, the user position can be determined. As mentioned, signals received beyond the constellation will be sparse and weak, which makes DPE a desirable positioning method.

BACKGROUND

The proposed techniques draw from several studies exploring the use of weak signals and provide a groundwork for developing robust direct positioning methods for navigating beyond the constellation. NASA has supported and conducted several of the studies in developing further research into the use of signals in this space.

A study done by Kar-Ming Cheung and his colleagues at the Jet Propulsion Laboratory propagates the orbits of satellites in GPS, Galileo, and GLONASS constellations, and simulates the “weak GPS” real-time positioning and timing performances at lunar distance. The authors simulated an NRHO lunar vehicle based on the assumption that the lunar vehicle is in view of a GNSS satellite as long as it falls within the 40-degree beamwidth of the satellite’s antenna. The authors also simulate the 3D positioning performance as a function of the satellites’ ephemeris and pseudorange errors. Preliminary results showed that the lunar vehicle can see five to 13 satellites and achieve a 3D positioning error (one-sigma) of 200 to 300 meters based on reasonable ephemeris and pseudorange error assumptions. The authors also considered using relative positioning to mitigate the GNSS satellites’ ephemeris biases. Our work differs from this study in several key ways, including using real data collected beyond the GNSS constellations and investigating the method of direct positioning estimation for sparse signals.

Luke Winternitz and colleagues at the Goddard Space Flight Center described and predicted the performance of a conceptual autonomous GPS-based navigation system for NASA’s planned Lunar Gateway. The system was based on the flight-proven Magnetospheric Multiscale (MMS) GPS navigation system augmented with an Earth-pointed high-gain antenna, and optionally, an atomic clock. The authors used high-fidelity simulations calibrated against MMS flight data, making use of GPS transmitter patterns from the GPS Antenna Characterization Experiment project to predict the system’s performance in the Gateway NRHO. The results indicated that GPS can provide an autonomous, real-time navigation capability with comparable, or superior, performance to a ground-based DSN approach using eight hours of tracking data per day.

In direct positioning or collective detection research, Penina Axelrad and her colleagues at the University of Colorado at Boulder and the Charles Stark Draper Laboratory explored the use of GPS for autonomous orbit determination in geostationary orbit (GEO). They developed a novel approach for directly detecting and estimating the position of a GEO satellite using a very short duration GPS observation period that had been presented and demonstrated using a hardware simulator, radio-frequency sampling receiver, and MATLAB processing.

Ultimately, these studies and more have directed our research in exploring novel methods for navigating beyond the constellation space.

DATA COLLECTION

The data we used was collected as part of the U.S. Air Force Academy-sponsored Falcon Gold experiment and the data was also post-processed by analysts from the Aerospace Corporation. A few of the key notions behind the design of the experiment was to place an emphasis on off-the-shelf hardware components. The antenna used on board the spacecraft was a 2-inch patch antenna and the power source was a group of 30 NiMH batteries. To save power, the spacecraft collected 40-millisecond snapshots of data and only took data every five minutes. The GPS L1 frequency was down-converted to a 308.88 kHz intermediate frequency and was sampled at a low rate of 2 MHz (below the Nyquist rate) and the samples were only 1- bit wide. Again, the processing was designed to minimize power requirements.

METHODS AND SIMULATIONS

To test our techniques, we used real data collected from the Falcon Gold experiment on a launch vehicle upper stage (we’ll call it the Falcon Gold satellite) which collected data above the constellation on a HEO orbit. The data collected was sparse, and the signals were weak. However, the correlation process has shown that the collected data contained satellite pseudorandom noise codes (PRNs). Through preliminary investigation, we find that the acquired Doppler frequency offset matches the predicted orbit of the satellite when propagated forward from an initial state. The predicted orbit of the satellite was derived from the orbital parameters estimated using a batch least-squares fit of range-rate measurements using Aerospace Corporation’s TRACE orbit-determination software. The propagation method uses a Dormand-Prince eighth-order integration method with a 70-degree, first-order spherical harmonic gravity model and accounting for the gravitation of the Moon and Sun. The specifics of this investigation are detailed below.

Figure 3: GPS constellation “birdcage” (grey tracks), with regions of visibility near the GPS antenna boresight in blue and green for the given line-of-sight from the Falcon Gold satellite along its orbit (orange).

The positions of the GPS satellites are calculated using broadcast messages (combined into so-called BRDC files) and International GNSS Service (IGS) precise orbit data products (SP3 files). GPS satellites broadcast signals containing their orbit details and timing information with respect to an atomic clock. Legacy GPS signals broadcast messages contain 15 ephemeris parameters, with new parameters provided every two hours. The IGS supports a global network of more than 500 ground stations, whose data is used to precisely determine the orbit (position and velocity in an Earth-based coordinate system) and clock corrections for each GNSS satellite. These satellite positions, along with the one calculated for the Falcon Gold satellite, allowed for the simulation of visibility conditions. In other words, by determining points along the Falcon Gold satellite trajectory, we determine whether the vehicle will be within the 50° beamwidth of a GPS satellite that is not blocked by Earth.

Figure 3 shows a plot rendering of the visibility conditions of the Falcon Gold satellite at a location along its orbit to the GPS satellite tracks. Figure 4 depicts three of the 12 segments where signals were detected and compares the predicted visibility to the satellites that were actually detected. A GPS satellite is predicted to be visible to the Falcon Gold satellite if the direct line-of-sight (DLOS) is not occluded by Earth and if the DLOS is within 25° of the GPS antenna boresight (see Figure 5).

Figure 4: Predicted visibility of direct line-of-sight to each GPS satellite where a blue line indicates the PRN is predicted to be visible but undetected. A green line is predicted to be visible and was detected, and a red line indicates that the satellite is predicted to not be visible, but was still detected.

Figure 5: Depiction of the regions of a GPS orbit where the Falcon Gold satellite could potentially detect GPS signals based on visibility.

As a preliminary step to evaluate the Falcon Gold data, we analyzed the Doppler shifts that were detected at 12 locations along the Falcon Gold trajectory above the constellation. By comparing the Doppler frequency shifts detected to the ones predicted by calculating the rate of change of the range between the GPS satellites and modeled Falcon Gold satellite, we calculated the range rate root-mean-square error (RMSE). Through this analysis, we were able to verify the locations on the predicted trajectory that closely matched the detected Doppler shifts.

These results are used to direct our investigations to regions of the dataset to parameterize our orbit track in a way to effectively search our delay and Doppler correlograms to populate our spatial correlograms within the DPE. Figure 6 shows the time history of the difference of predicted range rates on the trajectory and the detected range rates on the trajectory. That is, a constant detected range rate value is subtracted from a changing range rate for the duration of the trajectory and not just at the location on the trajectory at the detect time (dashed vertical line). From this we can see that the TRACE method gives range rates near the detected ranges at the approximate detection time for the 12 different segments.

Figure 6: Plots depicting the 12 segments of detection and the corresponding time history of differences of range-rate values for each GPS PRN detected. The time history is of the range-rate difference between the predicted range rate from the TRACE-estimated trajectory and the constant detected range rate at the detection time (vertical line).

Excluding Segment 12, which was below the MEO constellation altitude, Segment 6 has more detected range rates than that of the other segments. On closer inspection of this segment, and using IGS precise orbit data products, it appears that the minimum RMSE of the range rates from the detected PRNs is off from the reported detection time by several seconds (see Figure 7). Investigating regions along the Falcon Gold TRACE-estimated trajectory and assuming a mismatch in time tagging results in a location (in Earth-centered Earth-fixed coordinates) with a lower RMSE for the predicted range rates compared to detected range rates.

Figure 7: Range-rate difference between the predicted range rate from the TRACE-estimated trajectory and the constant detected range rate at the detection time (left). A portion of the trajectory around Segment 6 with the TRACE-estimated location at the time of detection (red) and the location with the minimum RMSE of range rate (black).

To determine the search space for the DPE, we first determine the location along the original TRACE-estimated trajectory with the minimum RMSE of range rates for each segment. Then we propagate the state (position and velocity) at the minimum location to the Segment 6 time stamp. If the time segment has more than three observed range rates (Segment 6 and Segment 12), we perform a least squares velocity estimate using the range-rate measurements, using the locations along the trajectory and selecting the location with the smallest RMSE. Then, for Segment 12, the position and velocity obtained from least squares is propagated backwards in time to the Segment 6 timestamp. All of these points along the trajectory as well as the original point from the TRACE estimated trajectory are used in a way similar to the method of using a sigma point filter. Specifically, the mean and covariance of the position and velocity values are used to sample a Gaussian distribution. This distribution will serve as the first iteration of the candidate locations for DPE. There were a total of three iteration steps and at each iteration the range of clock bias values over which to search was refined from a spacing of 1,000 meters, 100 meters, and 10 meters. Also on the third iteration, the sampled Gaussian distribution was resampled with 1,000 times the covariance matrix values in the directions perpendicular to the direction to Earth. This was done to gain better insight into the GPS satellites that were contributing to the DPE solution.

RESULTS

Figure 8 shows the correlation peaks for each of the signals reported to be detected using a 15-millisecond non-coherent integration time within the DPE acquisition. Satellite PRNs 4, 16 and 19 are clearly detected. Satellite PRN 29 is less obviously detected, but the maximum correlation value is associated with the reported detected frequency. However, this is the peak detected frequency only if the Doppler search band is narrowly selected around the reported detected frequency. Similarly, while the peak code delay shows a clear acquisition peak for PRNs 4, 16 and 19, for PRN 29 the peak value for code delay is more ambiguous with many peaks of similar magnitude of correlation power. Figure 8 depicts the regions around the max peak correlation chip delay.

Figure 8: Acquisition peak in frequency (left) and time (right) for PRN 4, 16, 19 and 29. The correlograms are centered on the frequency predicted from the range rate calculated along the trajectory.

For the first iteration of DPE, the peak coordinated acquisition values for PRN 16 and PRN 4 are chosen for the solution space. From the corresponding spatial correlogram, the chosen candidate solution is roughly 44 kilometers away from the original position estimated using TRACE.
For the second iteration of DPE, the clock bias is refined to search over a 100-meter spacing. The peak values for PRN 16 and PRN 19 are chosen for the solution space and the chosen candidate solution is roughly 38 kilometers away from the original position estimated using TRACE.
For the final iteration, Figures 9 and 10 depict the solutions with the 10-meter clock bias spacing and the approach of spreading the search space over the dimension perpendicular to the direction of Earth. Again, this was done to illustrate how the peak correlations appear to be drawing close to a single intersection location. However, the results fall short of the type of results shown in the spatial correlogram previously depicted in Figure 2 when many satellite signals were detected.

Figure 9: Acquisition peaks plotted in the time domain with the candidate location chosen at the location of the vertical black line for the detected PRNs for the third iteration of the DPE method.

Figure 10: Spatial correlogram with the candidate location chosen at the location of the black circle for the detected PRNs for the third iteration of DPE method. The original TRACE-estimated position is indicated by a red circle. The two positions are approximately 28 kilometers apart.

A similar iterative method was followed using not just the four detected PRNs, but any satellite that was predicted to be visible with the relaxed criteria allowing for visibility based on receiving signals from the first and second sidelobes of the antenna. This is predicted using a larger 40° away from the GPS antenna boresight criterion. The final spatial correlogram (Figure 11) shows similar results to the intersections shown in Figure 10. However, there is potentially another PRN shown with a peak contribution near the original intersection point. These results are somewhat inconclusive and will need to be investigated further.

Figure 11: Spatial correlogram with the candidate location chosen at the location of the black circle for the detected PRNs for the third iteration of DPE method using additional satellites. The original TRACE-estimated position is indicated by a red circle. The two positions are approximately 24 kilometers apart.

CONCLUSIONS AND FUTURE WORK

Our research investigated the DPE approach of positioning beyond the GNSS constellations using real data. We will further investigate ways to parameterize our estimated orbit for use within a DPE algorithm in conjunction with other orbit determination techniques (such as filtering) as our results were promising but inconclusive. Some additional methods that may aid in this research include investigating the use of precise SP3 orbit files over the navigation message currently used (BRDC) within our DPE approach. Also, some additional work will need to be completed in determining the possibility of time tagging issues that could result in discrepancies and formulating additional methods related to visibility prediction that could aid in partitioning the search space. Additionally, we plan to investigate other segments where few signals were detected, but where more satellites are predicted to be visible (a better test of DPE). Finally, using full 40-millisecond data segments rather than the 15 milliseconds used to date may provide the additional signal strength needed to give more conclusive results.

ACKNOWLEDGMENTS

This article is based on the paper “Direct Positioning Estimation Beyond the Constellation Using Falcon Gold Data Collected on Highly Elliptical Orbit” presented at ION ITM 2023, the 2023 International Technical Meeting of the Institute of Navigation, Long Beach, California, January 23–26, 2023.

KIRSTEN STRANDJORD is an assistant professor in the Aerospace Engineering Department at the University of Minnesota. She received her Ph.D. in aerospace engineering sciences from the University of Colorado Boulder.

FAITH CORNISH is a graduate student in the Aerospace Engineering Department at the University of Minnesota.

<p>The post Far Out: Positioning above the GPS constellation first appeared on GPS World.</p>